This action might not be possible to undo. Are you sure you want to continue?
DET NORSKE VERITAS
DNVOSF201
DYNAMIC RISERS
OCTOBER 2010
The electronic pdf version of this document found through http://www.dnv.com is the officially binding version
© Det Norske Veritas
Any comments may be sent by email to rules@dnv.com
For subscription orders or information about subscription terms, please use distribution@dnv.com
Computer Typesetting (Adobe Frame Maker) by Det Norske Veritas
If any person suffers loss or damage which is proved to have been caused by any negligent act or omission of Det Norske Veritas, then Det Norske Veritas shall pay compensation to such person
for his proved direct loss or damage. However, the compensation shall not exceed an amount equal to ten times the fee charged for the service in question, provided that the maximum compen
sation shall never exceed USD 2 million.
In this provision "Det Norske Veritas" shall mean the Foundation Det Norske Veritas as well as all its subsidiaries, directors, officers, employees, agents and any other acting on behalf of Det
Norske Veritas.
FOREWORD
DET NORSKE VERITAS (DNV) is an autonomous and independent foundation with the objectives of safeguarding life,
property and the environment, at sea and onshore. DNV undertakes classification, certification, and other verification and
consultancy services relating to quality of ships, offshore units and installations, and onshore industries worldwide, and carries
out research in relation to these functions.
DNV service documents consist of amongst other the following types of documents:
— Service Specifications. Procedual requirements.
— Standards. Technical requirements.
— Recommended Practices. Guidance.
The Standards and Recommended Practices are offered within the following areas:
A) Qualification, Quality and Safety Methodology
B) Materials Technology
C) Structures
D) Systems
E) Special Facilities
F) Pipelines and Risers
G) Asset Operation
H) Marine Operations
J) Cleaner Energy
O) Subsea Systems
DET NORSKE VERITAS
Offshore Standard DNVOSF201, October 2010
ACKNOWLEDGEMENT
This Offshore Standard has been developed in close cooperation with the industry. The basis for the standard was developed
within the recently completed 4 year Joint Industry Project “Design Procedures and Acceptance Criteria for Deepwater Risers”.
The JIP was performed by DNV, SINTEF and SeaFlex and supported by international oilcompanies and national authorities.
In addition to the feedback from the JIP steering committee the Standard has been circulated on extensive internal and external
hearing. The following organisations have made contributions to the standard.
DNV is grateful for the valuable cooperations and discussions with the individual personnel of these companies.
CHANGES
• General
As of October 2010 all DNV service documents are primarily
published electronically.
In order to ensure a practical transition from the “print” scheme
to the “electronic” scheme, all documents having incorporated
amendments and corrections more recent than the date of the
latest printed issue, have been given the date October 2010.
An overview of DNV service documents, their update status
and historical “amendments and corrections” may be found
through http://www.dnv.com/resources/rules_standards/.
• Main changes
Since the previous edition (January 2001), this document has
been amended, most recently in October 2009. All changes
have been incorporated and a new date (October 2010) has
been given as explained under “General”.
Coflexip Stena Offshore Norsk Hydro Stolt Offshore
DST NPD SINTEF
ELF Phillips Petroleum Stress Engineering
Europipe Saga Petroleum Shell
Exxon Prod. Research Company SeaFlex Statoil
MCS Norway
DET NORSKE VERITAS
Offshore Standard DNVOSF201, October 2010
DET NORSKE VERITAS
CONTENTS
SECTION 1 GENERAL......................................................1
A. General ............................................................................1
A 100 Introduction.................................................................. 1
A 200 Objectives ..................................................................... 1
A 300 Scope and Application ............................................... 1
A 400 Other Codes ................................................................. 2
A 500 Structure of Standard .................................................. 2
B. Normative References .................................................3
B 100 Offshore Service Specifications ............................... 3
B 200 Offshore Standards ..................................................... 3
B 300 Recommended Practices ............................................ 3
B 400 Rules ........................................................................... 3
B 500 Certification notes and Classification notes ........... 3
B 600 Guidelines..................................................................... 3
B 700 Other references .......................................................... 3
C. Definitions.......................................................................6
C 100 Verbal forms ................................................................. 6
C 200 Definitions.................................................................... 6
D. Abbreviations and Symbols .......................................9
D 100 Abbreviations .............................................................. 9
D 200 Symbols ...................................................................... 10
D 300 Greek Characters ....................................................... 11
SECTION 2 DESIGN PHILOSOPHY AND DESIGN
PRINCIPLES...............................................12
A. General ..........................................................................12
A 100 Objective..................................................................... 12
A 200 Application................................................................. 12
B. Safety Philosophy........................................................12
B 100 General ........................................................................ 12
B 200 Safety Objective ........................................................ 12
B 300 Systematic review...................................................... 13
B 400 Fundamental requirements ...................................... 13
B 500 Operational considerations...................................... 13
B 600 Design Principles....................................................... 14
B 700 Quality Assurance and Quality System................. 14
C. Design Format.............................................................14
C 100 Basic Considerations................................................ 14
C 200 Safety Class Methodology ...................................... 14
C 300 Design by LRFD Method ........................................ 15
C 400 Design by WSD Method.......................................... 16
C 500 Reliability Based Design ......................................... 17
C 600 Design by Testing..................................................... 17
SECTION 3 LOADS..........................................................18
A. General ..........................................................................18
A 100 Objective..................................................................... 18
A 200 Application................................................................. 18
A 300 Loads ......................................................................... 18
B. Pressure Loads ............................................................19
B 100 Definition.................................................................... 19
B 200 Determination of Pressure Loads........................... 19
B 300 Pressure Control System.......................................... 20
B 400 Pressure Ratings........................................................ 20
C. Functional Loads ........................................................ 20
C 100 Definition ....................................................................20
C 200 Determination of Functional Loads........................20
D. Environmental Loads ............................................... 20
D 100 Definition ....................................................................20
D 200 Environmental Load Condition...............................20
D 300 Waves ..........................................................................20
D 400 Current.........................................................................21
D 500 Floater Motion............................................................21
SECTION 4 ANALYSIS METHODOLOGY............. 22
A. General.......................................................................... 22
A 100 Objective .....................................................................22
A 200 Application .................................................................22
A 300 Riser Analysis Procedure .........................................22
B. Extreme Combined Load Effect Assessment ..... 22
B 100 Fundamentals .............................................................22
B 200 Generalised Load Effect ...........................................23
B 300 Load Cases .................................................................23
B 400 Design Based on Environmental Statistics ...........24
B 500 Design Based on Response Statistics .....................24
C. Global Analysis .......................................................... 24
C 100 General.........................................................................24
C 200 Fatigue Analysis ........................................................25
SECTION 5 DESIGN CRITERIA FOR RISER
PIPES ............................................................ 27
A. General.......................................................................... 27
A 100 Objective .....................................................................27
A 200 Application .................................................................27
A 300 Limit States.................................................................27
B. Load Effects................................................................. 28
B 100 Design Load Effects..................................................28
B 200 Load Effect Factors ...................................................29
C. Resistance..................................................................... 29
C 100 Resistance Factors.....................................................29
C 200 Geometrical Parameters............................................29
C 300 Material Strength.......................................................30
D. Ultimate Limit State .................................................. 31
D 100 General.........................................................................31
D 200 Bursting.......................................................................31
D 300 System Hoop Buckling (Collapse).........................32
D 400 Propagating Buckling................................................32
D 500 Combined Loading Criteria .....................................33
D 600 Alternative WSD Format..........................................33
D 700 Displacement Controlled Conditions .....................33
E. Fatigue Limit State .................................................... 34
E 100 General.........................................................................34
E 200 Fatigue assessment using SN curves ....................34
E 300 Fatigue assessment by crack propagation
calculations.................................................................35
E 400 Inservice Fatigue Inspections ................................35
F. Accidental Limit State .............................................. 36
F 100 Functional requirements ...........................................36
F 200 Categories of accidental loads.................................36
DET NORSKE VERITAS
F 300 Characteristic accidental load effects .................... 36
F 400 Design against accidental loads.............................. 37
G. Serviceability Limit State .........................................38
G 100 General ........................................................................ 38
G 200 Ovalisation limit due to bending ............................ 38
G 300 Riser stroke................................................................. 38
G 400 Examples..................................................................... 38
H. Special Considerations ..............................................39
H 100 Interference................................................................. 39
H 200 Unstable Fracture and Gross Plastic
Deformation ............................................................... 40
H 300 Global Buckling......................................................... 40
SECTION 6 CONNECTORS AND RISER
COMPONENTS..........................................42
A. General ..........................................................................42
A 100 Objective..................................................................... 42
B. Connector Designs .....................................................42
B 100 Functional Requirements ......................................... 42
B 200 Design and Qualification Considerations ............. 42
B 300 Seals ......................................................................... 43
B 400 Local Analysis ........................................................... 43
C. Documentation.............................................................43
C 100 Documentation........................................................... 43
C 200 Operating and maintenance manuals ..................... 43
SECTION 7 MATERIALS...............................................44
A. General ..........................................................................44
A 100 Objective..................................................................... 44
A 200 Application................................................................. 44
A 300 Material Selection ..................................................... 44
B. Additional Requirements .........................................45
B 100 General ........................................................................ 45
B 200 Long term properties ................................................ 45
SECTION 8 DOCUMENTATION AND
VERIFICATION........................................47
A. General ..........................................................................47
A 100 Objective..................................................................... 47
B. Documentation.............................................................47
B 100 Design47
B 200 Design basis ............................................................... 47
B 300 Design analysis .......................................................... 47
B 400 Manufacture and fabrication ................................... 47
B 500 Installation and Operation........................................ 48
B 600 DFI Résumé................................................................ 48
B 700 Filing of documentation........................................... 48
C. Verification...................................................................49
C 100 General requirements ............................................... 49
C 200 Verification during the design phase..................... 49
C 300 Verification during the fabrication phase............. 49
SECTION 9 OPERATION, MAINTENANCE AND
REASSESSMENT......................................50
A. General ..........................................................................50
A 100 Objective..................................................................... 50
B. Inservice Inspection, Replacement and
Monitoring ...................................................................50
B 100 General.........................................................................50
B 200 Riser Inspection .........................................................50
B 300 Riser monitoring.........................................................51
B 400 Guidelines for inspection intervals .........................51
B 500 Condition Summary...................................................51
C. Reassessment ...............................................................51
C 100 General.........................................................................51
C 200 Ultimate Strength.......................................................51
C 300 Extended Service life.................................................51
C 400 Material Properties.....................................................52
C 500 Dimensions and Corrosion Allowance...................52
C 600 Cracked Pipes and Components ..............................52
APPENDIX A GLOBAL ANALYSIS...............................53
A. General ..........................................................................53
A 100 Objective......................................................................53
B. Physical Properties of riser systems ......................54
B 100 General.........................................................................54
B 200 Top tensioned risers ..................................................54
B 300 Compliant riser configurations................................54
B 400 Nonlinearities..............................................................55
C. Global riser system analysis....................................55
C 100 Purpose of global analysis ........................................55
C 200 General modelling/analysis considerations...........56
C 300 Static finite element analysis ...................................56
C 400 Finite element eigenvalue analysis .........................57
C 500 Dynamic finite element analysis .............................57
D. Combined floater/slender structure analysis .....58
D 100 General.........................................................................58
D 200 Coupled system analysis ...........................................58
D 300 Efficient analysis strategies considering coupling
effects ..........................................................................58
D 400 Coupled floater motion analysis ..............................59
D 500 Decoupled floater motion analysis ........................59
E. Hydrodynamic loading on slender structures ....59
E 100 General.........................................................................59
E 200 Morison equation for circular crosssections........60
E 300 Morison equation for double symmetric cross
sections ........................................................................60
E 400 Principles for selection of hydrodynamic
coefficients..................................................................61
F. Marine growth.............................................................62
G. Hydrostatic pressure loading ..................................62
H. Internal fluid flow.......................................................62
H 100 General.........................................................................62
H 200 Steady flow..................................................................62
H 300 Accelerated uniform flow.........................................63
H 400 Slug flow......................................................................63
I. Forced Floater Motions ............................................63
J. Hydrodynamic loading in moonpool.....................64
J 100 General.........................................................................64
J 200 Moonpol kinematics ..................................................64
J 300 Hydrodynamic coefficients ......................................64
DET NORSKE VERITAS
K. Structural damping ....................................................64
K 100 Global Rayleigh damping model ............................ 64
K 200 Local Rayleigh damping models ............................ 65
L. References.....................................................................65
L 100 Standards, Guidelines and Handbooks.................. 65
L 200 Technical references ................................................. 65
APPENDIX B FATIGUE ANALYSIS..............................67
A. General ..........................................................................67
A 100 Objective..................................................................... 67
A 200 Application................................................................. 67
A 300 Fatigue design............................................................ 67
A 400 Methods for fatigue damage assessment............... 67
B. Fatigue analysis procedures ....................................68
B 100 General ........................................................................ 68
B 200 Basic fatigue damage methodology....................... 68
B 300 Global fatigue analysis procedures ........................ 69
C. Narrow Band Fatigue Damage Assessment.........69
C 100 General ........................................................................ 69
C 200 Narrow Band Gaussian Fatigue Damage .............. 70
C 300 Narrow Band NonGaussian Fatigue damage...... 70
D. Wide band Fatigue Damage Assessment..............71
D 100 General ........................................................................ 71
D 200 Cycle counting........................................................... 71
D 300 Semiempirical Solutions......................................... 71
D 400 Analytical Solutions for Bimodal Spectra ........... 71
E. Fatigue Capacity SN Curves ..................................71
E 100 General ........................................................................ 71
F. References.....................................................................73
APPENDIX C ASSESSMENT OF EXTREME LOAD
EFFECT FOR COMBINED LOADING74
A. General ..........................................................................74
A 100 Objective..................................................................... 74
B. Design principles.........................................................74
B 100 General ........................................................................ 74
B 200 Design based on environmental statistics ............. 74
B 300 Design based on response statistics ....................... 75
C. Implementation of the LRFD design format.......75
C 100 General ........................................................................ 75
C 200 Generalised load effect ............................................. 75
C 300 Shortterm acceptance criteria................................. 75
C 400 Long term acceptance criteria ................................. 76
C 500 ULS Analysis Procedure.......................................... 76
C 600 Post processing procedures ..................................... 77
C 700 Computer implementation ....................................... 77
D. Implementation of the WSD design format.........77
D 100 General ........................................................................ 77
D 200 Implementation in design analyses ........................ 77
E. Shortterm extreme load effect estimation..........77
E 100 General ........................................................................ 77
E 200 Envelope statistics.....................................................78
E 300 Extreme response estimation...................................78
E 400 Statistical uncertainty and simulation planning....79
F. Longterm load effect statistics.............................. 79
F 100 General.........................................................................79
F 200 Response surface approach......................................80
G. References .................................................................... 80
G 100 Standards, Guidelines and Handbooks ..................80
G 200 Technical references .................................................80
APPENDIX D VERIFICATION OF GLOBAL
ANALYSIS MODEL................................. 81
A. General.......................................................................... 81
A 100 Objective .....................................................................81
A 200 Introduction................................................................81
B. Verification of theoretical models ......................... 81
C. Verification of numerical procedures .................. 82
C 200 Spatial discretisation.................................................82
C 300 Frequency discretisation ..........................................83
C 400 Time discretisation....................................................83
D. References .................................................................... 84
APPENDIX E VIV ANALYSIS GUIDANCE................ 85
A. General.......................................................................... 85
A 100 Objective .....................................................................85
B. Fatigue Assessment.................................................... 85
B 100 Simplified Assessment of Fatigue Damage ..........85
B 200 Multimodal Response Analysis Based on
Empirical Models ......................................................86
B 300 Methods Based on Solution of the NavierStokes
equations .....................................................................86
C. Methods for reduction of VIV ................................ 86
C 200 Modification of Riser Properties.............................86
C 300 Vortex suppression devices .....................................87
D. References .................................................................... 87
APPENDIX F FRAMEWORK FOR BASIS OF
DESIGN........................................................ 88
A. General.......................................................................... 88
A 100 Objective .....................................................................88
A 200 Application .................................................................88
B. Design basis ................................................................. 88
B 100 General.........................................................................88
B 200 General design requirements ...................................88
B 300 Internal fluid data.......................................................88
B 400 Environmental data ...................................................89
B 500 Data for Floater and Stationkeeping System.......89
B 600 Riser system and interfaces......................................90
B 700 Analysis methods and load cases ...........................90
B 800 Miscellaneous ............................................................91
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 1
DET NORSKE VERITAS
SECTION 1 GENERAL
Contents
A. General
A 100 Introduction
A 200 Objectives
A 300 Scope and Application
A 400 Other Codes
A 500 Structure of Standard
B. Normative References
B 100 Offshore Service Specifications
B 200 Offshore Standards
B 300 Recommended Practices
B 400 Rules
B 500 Certification notes and Classification notes
B 600 Guidelines
B 700 Other references
C. Definitions
C 100 Verbal forms
C 200 Definitions
D. Abbreviations and Symbols
D 100 Abbreviations
D 200 Symbols
D 300 Greek Characters
A. General
A 100 Introduction
101 This standard gives criteria, requirements and
guidance on structural design and analysis of riser systems
exposed to static and dynamic loading for use in the
offshore petroleum and natural gas industries.
102 The major benefits in using this standard comprise:
÷ provision of riser solutions with consistent safety level
based on flexible limit state design principles;
÷ application of safety class methodology linking
acceptance criteria to consequence of failure;
÷ provision of stateoftheart limit state functions in a
Load and Resistance Factor Design (LRFD) format
with reliabilitybased calibration of partial safety
factors. As an alternative, a simple conservative
Working Stress Design (WSD) format is also given;
÷ guidance and requirements for efficient global analyses
and introduce a consistent link between design checks
(failure modes), load conditions and load effect
assessment in the course of the global analyses;
÷ allowance for the use of innovative techniques and
procedures, such as reliabilitybased design methods.
103 The basic design principles and functional
requirements are in compliance with stateoftheart industry
practice.
A 200 Objectives
201 The main objectives of this standard are to:
÷ provide an international standard of safety for steel
risers utilised for drilling, completion/ workover,
production/injection, or transportation of hydrocarbons
(import/export) in the petroleum and gas industries;
÷ serve as a technical reference document in contractual
matters; and
÷ reflect the stateoftheart and consensus on accepted
industry practice and serve as a guideline for riser
design and analysis.
A 300 Scope and Application
301 This standard applies to all new built riser systems
and may also be applied to modification, operation and
upgrading of existing risers.
302 The scope covers design, materials, fabrication,
testing, operation, maintenance and reassessment of riser
systems. Aspects relating to documentation, verification and
quality control are also addressed. The main purpose is to
cover design and analysis of top tensioned and compliant
steel riser systems operated from floaters and fixed
platforms. The standard applies for permanent operation
(e.g. production and export/import of hydrocarbons and
injection of fluids), as well as for temporary operation (e.g.
drilling and completion/workover activities).
303 This standard is applicable to structural design of all
pressure containing components that comprise the riser
system, with special emphasis on:
÷ single pipes with a ratio of outside diameter to wall
thickness less than 45;
÷ riser connectors and other riser components such as
tension joints and stress joints.
Guidance note:
This standard may also be applied to design of single steel
pipes used as components in more complex composite cross
sections (e.g. umbilical) if the loading on the pipe can be
adequately predicted.
Multitube crosssections (i.e. pipes inside pipes) are not
considered explicitly. However, this standard may be applied
for design of each individual tubular of such crosssections
provided a realistic (conservative) distribution of the loading
on each individual tubular are assumed. Boundary conditions
of the pipes, temperature and local contact loads should be
considered in particular.
 end  of  Guidance  note 
304 There are, in principle, no limitations regarding
floater type, water depth, riser application and configuration.
However, for novel applications where experience is
limited, special attention shall be given to identify possible
new failure mechanisms, validity/adequacy of analysis
methodology and new loads and load combinations.
DNVOSF201 Dynamic Risers, October 2010
Page 2 Section 1
DET NORSKE VERITAS
Guidance note:
For application of this standard to new riser types/concepts
(e.g. novel hybrid systems, complex riser bundles etc) it shall
be documented that the global load effects can be predicted
with same precision as for conventional riser systems. This
may typically involve validation of computational
methodology by physical testing.
As an alternative an appropriate conservatism in design should
be documented.
 end  of  Guidance  note 
305 Examples of typical floater and riser configurations
are shown schematically in Figure 12. Examples of some
typical components/important areas included in typical riser
systems are illustrated in Figure 13.
A 400 Other Codes
401 In case of conflict between requirements of this
standard and a reference document, the requirements of this
standard shall prevail.
402 Where reference is made to codes other than DNV
documents, the valid revision shall be taken as the revision
that was current at the date of issue of this standard, unless
otherwise noted, see B 700.
403 The framework within DNV Riser standards and
RP’s is illustrated in Figure 11.
404 This standard provides the design philosophy, loads
and global analysis aspects valid for all riser materials.
Specific acceptance criteria for steel are given in this
standard while titanium and composite materials are
currently under development in associated recommended
practices. These Recommend Practice (RP) documents
subscribe, for consistency, to the safety philosophy and
analyses methodology set forward by this standard.
405 This standard is compatible with the DNV Offshore
Standard for Submarine Pipeline Systems DNVOSF101.
DNVOSF101 forms the primary reference for materials,
testing and fabrication for riser pipes. Strain limits and
acceptance criteria for displacement controlled conditions of
pipes (e.g. for reeling) shall comply with DNVOSF101.
The limit state design checks for this standard and DNV
OSF101 is similar but due to difference in the governing
failure modes and prevailing uncertainties some difference
in safety factors exist. This is discussed in more details in
Section 5.
Guidance note:
The major differences/conflicts in design principles compared
to current industry practice reflected by API (RP2RD and
RP1111) are:
− in the ASME and API codes the hydrostatic pressure test is
fundamental and often drives the design of pipelines and
export risers. The limit state based DNVOS aim to design
for the actual modes of failure and the safety margin is
ensured by a combination of material requirements, and
testing;
− the API codes (RP2RD and RP1111) implicitly assumes
displacement controlled riser configuration with secondary
bending stress for ULS design checks. The DNVOS a
priori assumes that important riser locations (top and
touchdown point) are load controlled unless otherwise
argued and documented. This implies that the fatigue
criterion in API is used as an implicit control of excessive
bending rather than explicit ULS design checks where
relevant as in this standard.
 end  of  Guidance  note 
OSF201
Design Criteria
STEEL
Design Philosophy
Loads
Analyses
OSF101
Material
Testing
Installation
RPF202
COMPOSITE
Material
Testing
Design Criteria
Rules
FLEXIBLES
RPF201
TITANIUM
Material
Testing
Design Criteria
Figure 11 Framework for DNV Riser Standards and
RP’s
A 500 Structure of Standard
501 This standard consist of two parts:
1. a main part providing minimum requirements in terms
of explicit criteria where relevant and functional
requirements elsewhere;
2. appendices containing practical guidance and
background information on topical issues.
In addition a number of supporting documents may be
required as listed in section B.
502 The main part is organised as follows:
Section 1 contains the objectives and scope of the standard.
It further introduces essential concepts, definitions and
abbreviations.
Section 2 contains the fundamental design philosophy and
design principles. It introduces the safety class methodology
and normal classification of safety classes.
Section 3 contains a classification of loads into pressure
loads, functional loads and environmental loads. Important
internal pressure definitions are given.
Section 4 contains the framework for global analysis
methodology. It provides a consistent link between design
checks for combined loading, global analysis, load effect
assessment and load cases. The section is supported by
÷ appendix A providing additional information on global
analyses;
÷ appendix B on fatigue analyses;
÷ appendix C on assessment of extreme load effect for
combined loading;
÷ appendix D on verification of global analysis model,
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 3
DET NORSKE VERITAS
÷ section 5 on acceptance criteria for combined loading
Section 5 contains acceptance criteria for the riser pipe for
ULS, SLS, ALS and FLS. This includes a definition of
resistance and load effects and safety factors for explicit
limit states.
Section 6 contains the fundamental functional requirements
for connectors and riser components.
Section 7 contains requirements for materials, manufacture,
fabrication and documentation of riser pipe and components
where the principles and requirements in OSF101 is
adhered. If other codes are applied additional evaluations is
required.
Section 8 contains requirements for documentation and
verification of the riser system. Appendix F provides
additional information.
Section 9 contains basic requirements for operation and in
service operations.
B. Normative References
The latest revision of the following documents applies:
Guidance note:
Explicit reference to paragraphs in DNVOSF101 should
relate to January 2000 version.
 end  of  Guidance  note 
B 100 Offshore Service Specifications
DNVOSS301 Certification and Verification of Pipelines
B 200 Offshore Standards
DNVOSF101 Submarine Pipeline Systems
DNVOSC105 Structural Design of TLPs by the LRFD
Method
DNVOSC106 Structural Design of Deep Draught
Floating Units
B 300 Recommended Practices
DNV RP B401 Cathodic Protection Design
DNV RPC203 Fatigue Strength
DNV RPF101 Corroded Pipelines
DNV RPF104 Mechanical Pipeline Couplings
DNV RPF105 Free Spanning Pipelines
DNV RPF106 Factory applied Pipeline Coatings for
Corrosion Control
DNV RPF108 Fracture Control for Reeling of Pipelines
DNV RPF201 Titanium Risers
DNV RPF202 Composite Risers
DNV RP O501 Erosive Wear in Piping Systems
B 400 Rules
DNV Rules for Certification of Flexible Risers and Pipes
DNV Rules for Planning and Execution of Marine
operations
DNV Rules for Classification of Fixed Offshore
Installations
B 500 Certification notes and Classification notes
DNV CN 1.2 Conformity Certification Services, Type
Approval
DNV CN 1.5 Conformity Certification Services,
Approval of Manufacturers, Metallic
Materials
DNV CN 7 Ultrasonic Inspection of Weld Connections
DNV CN 30.2 Fatigue Strength Analysis for Mobile
Offshore Units
DNV CN 30.4 Foundations
DNV CN 30.5 Environmental Conditions and
Environmental Loads
DNV CN 30.6 Structural Reliability Analysis of Marine
Structures
B 600 Guidelines
DNV Guidelines for Flexible Pipes
B 700 Other references
BS 7910 Guide on methods for assessing the
acceptability of flaws in fusion welded
structures
API RP1111 Design, Construction, Operation, and
Maintenance of Offshore Hydrocarbon
Pipelines (Limit State Design)
API RP2RD Design of Risers for Floating
Production Systems (FPSs) and
TensionLeg Platforms (TLPs)
EUROCODE 3 Design of steel structures  Part 1.1:
General rules and rules for building.
ISO/FDIS 2394 General Principles on Reliability for
Structures
IS0/CD 136287 Petroleum and natural gas industries 
Design and operation of subsea
production systems  Part 7:
Completion/workover riser systems
Guidance note:
The latest revision of the DNV documents may be found in the
publication list at the DNV website www.dnv.com.
 end  of  Guidance  note 
DNVOSF201 Dynamic Risers, October 2010
Page 4 Section 1
DET NORSKE VERITAS
Figure 12 Examples of metallic riser configurations and floaters
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 5
DET NORSKE VERITAS
Figure 13 Examples of riser components
DNVOSF201 Dynamic Risers, October 2010
Page 6 Section 1
DET NORSKE VERITAS
C. Definitions
C 100 Verbal forms
101 “Shall”: Indicates requirements strictly to be
followed in order to conform to this standard and from
which no deviation is permitted.
102 “Should”: Indicates that among several possibilities,
one is recommended as particularly suitable, without
mentioning or excluding others, or that a certain course of
action is preferred but not necessarily required. Other
possibilities may be applied subject to agreement.
103 “May”: Verbal form used to indicate a course of
action permissible within the limits of the standard.
104 "Agreement", "by agreement": Unless otherwise
indicated, this means agreed in writing between
Manufacturer/ Contractor and Purchaser.
C 200 Definitions
201 Accidental loads: Loads acting on the riser system,
because of a sudden, unintended and undesirable event.
Typical accidental event has an annual probability of
occurrence less than 10
2
.
202 Auxiliary line: A conduit (excluding choke and kill
lines) attached to the outside of the riser main pipe such as
hydraulic supply line, buoyancy control line and mud boost
line.
203 Buckling, global: This is usually referred to as elastic
Euler buckling or bar buckling.
204 Buckling, local: Buckling mode implying
deformations of the cross section. This can e.g. be due to
external pressure (hoop buckling) and moment (wrinkling)
or a combination thereof.
205 Buoyancy modules: Structure of light weight
material, usually foamed polymers, strapped or clamped to
the exterior of riser joints, to reduce the submerged weight
of the riser.
206 Collapse capacity: Resistance against external over
pressures, i.e. hoop buckling failure (collapse).
207 Completion/Workover riser (C/WO riser):
Temporary riser used for completion or workover operations
and includes any equipment between the subsea tree/tubing
hanger and the workover floaters tensioning system.
208 Connector or coupling: A mechanical device use to
connect adjacent components in the riser system, e.g.
connecting two joints of riser pipe endtoend.
209 Corrosion allowance: The amount of wall thickness
added to the pipe or component to allow for
corrosion/erosion/wear.
210 Design checks: Design checks are investigations of
the structural safety of the riser under the influence of load
effects (design load cases) with respect to specified limit
states, representing one or more failure modes, in terms of
resistance of relevant structural models obtained in
accordance with specified principles.
211 Design load: The combination of load effects,
multiplied by their respective load effect factors.
212 Design resistance: The resistance divided by the
appropriate resistance factor(s).
213 Drilling riser: A riser utilised during drilling and
workover operations and isolates any wellbore fluids from
the environment. The major functions of drilling riser
systems are to provide fluid transportation to and from the
well; support auxiliary lines, guide tools, and drilling
strings; serve as a running and retrieving string for the BOP.
Drilling risers may also be used for well completion and
testing.
214 Dynamic Positioning (DP, automatic station
keeping): A computerised means of maintaining a floater on
location by selectively driving thrusters.
215 Effective tension: The axial wall force (axial pipe
wall stress times area) adjusted for the contributions from
external and internal pressure.
216 Environmental loads: Loads due to the environment,
such as waves, current, wind, ice and earthquake.
217 Export/import riser: Export/import risers transfer the
processed fluids from/to the floater/structure to/from another
facility, which may include an another platform/floater or
pipeline.
218 Failure: An event causing an undesirable condition,
e.g. loss of component or system function, or deterioration
of functional capability to such an extent that the safety of
the unit, personnel or environment is significantly reduced.
219 Fatigue: Cyclic loading causing degradation of the
material.
220 Fail safe: Term applied to equipment or a system so
designed that, in the event of failure or malfunction of any
part of the system, devices are automatically activated to
stabilise or secure the safety of the operation.
221 Flex joint: A laminated metal and elastomer
assembly, having a central throughpassage equal to or
greater in diameter than the interfacing pipe or tubing bore,
that is positioned in the riser string to reduce the local
bending stresses.
222 Floater: Buoyant installation, which is floating or
fixed to the sea bottom by mooring systems in temporary or
permanent phases, e.g. TLP, Ship, Semi, Spar, Deep Draft
Floater etc.
223 Floater offset: The total offset of the floater, taking
into account the floater mean offset, wave frequency
motions and low frequency wind and wave motions.
224 Floater mean offset: The offset created by steady
forces from current, wind and waves.
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 7
DET NORSKE VERITAS
225 Floater wave frequency motions: The motions that
are a direct consequence of first order wave forces acting on
the floater, causing the platform to move at periods typically
between 3 – 25 seconds, and termed the wave frequency
(WF) regime.
226 Flowline: Any pipeline connecting to the subsea tree
assembly.
227 Fracture analysis: Analysis where critical initial
defect sizes under design loads are identified to determine
the crack growth life to failure, i.e. leak or unstable fracture.
228 Functional loads: Loads caused by the physical
existence of the riser system and by the operation and
handling of the system, excluding pressure loads.
229 Global analysis: Analysis of the complete riser
system.
230 Hangoff: Riser when disconnected from seabed.
231 Installation: The operation related to installing the
riser system, such as running of riser joints, landing and
connecting or such as laying, tiein, etc. for a catenary riser.
232 Interface loads and displacements: Loads and
displacements at a particular boundary between two
systems.
233 Limit state: The state beyond which the riser or part
of the riser no longer satisfies the requirements laid down to
its performance or operation. Examples are structural failure
(rupture, local buckling) or operational limitations (stroke or
clearance).
234 Load: The term load refers to physical influences
which cause stress, strain, deformation, displacement,
motion, etc. in the riser.
235 Load effect: Response or effect of a single load or
combination of loads on the structure, such as bending
moment, effective tension, stress, strain, deformation, etc.
236 Load effect factor: Partial safety factor by which the
load effect is multiplied to obtain the design load (effect).
237 Location class: A geographic area classified
according to the distance from locations with regular human
activities.
238 Load and Resistance Factor Design (LRFD): Design
format based upon a Limit State and Partial Safety Factor
methodology. The partial safety factor methodology is an
approach where separate factors are applied for each load
effect (response) and resistance term.
239 Low Frequency (LF) motion: Motion response at
frequencies below wave frequencies at, or near surge, sway
and yaw eigenperiods for the floater. LF motions typically
have periods ranging from 30 to 300 seconds.
240 Material resistance factor: Partial safety factor
transforming a resistance to a lower fractile resistance.
241 Maximum operating condition: Maximum condition
in which the normal operations are carried out.
242 Mode of operation: The riser mode of operation
includes typically running, landing and connecting, overpull
testing, pressure testing, wellkill, connected production
(well access), connected shutin, disconnecting, emergency
disconnect, hangoff (disconnected).
243 Nominal value: Specified value.
244 Operating envelope: Limited range of parameters in
which operations will result in safe and acceptable
equipment performance.
245 Operation, Normal Operation: Conditions that are
part of routine (normal) operation of the riser system. This
should include steady flow conditions over the full range of
flow rates as well as possible packing and shutin conditions
where these occur as part of routine operation.
246 Operation, Incidental Operation: Conditions that are
not part of normal operation of the system. Such conditions
may lead to incidental pressures. Such conditions may for
example be surges due to bullheading, sudden closing of
valves, or failure of the pressure regulating system and
activation of the pressure safety system.
247 Out of roundness: The deviation of the perimeter
from a circle. This can be an ovalisation, i.e. an elliptic cross
section, or a local out of roundness, e.g. flattening. The
numerical definition of out of roundness and ovalisation is
the same.
248 Ovalisation: The deviation of the perimeter from a
circle. This has the form as an elliptic cross section. The
numerical definition of out of roundness and ovalisation is
the same.
249 Permanent riser: A riser, which will be in continuous
operation for a long time period, irrespective of
environmental conditions.
250 Pressure definitions
Mill Test
Pressure,
Incidental
Pressure, pinc
Maximum Allowable
Incidental Pressure,
Maximum Allowable
Operating Pressure,
Pressure
Safety
System
System
Regulating
Pressure
Tolerance
Tolerance
Set Point
Set Point
Figure 14 Pressure definitions
DNVOSF201 Dynamic Risers, October 2010
Page 8 Section 1
DET NORSKE VERITAS
251 Pressure, design is the maximum internal pressure
during normal operations. The design pressure must take
account of steady flow conditions over the full range of flow
conditions as well as possible packing and shutin
conditions.
252 Pressure, local: The internal pressure at any point in
the riser for the corresponding design pressure, incidental
pressure or test pressure, i.e., the pressure at the reference
height plus the static head of the transported/test medium
due to the difference between the reference height and the
height of the section being considered.
253 Pressure, incidental: The maximum internal pressure
that is unlikely to be exceeded during the duration/life of the
riser or the maximum permitted internal pressure due to
incidental operation of the riser. Incidental pressure is
referred to the same reference height as the design pressure
and includes the situations where the riser is subject to surge
pressure, unintended shutin pressure, bullheading (water
hammer) or any temporary incidental condition.
254 Pressure, initiation: External overpressure required
to initiate a propagating buckle from an existing local buckle
or dent.
255 Pressure, Maximum Allowable Incidental (MAIP):
The maximum pressure at which the riser/pipeline system
shall be operated during incidental (i.e. transient) operation.
The maximum allowable incidental pressure is defined as
the maximum incidental pressure less the positive tolerance
of the pressure safety system.
256 Pressure, Maximum Allowable Operating (MAOP):
The maximum pressure at which the riser/pipeline system
shall be operated during normal operation. The maximum
allowable operating pressure is defined as the design
pressure less the positive tolerance of the pressure regulating
system.
257 Pressure, minimum: The local minimum internal
pressure in the riser. This is equal to the minimum pressure
at the reference height plus the static head of the fluid. A
conservative estimate is to assume zero.
258 Pressure, propagating: The lowest pressure required
for a propagating buckle to continue to propagate.
259 Pressure regulating system: For export risers and in
relation to pipelines, this is the system which ensures that,
irrespective of the upstream pressure, a set pressure is
maintained at a given reference point.
260 Pressure safety system: The system which,
independent of the pressure regulating system, ensures that
the allowable incidental pressure is not exceeded.
261 Pressure, surge: The pressure produced by sudden
changes in the velocity of fluids inside the riser.
262 Pressure, System test : The surface internal pressure
or local internal test overpressure applied to the riser or riser
section during testing after completion of the installation
work to test the riser system for tightness. (normally
performed as hydrostatic testing).
263 Process shutdown: A controlled sequence of events
that ensures that the well is secured against accidental
release of hydrocarbons to the environment.
264 Production/injection riser: Production risers
transport fluids produced from the reservoir. Injection risers
transport fluids to the producing reservoir or a convenient
disposal or storage formation. The production riser may be
used for well workovers, injection, completion and other
purposes.
265 Ratcheting: Accumulated plastic deformation during
cyclic loading.
266 Resistance: Capability of a structure or part of a
structure to resist load effects also noted strength or load
carrying capacity.
267 Resistance, characteristic: The nominal value of a
strength parameter to be used in determination of design
resistance. The (characteristic) resistance is normally based
on a defined fractile in the lower end of the distribution
function for the resistance.
268 Riser component: Any part of the riser system that
may be subjected to pressure by the internal fluid. This
includes items such as flanges, connectors, stress joints,
tension joints, flexjoints, ball joints, telescopic joints, slick
joints, tees, bends, reducers and valves.
269 Riser disconnect: The operation of unlatching of a
riser connector.
270 Riser joint: A riser joint consists of a pipe member
mid section, with riser connectors at each end. Riser joints
are typically provided in 30 ft. to 50 ft. (9,14m to 15,24m)
lengths. Shorter joints, “pup joints”, may also be provided to
ensure proper spaceout.
271 Riser pipe (riser tube): The pipe, which forms the
principal conduit of the riser joint. For example, the riser
pipe is the conduit for containing the production fluid flow
from the well into the surface tree.
272 Riser system: A riser system is considered to
comprise the riser, all integrated riser components and
corrosion protection system.
273 Riser tensioner stroke: The total upward and
downward vertical movements of the riser relative to the
floater.
274 Riser tensioner system: A device that applies a
tension to the riser string while compensating for the relative
vertical motion (stroke) between the floater and riser.
Tension variations are controlled by the stiffness of the unit.
275 Risk analysis: Analysis including a systematic
identification and categorisation of risk to people, the
environment and to assets and financial interests.
276 Safety class: The concept adopted herein to classify
the criticality of the riser system.
277 Safety class resistance factor: Partial safety factor
multiplied on the resistance reflecting the safety class.
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 9
DET NORSKE VERITAS
278 Serviceability: A condition in which a structure is
considered to perform its design function satisfactorily.
279 Service life: The length of time assumed in design
that a component will be in service.
280 SN fatigue curve: Stress range versus number of
cycles to failure.
281 Specified Minimum Tensile Strength (SMTS): The
minimum tensile strength (stress) at room temperature
prescribed by the specification or standard under which the
material is purchased.
282 Specified Minimum Yield Stress (SMYS): The
minimum yield strength (stress) at room temperature
prescribed by the specification or standard under which the
material is purchased. The tensile stress at 0.5 % elongation
of the specimen gage length.
283 Specified weather window: Limits to environmental
conditions specified in operation manual.
284 Splash zone: The external region of the riser that is
periodically in and out of the water. The determination of
the splash zone includes evaluations of all relevant effects
including wave height, wave diffraction effects, tidal
variations, settlements, subsidence and vertical motions of
the riser in the splash zone.
285 Stress Concentration Factor (SCF): Equal to the
local peak alternating principal stress in a component
(including welds) divided by the nominal alternating
principal stress near the location of the component. This
factor is used to account for the increase in the stresses
caused by geometric stress amplifiers, which occur in the
riser component.
286 Stress joint: A specialised riser joint designed with a
tapered cross section, to control curvature and reduce local
bending stresses.
287 Submerged weight: Weight minus buoyancy
(commonly referred to as weight in water, wet weight, net
lift, submerged weight or effective weight). Also named
apparent weight.
288 System Effects: System effects are relevant in cases
where many riser pipe sections are subjected to similar
loading conditions, and potential structural failure may
occur in connection with the lowest structural resistance
among riser pipe sections.
289 Temporary riser: A riser which is used intermittently
for tasks of limited duration, and which can be retrieved in
severe environmental conditions, essentially marine/drilling
risers and completion/workover risers.
290 Tensioned riser: A riser, which is essentially kept
straight and tensioned in all parts, by applying a top tension
to it.
291 Tubing: Pipe used in wells to conduct fluid from the
well's producing formation into the subsea tree or to the
surface tree.
292 Water Level. The tidal range is defined as the range
between the highest astronomical tide (HAT) and lowest
astronomical tide (LAT). The mean water level (MWL) is
defined as the mean level between HAT and LAT. The
design maximum still water level (SWL) is to include
astronomical tidal influences, wind and pressure induced
storm surge and settlements and subsidence if relevant.
Maximum Still Water Level (SWL)
Highest astronomical tide (HAT)
Mean water level ( MWL)
Lowest astronomical tide ( LAT)
Storm surge
Astronomical
tidal range
Figure 15 Definition of water levels
293 Wave Frequency (WF) motion: Motion of the floater
at the frequencies of incident waves.
294 Wellbore annulus: Annular space between the
production tubing and the well casing.
295 Working Stress Design (WSD): Design method where
the structural safety margin is expressed by one central
safety factor for each limit state. The central safety factor is
the ratio between a resistance and the load effect.
D. Abbreviations and Symbols
D 100 Abbreviations
ALS Accidental Limit State
BOP Blow Out Preventer
CMn Carbon Manganese steel
CRA Corrosion Resistant Alloys
CTOD Crack Tip Opening Displacement
DDF Deep Draft Floater
DFF Design Fatigue Factor
DFCGF Design Fatigue Crack Growth Factor
DFI Design, Fabrication and Installation
DP Dynamic Positioning
ECA Engineering Criticality Assessment
FAT Factory Acceptance Tests
FD Frequency Domain
FLS Fatigue Limit State
FMEA Failure Mode Effect Analysis
FPS Floating Production System
HAZ Heat Affected Zone
HAZOP Hazard and Operational Analysis
HIPC Hydrogen Induced Pressure Cracking
HIPPS High Integrity Pressure Protection System
HSE Health, Safety and Environment
DNVOSF201 Dynamic Risers, October 2010
Page 10 Section 1
DET NORSKE VERITAS
IM Installation Manual
LF Low Frequency
LRFD Load and Resistance Factor Design
MQL Material Quality Level
MWL Mean water level
NDT Non Destructive Testing
RFC Rain Flow Counting
SCR Steel Catenary Riser
QA Quality Assurance
QC Quality Control
QL Material Quality Level
QRA Quantitative Risk Analysis
SCF Stress Concentration Factor
SLS Serviceability Limit State
SML Seamless Pipe
SMTS Specified Minimum Tensile Strength
SMYS Specified Minimum Yield Stress
SRA Structural Reliability Analysis
SWL Still Water Level
TD Time Domain
TRB Three Roll Bending
TLP Tension Leg Platform
ULS Ultimate Limit State
UO Pipe fabrication process for welded pipes
UOE Pipe fabrication process for welded pipes,
expanded
VIV Vortex Induced Vibrations
WF Wave Frequency
WSD Working Stress Design
D 200 Symbols
A Cross section area
A
i
Internal fluid area ( )
2
t 2 D
4
⋅ −
π
A
e
External cross sectional area
2
D
4
π
A
s
( ) t t D ⋅ − π Pipe steel cross section area
D Nominal outside diameter.
D
fat
Accumulated fatigue damage or Miner sum
D
i
D2t
nom
=Nominal internal diameter
D
max
Greatest measured inside or outside diameter
D
min
Smallest measured inside or outside diameter
E Young's Modulus
f
0
Ovality,
D
D D
min max
−
f
y
Yield strength to be used in design
f
u
Tensile strength to be used in design
f
k
Material strength
g Gravity acceleration
g(t) Generalised load effect
h Height from the riser section to the reference
point for design pressure
H
s
Significant wave height
) log(a intercept of SN curve
) log(
1
a
intercept of right leg of bilinear SN curve
) log(
2
a
intercept of left leg of bilinear SN curve
m inverse SN curve slope
m
1
inverse SN curve slope for right leg of
bilinear SN curve
m
2
inverse SN curve slope for left leg of bilinear
SN curve
M Bending moment
M
A
Bending moment from Accidental loads
M
d
Design bending moment
max
d
M
Maximum design bending moment, e.g. in
short term sea state
M
E
Bending moment from Environmental loads
M
F
Bending moment from Functional loads
M
k
Plastic bending moment resistance
N Axial force in pipe wall ("true" force) (tension
is positive)
n
i
Number of stress blocks
N
i
Number of stress cycles to failure at constant
amplitude
O Out of roundness, D
max
– D
min
p
b
Burst resistance pressure
p
c
Collapse pressure
p
d
Design pressure at reference point
p
e
External pressure
p
el
Elastic collapse pressure
p
i
Internal pressure
p
inc
Incidental pressure
p
ld
Local internal design pressure
p
li
Local incidental pressure
p
min
Local minimum internal pressure taken as the
most unfavourable internal pressure plus static
head of the internal fluid.
p
p
Plastic collapse pressure
p
pr
Propagating pressure
R
k
Vector of resistances
S
SW
stress range at knee of bilinear SN curve
t time
t
1
, t
2
,t
3
Pipe wall thickness, see section 5
DNVOSF201 Dynamic Risers, October 2010
Section 1 Page 11
DET NORSKE VERITAS
t
corr
Internal and external corrosion allowance
t
fab
Absolute value of the negative tolerance taken
from the material standard/specification of the
pipe
t
nom
Nominal wall thickness of pipe (uncorroded),
as specified on the drawing/specification
T
e,A
Effective tension from Accidental loads
T
e,E
Effective tension from Environmental loads
T
e,F
Effective tension from Functional loads
T
e
Effective tension (axial force) (Tension is
positive), wave period or calculation
(operating, design) temperature
max
ed
T
Maximum design effective tension, e.g. in
short term sea state
T
ed
Design effective tension (force)
T
k
Plastic axial force resistance
T
p
Wave peak period
T
w
True wall tension
T
z
Wave zeroupcrossing period
D 300 Greek Characters
α
c
Flow stress parameter accounting for strain
hardening
α
fab
Manufacturing process reduction factor
α
U
Material quality factor
γ
A
Load factor for accidental loads
γ
c
Condition factor
γ
E
Load effect factor for environmental loads
γ
F
Load effect factor for functional loads
γ
m
Resistance factor to take into account
uncertainties in material properties
γ
SC
Resistance factor dependent on safety class
(consequence of failure)
λ
n
n
th
order spectral moment
ν Poisson’s ratio for pipe wall material
η usage factor
ρ
e
Density of external fluid (e.g. sea water)
ρ
i
Density of internal fluid (contents)
DNVOSF201 Dynamic Risers, October 2010
Page 12 Section 2
DET NORSKE VERITAS
SECTION 2 DESIGN PHILOSOPHY AND DESIGN PRINCIPLES
Contents
A. General
A 100 Objective
A 200 Application
B. Safety Philosophy
B 100 General
B 200 Safety Objective
B 300 Systematic review
B 400 Fundamental requirements
B 500 Operational considerations
B 600 Design Principles
B 700 Quality Assurance and Quality System
C. Design Format
C 100 Basic Considerations
C 200 Safety Class Methodology
C 300 Design by LRFD Method
C 400 Design by WSD Method
C 500 Reliability Based Design
C 600 Design by Testing
A. General
A 100 Objective
101 The purpose of this section is to present the safety
philosophy and corresponding limit state design format
applied in this standard.
A 200 Application
201 This section applies to all risers that are to be built in
accordance with this standard. The section also provides
guidance for extension of this standard in terms of new
criteria etc.
B. Safety Philosophy
B 100 General
101 The objective of this standard is that design,
materials, fabrication, installation, commissioning,
operation, repair, requalification, and abandonment of riser
systems are safe and conducted with due regard to public
safety and protection of the environment.
102 The integrity of a riser system constructed to this
standard is ensured through a safety philosophy integrating
the different aspects illustrated in Figure 21.
Safety
Objective
Systematic Review
Fundamental
Requirements
Operational
Considerations
Quality Assurance Design Principles
Figure 21 Safety hierarchy
B 200 Safety Objective
201 An overall safety objective shall be established,
planned and implemented covering all phases from
conceptual development until retrieval or abandonment.
Guidance note:
All companies have policy regarding human aspects,
environment and financial issues. These are typically on an
overall level, but more detailed objectives and requirements in
specific areas may follow them. These policies should be used
as a basis for defining the Safety Objective for a specific riser
system. Typical statements can be:
− all work associated with the transportation, installation/
retrieval, operation and maintenance of the riser system
shall be such as to ensure that no single failure will lead to
lifethreatening situations for any person, or to
unacceptable damage to material or the environment;
− the impact on the environment shall be reduced to as low as
reasonably possible (ALARP);
− no releases of fluid content will be accepted during
operation of the riser and pipeline system;
Statements such as those above may have implications for all
or individual phases only. They are typically more relevant for
the work execution and specific design solutions. Having
defined the Safety Objective, it can be a point of discussion as
to whether this is being accomplished in the actual project. It is
therefore recommended that the overall Safety Objective be
followed up by more specific, measurable requirements.
If no policy is available, or if it is difficult to define the safety
objective, one could also start with a risk assessment. The risk
assessment could identify all hazards and their consequences,
and then enable backextrapolation to define acceptance
criteria and areas that need to be followed up more closely.
In this standard, the structural failure probability is reflected in
the choice of safety class. The choice of safety class should
also include consideration of the expressed safety objective.
 end  of  Guidance  note 
DNVOSF201 Dynamic Risers, October 2010
Section 2 Page 13
DET NORSKE VERITAS
B 300 Systematic review
301 A systematic review or analysis shall be carried out
at all phases in order to identify and evaluate the
consequences of single failures and series of failures in the
riser system, such that necessary remedial measures can be
taken. The consequences include consequences of such
events for people, for the environment and for assets and
financial interests.
302 The Operator shall determine the extent of risk
assessments and the risk assessment methods. The extent of
the review or analysis shall reflect the criticality of the riser
system, the criticality of the planned operation and previous
experience with similar systems or operations.
Guidance note:
A methodology for such a systematic review is quantitative
risk analysis (QRA). This may provide an estimation of the
overall risk to human health and safety, environment and assets
and comprises:
− hazard identification,
− assessment of probabilities of failure events,
− accident developments, and
− consequence and risk assessment.
It should be noted that legislation in some countries requires
risk analysis to be performed, at least at an overall level to
identify critical scenarios that might jeopardise the safety and
reliability of a riser system. Other methodologies for
identification of potential hazards are Failure Mode and Effect
Analysis (FMEA) and Hazard and Operability studies
(HAZOP).
 end  of  Guidance  note 
B 400 Fundamental requirements
401 A riser shall be designed, manufactured, fabricated,
operated and maintained in such a way that:
÷ with acceptable probability, it will remain fit for the use
for which it is intended, having due regard to its service
life and its cost, and
÷ with appropriate degree of reliability, it will sustain all
foreseeable load effects and other influences likely to
occur during the service life and have adequate
durability in relation to maintenance costs.
402 In order to maintain the required safety level, the
following requirements apply:
÷ the design shall be in compliance with this standard;
÷ risers shall be designed by appropriately qualified and
experienced personnel ;
÷ the materials and products shall be used as specified in
this standard or in the relevant material or product
specification ;
÷ adequate supervision and quality control shall be
provided during manufacture and fabrication, on site
and during operation ;
÷ manufacture, fabrication, handling, transportation and
operation shall be carried out by personnel having the
appropriate skill and experience. Reference is made to
recognised standards for personnel qualifications ;
÷ the riser shall be adequately maintained including
inspection and preservation when applicable ;
÷ the riser shall be operated in accordance with the design
basis and the installation and operating manuals;
÷ relevant information between personnel involved in the
design, manufacture, fabrication and operation shall be
communicated in an understandable manner to avoid
misunderstandings, see e.g. Section 9;
÷ design reviews shall be carried out where all
contributing and affected disciplines (professional
sectors) are included to identify and solve any
problems;
÷ verification shall be performed to check compliance
with provisions contained herein in addition to national
and international regulations. The extent of the
verification and the verification method in the various
phases, including design and fabrication, shall be
assessed, see Section 8.
B 500 Operational considerations
501 Operational requirements are system capabilities
needed to meet the functional requirements. Operational
considerations include matters which designers should
address in order to obtain a design that is safe and efficient
to install, operate and maintain. Operational requirements
include operational philosophy, floater motions and
environmental limits, floater interfaces, riser installation and
retrieval, inservice operations, inspection and maintenance
philosophy.
502 Safe operation of a riser requires that:
÷ the designer shall take into account all realistic
conditions under which the riser will be operated;
÷ the operations personnel shall be aware of, and comply
with, limits for safe operations.
Guidance note:
Risers generally fit into two main operational types:
− permanent risers; risers installed and left for (many) years
until subsequent retrieval, e.g. for production/injection and
export/import of fluids and temporary risers for
drilling/workover where it is not allowable to disconnect in
extreme conditions (e.g. TLP, DDF, Spar), and
− temporary risers; risers run and retrieved many times
during their service life, e.g. for drilling and/or workover
operations.
Permanent risers are normally designed to stay connected and
operate when subjected to the extreme environment. However,
operating limits may be introduced for some temporary
conditions, e.g. shut down, bullheading etc.
A temporary riser may be designed to be disconnected,
retrieved or hungoff when the operating limit for the riser is
about to be exceeded. Temporary riser system operational
parameters normally are closely monitored at all times to
ensure that the riser is being operated within prescribed limits.
The operational parameters may include parameters such as
internal pressure and density, wave height, relative vertical
motions between riser and floater (stroke), floater offset, top
tension, flex joint/ball joint angle and stress joint stresses.
DNVOSF201 Dynamic Risers, October 2010
Page 14 Section 2
DET NORSKE VERITAS
Both temporary risers and permanent risers normally have
certain operations, such as riser installation including
connection, retrieval including disconnection and pressure
testing, which are normally limited due to weather conditions.
There are two levels of riser disconnection: normal or planned
disconnection and rapid or emergency disconnection. Rapid or
emergency disconnection of the riser system may be necessary
if floater or well system emergencies occur, the floater station
keeping/tensioning system fails, or the weather suddenly and
unpredictably deteriorates beyond the riser's operating
threshold. If riser recovery is required following an emergency
disconnect event, all subsea valves should be closed before the
riser system is removed. All equipment should be designed to
be failsafe to prevent the escape of fluids from the riser/well
bore/pipeline to the environment during disconnection.
end  of  Guidance  note –
B 600 Design Principles
601 In this standard, structural safety of the riser is
ensured by use of a safety class methodology, see C 200.
602 The riser system including riser pipe and interfaces,
details and components, shall be designed according to the
following basic principles:
÷ the riser system shall satisfy functional and operational
requirements as given in the design basis.
÷ the riser system shall be designed such that an
unintended event does not escalate into an accident of
significantly greater extent than the original event;
÷ permit simple and reliable installation, retrieval, and be
robust with respect to use;
÷ provide adequate access for inspection, maintenance,
replacement and repair;
÷ the riser joints and components shall be made such that
fabrication can be accomplished in accordance with
relevant recognised techniques and practice;
÷ design of structural details and use of materials shall be
done with the objective to minimise the effect
corrosion, erosion and wear;
÷ riser mechanical components shall, as far as practicable,
be designed “fail safe”. Consideration is to be given in
the design to possible early detection of failure or
redundancy for essential components, which cannot be
designed according to this principle;
÷ the design should facilitate monitoring of its behaviour
in terms of tension, stresses, angles, vibrations, fatigue
cracks, wear, abrasion, corrosion etc.
B 700 Quality Assurance and Quality System
701 The design format within this standard requires that
the possibility of gross errors (human errors) shall be
prevented by requirements to the organisation of the work,
competence of personnel performing the work and
verification activities during the design, manufacture and
fabrication phases and quality assurance during all relevant
phases.
702 A quality system shall be applied to the design,
manufacturing, fabrication, testing, operation and
maintenance activities to assist compliance with the
requirements of this standard.
Guidance note:
ISO/CD 136287 give guidance on the selection and use of
quality systems.
end  of  Guidance  note –
C. Design Format
C 100 Basic Considerations
101 The design objective is to keep the failure probability
(i.e. probability of exceeding a limit state) below a certain
value. All relevant failure modes for the riser shall be
identified and it shall be verified that no corresponding limit
state is exceeded.
102 The following design methods may be applied:
÷ Load and Resistance Factor Design (LRFD) method,
see C 300
÷ Working Stress Design (WSD) method, see C 400
÷ Reliability analysis, see C 500
÷ Design by testing, see C 600
Guidance note:
The LRFD method separates the influence of uncertainties and
variability originating from different causes by means of partial
safety factors.
The WSD method adopted herein addresses the same limit
states as the LRFD but accounts for the influence of
uncertainty in only a single usage factor. The LRFD method
allows for a more flexible and optimal design with uniform
safety level and is considered superior to the WSD method.
The WSD format is included as a more easytouse
conservative alternative.
Reliability analysis is mainly considered as applicable to
unique, special case design problems, for conditions where
limited experience exists and for (re) calibration of
safety/usage factors.
As an alternative or supplement, testing (fullscale or model)
conducted in accordance with valid experimental methods may
be used to determine or verify riser system load effects,
structural resistance and resistance against material
degradation.
 end  of  Guidance  note 
C 200 Safety Class Methodology
201 This standard gives the possibility to design risers
with different safety requirements, depending on the safety
class to which the riser belongs. The riser system shall (on a
component level if relevant) be classified into one or more
safety classes based on the failure consequences. The safety
class of a riser depends on:
÷ the hazard potential of the fluid in the riser, i.e. fluid
category;
÷ the location of the part of the riser that is being
designed;
÷ whether the riser is in operating or temporary state.
DNVOSF201 Dynamic Risers, October 2010
Section 2 Page 15
DET NORSKE VERITAS
202 Fluids in the riser system shall be categorised
according to their hazard potential as given in Table 21.
Contents not specifically identified shall be classified in the
category containing substances most similar in hazard
potential to those quoted. If the category is not evident, the
most hazardous category shall be assumed.
Table 21 Classification of fluids
Category Description
A Typical nonflammable waterbased fluids.
B Flammable and/or toxic substances which are liquids
at ambient temperature and atmospheric pressure
conditions. Typical examples would be oil,
petroleum products, toxic liquids and other liquids,
which could have an adverse effect on the
environment if released.
C Nonflammable substances which are gases at
ambient temperature and atmospheric pressure
conditions. Typical examples would be nitrogen,
carbon dioxide, argon and air.
D Nontoxic, singlephase gas which is mainly
methane.
E Flammable and toxic substances, which are gases at
ambient temperature and atmospheric pressure
conditions and which, are conveyed as gases or
liquids. Typical examples would be hydrogen,
methane (not otherwise covered under category D),
ethane, ethylene, propane, butane, liquefied
petroleum gas, natural gas liquids, ammonia, and
chlorine.
203 The riser system shall be classified into a location
class 1 and 2 as defined in Table 22.
Table 22 Classification of locations
Location Description
1 Area where no frequent human activity is anticipated
2 The part of the riser in the near platform (manned)
area or in areas with frequent human activity. The
extent of location class 2 should be based on
appropriate risk analyses. If no such analyses are
performed, a minimum horizontal distance of 500 m
may be adopted.
204 Riser design shall be based on potential failure
consequences. This is implicit by the concept of safety
classes defined in Table 23.
Table 23 Classification of safety classes
Safety
class
Definition
Low Where failure implies low risk of human injury and
minor environmental and economic consequences.
Normal For conditions where failure implies risk of human
injury, significant environmental pollution or very
high economic or political consequences.
High For operating conditions where failure implies high
risk of human injury, significant environmental
pollution or very high economic or political
consequences.
205 The safety class is a function of the riser status
(phase) and location class. For normal riser use, the safety
classes in Table 24 apply. Other classifications may exist
depending on the conditions and criticality of the riser. The
operator shall specify the safety class to which the riser shall
be designed.
Table 24 Normal classification of safety classes
3), 4), 5)
Riser content
Fluid category A,C Fluid category B Fluid category D, E
Location class Location class Location class
Riser status (phase)
1 2 1 2 1 2
Testing
1)
Low Low Low Low NA NA
Temporary with no pipeline/well access
2)
Low Low Low Low Low Normal
Inservice with pipeline/well access Low Normal Normal Normal Normal High
NOTES
1) Testing like overpull to test connection (e.g. bottom connection) and system pressure test performed with incompressible medium is classified as safety
class low.
2) Temporary conditions include handling, transportation, installation, landing, connecting, disconnection, retrieval and hangoff.
3) Riser with nonflammable content but under pressure may require to be classified as safety class Normal.
4) Risers that are pressurised in temporary condition may require to be treated as inservice risers.
5) If deemed necessary, a riser can always be designed to the requirements of a more strict safety class.
C 300 Design by LRFD Method
301 The fundamental principle of Load and Resistance
Factored Design (LRFD) method (also denoted partial safety
factor method) is to verify that factorised design load effects
do not exceed factored design resistance for any of the
considered limit states (i.e., failure modes).
302 In the LRFD approach it is distinguished between:
÷ pressure load effect (static) ;
÷ functional load effects (static) ;
÷ environmental load effects (mainly dynamic) and
÷ accidental load effects.
Guidance note:
This separation of loads is done in order to cope with sources
of uncertainties in a rational way; e.g. uncertainties in the
DNVOSF201 Dynamic Risers, October 2010
Page 16 Section 2
DET NORSKE VERITAS
environmental load effects are typically larger compared to
those in pressure or functional load effects implying a higher
safety factor
 end  of  Guidance  note 
303 The general LRFD safety format can be expressed as:
( ) 1 t ; ; ; ; R ; S ; S ; S ; S g
c m SC k A A E E F F P
≤ γ γ γ ⋅ γ ⋅ γ ⋅ γ
(2.1)
g(•) is the generalised load effect. g(•)<1 implies a safe
design and g(•)> 1 implies failure. Further,
S
P
= Pressure loads
S
F
= Load effect from functional loads ( vector or
scalar)
S
E
= Load effect from environmental load ( vector
or scalar)
S
A
= Load effect from accidental loads ( vector or
scalar)
γF
= Load effect factor for functional loads( vector
or scalar)
γE
= Load effect factor for environmental loads
γA
= Load effect factor for accidental loads
R
k
= Generalised resistance ( vector or scalar)
γSC
= Resistance factor to take into account the
safety class (i.e. failure consequence)
γm
= Resistance factor to account for material and
resistance uncertainties
γc
= Resistance factor to account for special
conditions
t = Time
Guidance note:
g(•) is a function of time for systems exposed to time varying
excitations. The timedependent generalised load effect g( •)
defined above covers the general case for combined loading.
For design criteria where the load effects and resistance can be
separated the LRFD format can be written in the more familiar
format:
( )
c m SC
k
A A E E F F P d
R
; S ; S ; S ; S S
γ ⋅ γ ⋅ γ
≤ ⋅ γ ⋅ γ ⋅ γ
The generalised load effect g( •) is discussed in more detail in
section 4.
 end  of  Guidance  note 
304 The acceptance criteria presented in this standard are
calibrated using a reliabilitybased methodology for the
different safety classes. The following comments apply:
÷ the load effect factors and resistance factors depend on
the limit state category
÷ identical load effect factors will apply to limit states and
safety classes;
÷ the set of resistance factors are adapted to the particular
failure mode being considered and safety class;
÷ an additional safety factor, γ
c
is applied where
appropriate in order to account for conditions with
specific load effects or resistances. (e.g. in case of
prevailing system effects where many pipe sections are
exposed to the same loading)
Guidance note:
Load effect factors typically account for natural variability in
loads and model uncertainties due to incomplete knowledge or
models leading to possible inaccurate calculation of load
effects.
Resistance factors typically account for variability in strength
and basic variables including the effect of dimensional
tolerances and model uncertainties due to incomplete resistance
model.
 end  of  Guidance  note 
305 The load effects and resistance in this standard are
usually given as percentile values (i.e. return period values
for load effects) of the respective probability distributions.
They shall be based on reliable data, using recognised
statistical techniques.
Guidance note:
The characteristic resistances in this standard do not
necessarily reflect either mean values or certain percentile
values. The resulting design formulas provide design criteria as
a totality of model uncertainty, bias loads etc. Hence, care shall
be taken when recalibrating these formulas to ensure this
totality.
 end  of  Guidance  note 
C 400 Design by WSD Method
401 The Working (allowable) Stress Design (WSD)
method is a design format where the structural safety margin
is expressed by one central safety factor or usage factor for
each limit state.
402 The WSD method adopted herein applies explicit
design checks similar to the LRFD method but accounts for
the influence of uncertainty in only a single usage factor.
Guidance note:
The usage factor accounts for the integrated uncertainty and
possible bias in load effects and resistance. The usage factor, η,
may be interpreted as an inverted weighted product of partial
safety factors.
The usage factor is also named Allowable Stress factor or
Design Factor in some WSD codes and standards.
 end  of  Guidance  note 
403 The general WSD design format can be expressed
as:
1 ) t , , R , S ( g
k
≤ η
(2.2)
404 where S is the total load effect, R
k
is the resistance, η
is the usage factor and ) ( g • is the generalised load effect as
discussed for the LRFD safety format. It is emphasised that
S is the total load effect (scalar or vector), due to combined
DNVOSF201 Dynamic Risers, October 2010
Section 2 Page 17
DET NORSKE VERITAS
action from pressure, functional, environmental and
accidental loads as relevant for the actual limit state and
load case.
Guidance note:
It should be observed that the generalised load effect for the
WSD formulation could be derived as a special case of the
generalised load effect for the LRFD formulation.
For design criteria where the load effect and resistance can be
separated the WSD format can be expressed in the more
familiar format:
k d
R ) S ( S η ≤
 end  of  Guidance  note 
C 500 Reliability Based Design
501 As an alternative to the design formats specified in
this standard, a probabilistic design approach based on a
recognised structural reliability analysis may be applied
provided that:
÷ it is used for calibration of explicit limit states outside
the scope of this standard;
÷ the method complies with DNV Classification Note no.
30.6 or ISO 2394;
÷ the approach is demonstrated to provide adequate safety
for familiar cases, as indicated by this standard.
502 Suitably competent and qualified personnel shall
perform the structural reliability analysis, and extension into
new areas of application shall be supported by technical
verification.
503 As far as possible, target reliability levels shall be
calibrated against identical or similar riser designs that are
known to have adequate safety based on this standard. If this
is not feasible, the target safety level shall be based on the
failure type and class as given in Table 25. The values are
nominal values reflecting structural failure due to normal
variability in load and resistance but excluding gross error.
Table 25 Acceptable failure probabilities
1)
vs. safety
class
Safety classes Limit
state
Probability
bases
2,3)
Low Normal High
SLS
4)
Annual per riser 10
1
10
1
10
2
10
2
10
3
ULS Annual per riser
FLS
5)
Annual per riser
ALS Annual per riser
10
3
10
4
10
5
NOTES
1) The failure probability from a structural reliability analysis is a
nominal value and cannot be interpreted as an expected frequency
of failure.
2) The probability basis is failures per year for permanent conditions
and for the actual period of operation for temporary conditions.
3) Per riser imply for the riser in each location class
4) The failure probabilities provided for SLS are not mandatory. SLS
are used to select operational limitations and can be defined
according to the operator’s preference. Note that exceedence of a
SLS conditions require a subsequent ALS design check.
5) The FLS probability basis is failures per year, i.e., often last year
of service life or last year before inspection.
C 600 Design by Testing
601 Testing (fullscale or model) conducted in
accordance with valid experimental methods may be used to
determine or verify riser system load effects, structural
resistance and resistance against material degradation.
Design by testing or observation of performance shall be
supported by analytical design methods.
Guidance note:
Load effect model tests are normally performed to determine
the floater responses as wave induced motions and drift
motions. In general, load effect model tests should be
considered to verify methods for predicting systems load effect
(response) for concepts with little or no field experience and
cases with high uncertainty in analysis models. These tests may
include tests for evaluation of hydrodynamic coefficients,
shielding effects, vortexinduced vibrations, interference and
soilstructure interaction i.e. for touch down regions.
Certain vital riser components and materials including seals
may, due to their specialised and unproven function, require
extensive engineering and prototype testing to determine and
confirmation of anticipated design performance including
fatigue characteristics, fracture characteristics, corrosion
characteristics, wear characteristics, mechanical characteristics.
 end  of  Guidance  note 
602 When implementing experimental test results into
design, all relevant deviations between the model test and
reality shall be considered including:
÷ scaling effects,
÷ model/testing simplifications and uncertainties,
÷ data acquisition and processing simplifications and
uncertainties,
÷ uncertainties with regard to longterm effects and
failure modes.
Statistical uncertainties with respect to a limited number of
test results are to be included in the determination of model
load effects or resistance.
DNVOSF201 Dynamic Risers, October 2010
Page 18 Section 3
DET NORSKE VERITAS
SECTION 3 LOADS
Contents
A. General
A 100 Objective
A 200 Application
A 300 Loads
B. Pressure Loads
B 100 Definition
B 200 Determination of Pressure Loads
B 300 Pressure Control System
B 400 Pressure Ratings
C. Functional Loads
C 100 Definition
C 200 Determination of Functional Loads
D. Environmental Loads
D 100 Definition
D 200 Environmental Load Condition
D 300 Waves
D 400 Current
D 500 Floater Motion
A. General
A 100 Objective
101 This section defines the loads to be considered in the
design of riser systems. The loads are classified into
different load categories.
Guidance note:
The aim of the load classification is to relate the load effect to
the different uncertainties and occurrences.
 end  of  Guidance  note 
A 200 Application
201 This section describes the loads to be applied in the
adopted LRFD criteria.
A 300 Loads
301 Loads and deformations shall be categorised into four
groups as follows:
÷ pressure (P) loads (section B);
÷ functional (F) loads, (section C);
÷ environmental (E) loads, (section D)
÷ accidental (A) loads, (section 5.F 400)
Table 31 gives some examples on how the various loads are
categorised.
Table 31 Examples of categorisation of loads
1)
Floads Eloads Ploads
7)
Weight and buoyancy
6)
of riser, tubing, coatings
6)
,
marine growth
2)
, anodes, buoyancy modules,
contents and attachments
Weight of internal fluid
Applied tension for toptension risers
Installation induced residual loads or prestressing
Preload of connectors
Applied displacements and guidance loads,
including active positioning of support floater
Thermal loads
Soil pressure on buried risers
Differential settlements
Loads from drilling operations
Construction loads and loads caused by tools
Waves
Internal waves and other effects due to
differences in water density.
Current
Earthquake
4)
Ice
3)
Floater motions induced by wind, waves
and current, i.e.:
÷ Mean offset including steady wave
drift, wind and current forces
÷ Wave frequency motions
÷ Low frequency motions
External hydrostatic pressure
Internal fluid pressure: hydrostatic,
static and dynamic
5)
contributions,
as relevant
Water Levels
NOTES
1) Accidental loads, both size and frequency, for a specific riser and floater may be defined by a risk analysis.
2) For temporary risers, marine growth can often be neglected due to the limited duration of planned operations.
3) Ice effects shall be taken into account in areas where ice may develop or drift.
4) Earthquake load effects shall be considered in the riser design for regions considered being seismically active.
5) Slugs and pressure surges may introduce global load effects for compliant configurations.
6) Includes also absorbed water.
7) Possible dynamic load effects from Ploads and Floads shall be treated as Eloads, e.g. slug flow.
DNVOSF201 Dynamic Risers, October 2010
Section 3 Page 19
DET NORSKE VERITAS
B. Pressure Loads
B 100 Definition
101 Pressure loads, P, are loads that are strictly due to the
combined effect of hydrostatic internal and external
pressures see Table 31. Such loads are often included in the
general class of functional loads, however, they are
considered separately in this standard.
102 The following internal pressure definitions apply at
the surface (top) of the riser, see Table 32:
÷ Design pressure, p
d
, is the maximum surface pressure
during normal operations.
÷ Incidental pressure, p
inc
, is the surface pressure that is
unlikely to be exceeded during the life of the riser.
Table 32 Internal pressure definitions at riser surface (top)
2)
Riser Type Design pressure, p
d
Incidental pressure, p
inc
Drilling riser above subsea BOP stack Zero Maximum diverter line back pressure
Drilling riser with surface stack Zero (or if drilling underbalanced,
maximum under –balance pressure)
Design as an extension of the last casing
string that will be drilled through. This applies
to both outer riser and inner riser, if used.
Drilling riser with both surface and
subsea BOP stacks
Zero (or if drilling underbalanced,
maximum underbalance pressure)
Surface pressure that will handle most well
control situations. Assume subsea BOP will
be closed before pressure rises higher.
Production or injection riser used as
extension of production casing
Specified maximum annulus pressure
1)
or maximum sustained pressure allowed
by regulation or company policy
Pressure caused by nearsurface leak of shut
in tubing (maximum).
Outer casing of dual casing production
or injection riser with surface tree
No requirement or specified pressure. Pressure caused by nearsurface or near
bottom leak of inner tubing/casing maximum
operating pressure.
Tubing (single pipe) riser or flowline
from subsea satellite well
Surface shutin pressure with subsea
valves open
Maximum surge pressure or maximum well
kill pressure.
Import riser from subsea manifold Surface shutin pressure with subsea
valves open unless pressure can be
reliably limited to a lower value by e.g.
a pressure reduction system (HIPPS)
Maximum surface shutin pressure with
subsea valves open unless pressure can be
reliably limited to a lower value
Export/import riser from/to pipeline Maximum export/import pressure
during normal operations
Maximum surge pressure defined with low
lifetime probability of occurrence. Normally
to be taken as 1.1*p
d
Other riser type Highest pressure that will be seen for an
extended time
Pressure that is unlikely to be exceeded during
life/period of operation of riser
NOTES
1) Annulus refers to the space between the external riser pipe and the tubing/workstring/drillstring in the case of a singlecasing
production/workover/drilling riser, or the space between the inner casing and the tubing/work string in the case of a dualcasing riser. The
content and pressure of the outer annulus for a dualcasing riser can normally be assumed constant and as specified.
2) Internal pressure may also be specified at subsea wellhead.
B 200 Determination of Pressure Loads
201 It is the responsibility of the owner to determine
design surface and incidental surface internal pressures
together with internal content density and temperature
based on the guidelines given above and Table 32. The
owner shall also specify surface operating pressure and
minimum surface stresses with corresponding temperature
and density. It may be necessary to specify pressure
temperaturedensity values (p, T, ρ), which determine an
envelope of the (p, T, ρ)  regime of the credible extreme
values.
202 The local internal design pressure p
ld
and local
incidental pressure p
li
are determined based on the
definitions given in B 100 as follows
h g p p
h g p p
i inc li
i d ld
⋅ ⋅ ρ + ·
⋅ ⋅ ρ + ·
(3.1)
where ρ
i
is the density of the internal fluid, h is the height
difference between the actual location and the internal
pressure reference point, and g is the acceleration of gravity.
Guidance note:
Gas mixed with oil in the riser could reduce the hydrostatic
internal pressure acting downstream of the closed valve. This
should be taken into account when calculating the maximum
allowable shutin pressure for the specific application
 end  of  Guidance  note 
203 The hydrostatic seawater pressure governs the
external pressure on pipes directly exposed to seawater (e.g.
single pipe risers or outer riser of multitube risers). Annual
average seawater density and mean sea levels shall be used
to establish the external hydrostatic pressure.
Guidance note:
The external pressure should not be taken as higher than the
water pressure at the considered location corresponding to low
DNVOSF201 Dynamic Risers, October 2010
Page 20 Section 3
DET NORSKE VERITAS
tide when external pressure increase the resistance and high
tide when external pressure decreases the resistance.
 end  of  Guidance  note 
204 The hydrostatic annulus pressure governs the
external pressure on the inner riser and tubing in multitube
risers. The hydrostatic annulus pressure should be defined in
terms of the density of the annulus content together with a
reference pressure at a given location (i.e., similar to internal
pressure).
B 300 Pressure Control System
301 A pressure control system may be used to prevent the
internal pressure at any point in the riser system rising to an
excessive level. The pressure control system comprises the
pressure regulating system, pressure safety system and
associated instrumentation and alarm systems, see DNV
OSF101.
B 400 Pressure Ratings
401 The local differential pressure may form the basis for
selection of pressure rated components. Pressure rated
components like valves, flanges and other equipment shall
have pressure rating not less than the surface pressure or
local overpressure of the riser.
Guidance note:
Riser components at any point along the riser should be
designed for or selected to withstand the maximum differential
pressure between internal and external pressure to which the
components will be exposed to during operating conditions.
Pressurecontrolling components (such as valve bore sealing
mechanism and tubing plugs) may be isolated from the external
ambient pressure under certain operating conditions.
In most cases, valves in subsea gas service cannot be used in
applications where the shutin pressure would exceed the
maximum rated working pressure stamped on the equipment.
Pressurecontrolling components on subsea oil wells may
benefit from “external” downstream pressure due to hydrostatic
head of the oil column in the riser. In such cases, the equipment
could be used at pressures above the marked pressure rating.
 end  of  Guidance  note 
C. Functional Loads
C 100 Definition
101 Functional loads, F, are defined as loads that occur as
a consequence of the physical existence of the system and
by operating and handling of the system, without
environmental or accidental load. Examples of functional
loads are listed in Table 31.
C 200 Determination of Functional Loads
201 The following apply when the characteristic values of
the Fload shall be determined:
÷ In the case of welldefined functional loads, the
expected value of the load shall be used. Examples are
accurate data of the riser weight, buoyancy, contents
and applied tension;
÷ In the case of variable functional loads, the most
unfavourable with respect to the combined P, F, E
loading condition shall be considered. Sensitivity
analyses should be performed to quantify criticality.
Example is change in weight due to corrosion and
effects due to marine growth (weight and effects on
hydrodynamic loading);
÷ In the case of functional load caused by deformation,
the extreme value shall be used. Example is intended
vessel offset.
Guidance note:
The effect of marine growth on riser shall be considered, taking
into account biological and other environmental phenomena
relevant for the location. Such biological and environmental
factors include water salinity, oxygen content, pH, current and
temperature.
The estimation of hydrodynamic load on risers subjected to
accumulated marine growth shall account for the increase in
effective diameter and surface roughness.
 end  of  Guidance  note 
D. Environmental Loads
D 100 Definition
101 Eloads are loads imposed directly or indirectly by
the ocean environment, see Table 31. The principal
environmental parameters are waves, currents and floater
motions.
D 200 Environmental Load Condition
201 Environmental phenomena that are relevant for the
particular location and operations in question shall be taken
into account; see Table 31. The principles and methods as
described in DNV CN 30.5 may be used as basis for
establishing the environmental load conditions.
D 300 Waves
301 Wind driven surface waves are a major source of
dynamic environmental forces on the risers. Such waves are
irregular in shape, can vary in length and height, and can
approach the riser from one or more directions
simultaneously.
302 Wave conditions may be described either by a
deterministic design wave or by stochastic methods applying
wave spectra.
Guidance note:
Most spectra is described in terms of a few statistical wave
parameters such as significant wave height, H
s
, spectral peak
period, T
p
, spectral shape and directionality.
Other parameters of interest, such as the maximum wave height
H
max
and the associated wave period T
Hmax
can be derived from
these.
 end  of  Guidance  note 
DNVOSF201 Dynamic Risers, October 2010
Section 3 Page 21
DET NORSKE VERITAS
303 The selection of appropriate wave theories depends
on the actual application and link to assumptions used for
adjacent structures e.g. floater motion transfer function
Guidance note:
Normally, linear wave theory combined with wheeler
stretching should be considered in addition to disturbed
kinematics if relevant.
For part of the riser below the splashzone linear wave theory
is usually adequate in connection with irregular seastates.
Note however that disturbed kinematics e.g. for semi
submersibles and TLP’s may effect the kinematics close to the
floater.
 end  of  Guidance  note 
304 Combination of wind driven waves and swell from
different directions must be taken into account in design.
Guidance note:
This has relevance e.g. for monohull vessels (FPSO's and Drill
Ships) where large roll motions may introduce high bending
moments due to beam swell sea in combination with wind
driven head sea.
 end  of  Guidance  note 
D 400 Current
401 The design current velocity, profile and direction
shall be selected using the best statistics available. The
resulting current velocities shall include contributions from
tidal current, wind induced current, storm surge current,
density induced current, global ocean current, eddies that
spin off from a circulating current and other possible current
phenomena.
D 500 Floater Motion
501 Floater offset and motions constitute a source of both
static and dynamic loading on the riser. The main data
regarding floater motions needed for riser designs are:
÷ static offset  mean offset due to wave, wind and current
loads ;
÷ wave frequency motions  first order wave induced
motions ;
÷ low frequency motions  motions due to wind gust and
second order wave forces ;
÷ pulldown/set down  due to the combined effect of
mooring lines/tether constraints and floater offset (e.g.
for TLP’s);
For further details, reference is made to appendix F.
DNVOSF201 Dynamic Risers, October 2010
Page 22 Section 4
DET NORSKE VERITAS
SECTION 4 ANALYSIS METHODOLOGY
Contents
A. General
A 100 Objective
A 200 Application
A 300 Riser Analysis Procedure
B. Extreme Combined Load Effect Assessment
B 100 Fundamentals
B 200 Generalised Load Effect
B 300 Load Cases
B 400 Design Based on Environmental Statistics
B 500 Design Based on Response Statistics
C. Global Analysis
C 100 General
C 200 Fatigue Analysis
A. General
A 100 Objective
101 The purpose of this section is to provide
requirements for global analysis. Focus is on assessment of
global structural load effects in connection with design
criteria specified in Section 5.
A 200 Application
201 Combined load effects from pressure, functional and
environmental loads are provided below. For accidental load
and load effects see also Section 5. F.
202 Section B considers extreme load effect assessment
for SLS, ULS and ALS while FLS is discussed in C 200.
A 300 Riser Analysis Procedure
301 An overview of the (ULS)design approach is shown
in Figure 41. The design approach may be summarised as:
÷ identify all relevant design situations and limit states,
e.g. by FMEA, HAZOP and design reviews;
÷ consider all relevant loads defined in Section 3.;
÷ perform preliminary riser design and static pressure;
design checks (bursting, hoop buckling and propagating
buckling) specified in Section 5;
÷ establish loading conditions defined in B 300;
÷ define generalised load effect for combined design
criteria defined in Section 5;
÷ conduct riser analysis using appropriate analysis models
and methods defined in C;
÷ establish extreme generalised load effect estimate based
on environmental statistics B 400, or on response
statistics, B 500.
÷ check that no relevant limit state is exceeded.
Riser system •Service life
•Diameter(s)
•Internal fluid data
•Environmental
data
•Functional
requirements
•Operational
requirements
•Vessel data
•Interfaces
•Material selection
Design basis
Load cases
Environmental Statistics
OK?
Final riser design
Interface loads
Top tensions
Yes No
Safety Class
Adjust:
−
Design
−
Vessel
−
Operational
requirements
Pressure
P
Functional
F
Environmental
E
Accidental
A
Loading Conditions
ALS SLS ULS FLS
Combined Loading
Limit State Checks
gmax <1
Load cases
Response Statistics
Riser analysis
Static Dynamic
See Chapter 5 Fatigue C200
Preliminary riser design
Pressure design checks
Updated Riser Design
Design Criteria
gmax(short term)
Short Term Assessment
of Extreme Load effects
F(gmax)=11/N
Long Term Assessment
of Extreme Load Effects
Define
Generalised Load Effect
Figure 41 Design Approach
B. Extreme Combined Load Effect
Assessment
B 100 Fundamentals
101 The characteristic load condition for SLS, ULS and
ALS limit states shall reflect the most probable extreme
combined load effect over a specified design time period.
Guidance note:
For permanent conditions, the most probable extreme
generalised load effect during D years is commonly also
denoted the Dyear return period value. A Dyear return period
value corresponds to an annual exceedence probability of 1/D.
DNVOSF201 Dynamic Risers, October 2010
Section 4 Page 23
DET NORSKE VERITAS
 end  of  Guidance note 
102 For permanent operational conditions a 100year
return period (10
2
annual exceedence probability) apply.
103 For temporary operational conditions the load effect
return period value depends on the seasonal timing and
duration of the temporary period. The return periods shall be
defined such that the probability of exceedence in the
temporary state is no greater than that of the longterm
operational state.
Guidance note:
If more information is not available the following return
period values may be applied:
− a 100 year return period if duration in excess of 6 months.
− a 10 year return period for the actual seasonal
environmental condition if duration is in excess of 3 days
but less than 6 months,.
− For temporary conditions with duration less than 3 days or
operations which can be terminated within a 3 days
window an extreme load condition may be specified and
startup /shut down of the operation is then based on
reliable weather forecasts.
 end  of  Guidance note 
B 200 Generalised Load Effect
201 For combined loading, the acceptance criteria can be
expressed by the following generic equation:
. 1 ) , , p ), t ( T ), t ( M ( g ) t ( g
k ed d
≤ Λ ∆ · R
(4.1)
202 Where g(t) is the generalised load effect and M
d
, T
ed
denote design values for bending moment and effective
tension, respectively, see Section 5. Furthermore, ∆p
denotes the local differential pressure,
k
R is a vector of
crosssectional capacities and Λ is a vector of safety factors
(i.e. material, safety class and condition factors). Such a
generic formulation covers LRFD as well as WSD
acceptance criteria for combined loading; see Appendix C
for details.
Guidance note:
The generalised load effect indicates the level of utilisation.
g(t)< 1 imply a safe design and g(t)>1 imply failure. See also
section 3.
 end  of  Guidance note
203 The code checks for combined loading is hence
equivalent to extreme value prediction (e.g. the 100 year
return period value) of the generalised load effect, i.e.
1
max
≤ g (4.2)
Guidance note:
The importance of this formulation is that the combined time
dependent action of bending moment and effective tension is
transformed into a scalar process expressed by the generalised
load effect. This approach will automatically account for the
correlation between effective tension and bending moment
components and is hence capable of optimal design (i.e. allows
for maximum utilisation).
The generalised load effect applies to design based on response
statistics in establishing the long term probability distribution
as well as shortterm extreme load assessment for design based
on environmental statistics.
In case of design based on environmental statistics the standard
framework for response processing of results from time domain
analyses can be directly applied for code checks. This will
typically include application of response envelopes in case of
regular wave analysis and statistical extreme value prediction
in case of irregular wave analysis.
Conservative shortterm estimates can be obtained by separate
estimation of design values for effective tension and resulting
bending moment disregarding correlation effects, i.e.
( ) 1 , , p , T , M g
k
max
ed
max
d
≤ Λ ∆ R
(4.3)
where the indices “max” indicate extreme values. This
approach may yield acceptable results when the design is
driven by one dominating dynamic component. An advantage
of this formulation is that it is applicable to time domain as
well as frequency domain analyses.
 end  of  Guidance note 
B 300 Load Cases
301 The load cases form the basis for riser analysis,
which determines the generalised load effects to be used for
limit states controls. An adequate set of load cases (loading
conditions) should be examined in order to:
÷ reflect extreme combined load effects;
÷ represent all relevant limit states;
÷ represent both permanent and temporary conditions ;
÷ represent the range of operating conditions and
functional applications, and
÷ study sensitivities to the variation of critical parameters
at different locations along the riser.
302 Different conditions may be selected for various
stages in the operation, depending on the duration of the
operations and the consequences of exceeding the selected
conditions.
303 Environmental load effects generally depend on the
applied Floads since Floads may influence the dynamical
properties of the system (e.g. applied top tension and mass
per unit length will influence the dynamic properties of the
system.) Sensitivity studies shall therefore be performed to
identify the most unfavourable Fload with respect to
combined load effects at critical locations
304 For operating extreme conditions for combined load
effects the pressure should be taken as the design pressure or
a minimum value whichever is the more conservative.
Guidance note:
This implies that it is assumed that the design pressure (or
minimum pressure) is likely to occur during an extreme
environmental condition.
 end  of  Guidance note 
DNVOSF201 Dynamic Risers, October 2010
Page 24 Section 4
DET NORSKE VERITAS
B 400 Design Based on Environmental Statistics
401 Design criteria based on environmental statistics may
be applied to establish characteristic load effects. A
sufficient number of loading conditions in terms of
stationary environmental conditions must be analysed in
order to capture the extreme generalised load effects for all
critical locations on the riser.
Guidance note:
It has traditionally been common practice to adopt the most
unfavourable load effect found by exposing the riser system to
multiple stationary environmental conditions as the extreme
load effect. Each design condition is described in terms of a
limited number of environmental parameters (e.g. significant
wave height, peak period etc) and a given duration (e.g. 36
hours). Different combinations of wind, waves and current
yielding the same return period (e.g. 100 years) for the
combined environmental condition are typically applied.
Furthermore, the most severe directional combination of wind,
waves and current consistent with the environmental conditions
at the actual site is normally applied.
The main challenge is that the return period for the
characteristic load effect is unknown due to the nonlinear
dynamic behaviour of most riser systems. This will in general
lead to an inconsistent safety level for different design concepts
and failure modes. Acceptable results can however be expected
for quasistatic systems with moderate nonlinearities.
Guidance to computational strategies for shortterm assessment
of extreme load effects is given in Appendix C.
 end  of  Guidance note 
402 If the design is based on environmental statistics,
verification and/or calibration of results should be
performed in the following cases:
÷ new concepts;
÷ systems with significant nonlinear response
characteristics, and
÷ dynamically sensitive systems;
The methodology in B 500 may be applied for verification
and/or calibration purposes.
403 Wave period variation shall be considered for regular
and irregular wave analyses to identify most unfavourable
loading condition. This is of special importance for regular
wave analyses, which may be subjected to severe bias for
dynamically sensitive systems. The period variation shall be
performed with due consideration of the following:
÷ statistical variation of wave period;
÷ eigenvalues of the riser system;
÷ peaks in floater motion transfer function;
÷ period dependencies in load intensity (e.g. splash zone
loads in case of disturbed kinematics).
B 500 Design Based on Response Statistics
501 Design based on response statistics is generally the
recommended procedure for consistent assessment of
characteristic load effects.
Guidance note:
Design based on response statistics is the more correct
approach and should be considered when deemed important.
Consistent assessment of the Dyear generalised load effect
will in general require a probabilistic description of the load
effect due to the longterm environmental load on the riser
system. The main challenge is to establish the longterm load
effect distribution due to the nonlinear dynamic behaviour
experienced for most riser systems.
A feasible approach for establishing long term response
statistics is proposed in Appendix C.
 end  of  Guidance note 
C. Global Analysis
C 100 General
101 Global riser analysis shall be conducted for the
specified design cases, see B 100 to check the relevant limit
states for the riser system and establish component load
effects and riser interface data. A general guidance on global
load effect analysis of risers is given in Appendix A.
102 The global analyses shall be based on accepted
principles of static and dynamics analysis, model
discretisation, strength of materials, environmental loading
and soil mechanics to determine reliable load effects on the
riser system. The load effect analysis may be based on
analytical calculations, numerical simulations or physical
testing or a combination of these methods.
103 The global riser model shall include the complete
riser system considering accurate modelling of stiffness,
mass, damping and hydrodynamic load effects along the
riser in addition to top and bottom boundary conditions. In
particular, appropriate drag and inertia coefficients for the
selected method shall be applied.
104 The riser shall be discretised with sufficient number
of elements to represent environmental loading and
structural response and to resolve load effects in all critical
areas. Time and/or frequency discretisation shall be verified
to ensure that the desired accuracy is obtained. The
principles for model validation as outlined in Appendix D
should be adopted.
105 Sensitivity studies shall be performed to investigate
the influence from uncertain system parameters (e.g. soil
data, hydrodynamic coefficients, corrosion allowance,
disturbed wave kinematics, component modelling, structural
damping etc.) The main purpose is to quantify model
uncertainties, support rational conservative assumptions and
identify areas where a more thorough investigation is
needed to achieve an acceptable modelling (e.g. calibration
of computer model against physical testing)
106 Static analyses should be carried out using a full
nonlinear approach. Several alternatives are available in
subsequent dynamic analysis restarted from the static
equilibrium configuration. Treatment of nonlinearities is the
distinguishing feature among available dynamic analysis
techniques. Knowledge of governing nonlinearities for the
DNVOSF201 Dynamic Risers, October 2010
Section 4 Page 25
DET NORSKE VERITAS
actual system as well as treatment of nonlinearities in
established analysis techniques is crucial for the accuracy
and hence the choice of adequate analysis strategy.
107 An overview of commonly used dynamic FE analysis
methods is given in Table 41. Typical application of the
main techniques for dynamic analysis is indicated in Table
42. Reference is made to Appendix A for a more detailed
discussion.
Table 41 Global analysis. Finite element (FE) methods overview
Nonlinearities Method
Environmental Loads Special loads Structure
Nonlinear Time domain
(NTD)
Slug flow.
Collision/interaction
with other slender
structures
Geometric stiffness
Nonlinear material
Seafloor contact. Variable hull contact
Large 3D rotations
Linearised Time
domain
(LTD)
Morison loading
Integration to actual surface
elevation.
NA
Linearised at static equilibrium position
Frequency domain
(FD)
Linearised at static equilibrium
position (stochastic linearisation
in case of irregular excitation)
NA
Linearised at static equilibrium position
Table 42 Typical analyses techniques versus
applications
Method Typical applications
NTD Extreme response analysis of systems with
significant nonlinearities, in particular compliant
configurations exposed to 3D excitation.
Special FLS analyses for systems or parts of systems
with highly nonlinear response characteristics (e.g.
touchdown area of compliant configurations)
Verification/validation of simplified methods (e.g.
LTD, FD)
LTD Extreme analysis of systems with small/moderate
structural nonlinearities and significantly nonlinear
hydrodynamic loading (e.g. top tensioned risers )
FD Screening analyses.
FLS analyses of systems with small/moderate
nonlinearities
108 One or combinations of the following methods
should be applied:
÷ irregular wave analysis in the time domain (design
storm);
÷ regular wave analysis in time domain (design wave);
÷ irregular wave analysis in the frequency domain
109 The irregular wave analysis refers to modelling of
water particle kinematics and floater motions. Extreme load
effect analyses should preferably be carried out by use of
time domain analyses. However, frequency domain analyses
may be applied provided that the adequacy of such analyses
is documented by verification against time domain analysis.
110 It shall be documented that the duration of irregular
time domain analyses is sufficient to obtain extreme load
effect estimates with sufficient statistical confidence. This is
of particular concern in case of combined WF and LF
loading. The methodology as outlined in Appendix C may
be applied.
111 Any use of simplified modelling and/or analysis
techniques should be verified by more advanced modelling
and/or analyses. In particular, the validation as specified in
Table 43 should be considered for representative (critical)
load cases. For further details see Appendix D.
Table 43 Validation analysis methods overview
Applied method Method for validation
Linearised time domain
analysis
Nonlinear time domain
analysis
Frequency domain analysis Time domain analysis
Regular wave analysis Irregular wave analysis
C 200 Fatigue Analysis
201 Fatigue analysis of the riser system shall consider
all relevant cyclic load effects including:
÷ first order wave effects (direct wave loads and
associated floater motions) ;
÷ second order floater motions ;
÷ thermal and pressure induced stress cycles
÷ vortex induced vibrations, see Appendix E.
÷ collisions
All modes of operations including connected, running and
hangoff must be considered if relevant.
202 The fatigue response due to the first two contributors
may be calculated with the same methods as for extreme
response calculation. If frequency domain analysis is used,
validation against irregular sea, time domain analysis shall
be performed.
DNVOSF201 Dynamic Risers, October 2010
Page 26 Section 4
DET NORSKE VERITAS
203 Fatigue analyses normally apply nominal values.
Sensitivity analysis is needed to map criticality and give
input to DFI, e.g. using half the corrosion allowance in the
cross section values for inservice assessment
204 Recommended procedures for shortterm fatigue
damage calculation for commonly used global analysis
strategies are given in Table 44. For further details see
Appendix B.
Table 44 Fatigue analysis methods overview
Method of Analysis Fatigue damage assessment
WF response LF response WFdamage LFdamage Combined WF+LF damage
FD FD NB NB Summation / bimodal
FD TD NB RFC Summation
TD TD RFC RFC Summation
TD for combined WF+LF excitation RFC for combined WF+LF response
Where :
FD = Global frequency domain analysis
TD = Global time domain analysis
WF = Wave frequency
LF = Low frequency
NB = Narrow band approximation
RFC = Rain flow cycle counting
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 27
DET NORSKE VERITAS
SECTION 5 DESIGN CRITERIA FOR RISER PIPES
Contents
A. General
A 100 Objective
A 200 Application
A 300 Limit States
B. Load Effects
B 100 Design Load Effects
B 200 Load Effect Factors
C. Resistance
C 100 Resistance Factors
C 200 Geometrical Parameters
C 300 Material Strength
D. Ultimate Limit State
D 100 General
D 200 Bursting
D 300 System Hoop Buckling (Collapse)
D 400 Propagating Buckling
D 500 Combined Loading Criteria
D 600 Alternative WSD Format
D 700 Displacement Controlled Conditions
E. Fatigue Limit State
E 100 General
E 200 Fatigue assessment using SN curves
E 300 Fatigue assessment by crack propagation
calculations
E 400 Inservice Fatigue Inspections
F. Accidental Limit State
F 100 Functional requirements
F 200 Categories of accidental loads
F 300 Characteristic accidental load effects
F 400 Design against accidental loads
A. General
A 100 Objective
101 The section provides the general framework for
design of riser systems including provisions for checking of
limit states for pipes in riser systems. Design of connectors
and riser components are covered in Section 6.
A 200 Application
201 This standard provides design checks with emphasis
on ULS, FLS, SLS and ALS load controlled conditions.
Design principles for displacement controlled conditions are
discussed in D 700.
202 Requirements for materials, manufacture, fabrication
and documentation of riser pipe, components, equipment
and structural items in the riser system are given in Section
7.
203 Mill pressure test and system pressure test shall be
performed in compliance with DNVOSF101.
A 300 Limit States
301 The limit states are grouped into the following four
categories:
÷ Serviceability Limit State (SLS) requires that the riser
must be able to remain in service and operate properly.
This limit state corresponds to criteria limiting or
governing the normal operation (functional use) of the
riser;
÷ Ultimate Limit State (ULS) requires that the riser must
remain intact and avoid rupture, but not necessary be
able to operate. For operating condition this limit state
corresponds to the maximum resistance to applied loads
with 10
2
annual exceedence probability;
÷ Accidental Limit State (ALS) is a ULS due to accidental
loads (i.e. infrequent loads)
÷ Fatigue Limit State (FLS) is an ultimate limit state from
accumulated excessive fatigue crack growth or damage
under cyclic loading.
302 As a minimum requirement, the riser pipes and
connectors shall be designed for (not limited to) the
potential modes of failures as listed in Table 51 for all
relevant conditions expected during the various phases of its
life.
DNVOSF201 Dynamic Risers, October 2010
Page 28 Section 5
DET NORSKE VERITAS
Table 51 Typical limit states for the riser system
Limit State Category Limit State Failure definition/ Comments
Clearance No contact between e.g. riserriser, risermooring line, riserhull, surface tree
floater deck, subsea treeseabed, surface jumper floater deck.
Excessive angular response Large angular deflections that are beyond the specified operational limits, e.g.
inclination of flex joint or ball joint.
Excessive top displacement Large relative top displacements between riser and floater that are beyond the
specified operational limits for top tensioned risers, e.g. stroke of telescope joint,
slick joint and tensioner, coiled tubing, surface equipment and drill floor. Note that
systems can be designed for exceeding displacement limits if the structural integrity
is maintained.
SLS
Mechanical function Mechanical function of a connector during makeup/breakout.
Bursting Membrane rupture of the pipe wall due to internal overpressure only.
Hoop buckling (collapse) Gross plastic deformation (crushing) and/or buckling (collapse) of the pipe cross
section caused by external overpressure only.
Propagating buckling Propagating hoop buckling initiated by hoop buckling.
Gross plastic deformation
and local buckling
Gross plastic deformation (rupture/crushing) of the pipe crosssection in
combination with any local buckling of pipe wall (wrinkling) due to bending
moment, axial force and internal overpressure.
Gross plastic deformation,
local buckling and hoop
buckling
Gross plastic deformation and hoop buckling of the pipe cross section and/or local
buckling of the pipe wall due to the combined effect of external overpressure,
effective tension and bending moment.
Unstable fracture and gross
plastic deformation
Unstable crack growth or rest ligament rupture or cross section rupture of a cracked
component.
Liquid tightness Leakage in the riser system including pipe and components.
ULS
Global buckling Overall column buckling (Euler buckling) due to axial compression (negative
effective tension).
ALS Same as ULS and SLS Failure caused by accidental loads directly, or by normal loads after accidental
events (damage conditions).
FLS Fatigue failure Excessive Miner fatigue damage or fatigue crack growth mainly due to
environmental cyclic loading, directly or indirectly. Limiting size of fatigue cracks
may be wall thickness (leakage) or critical crack size (unstable fracture/gross plastic
deformation).
B. Load Effects
B 100 Design Load Effects
101 Design load effects are obtained by multiplying the
load effect of each category by their corresponding load
effect factor. Specific examples are given below for bending
moment and effective tension.
102 Design bending moment for functional and
environmental induced load effects:
A A E E F F d
M M M M ⋅ γ + ⋅ γ + ⋅ γ · (5.1)
where:
M
F
= Bending moment from functional loads
M
E
= Bending moment from environmental loads
M
A
= Bending moment from accidental loads
103 Design effective tension for functional and
environmental induced load effects:
eA A eE E eF F ed
T T T T ⋅ γ + ⋅ γ + ⋅ γ · (5.2)
where
T
eF
= Effective tension from functional loads
T
eE
= Effective tension from environmental loads
T
eA
= Effective tension from accidental loads
Guidance note:
Accidental loads are included in the above design load effects
for completeness. Normally, F+E loads and A loads is not
considered simultaneously in global analyses
 end  of  Guidance  note 
104 The effective tension, T
e
is given by, see Appendix
A: (tensile force is positive):
e e i i W e
A p A p T T + − ·
(5.3)
Where
T
w
= True wall tension (i.e. axial stress resultant
found by integrating axial stress over the
crosssection)
p
i
= Internal (local) pressure
p
e
= External (local) pressure
A
i
= Internal crosssectional area
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 29
DET NORSKE VERITAS
A
e
= External crosssectional area
B 200 Load Effect Factors
201 The design load effect is used in the design checks.
Several combinations may have to be checked when load
effects from several load categories enter one design check.
The load effect factors shown in Table 52 shall be used
wherever the design load effect is referred to for all limit
states and safety class.
Table 52 Load effect factors
Fload
effect
Eload
effect
Aload
effect
Limit state
F
γ
1)
E
γ
2)
A
γ
ULS 1.1
1)
1.3
2)
NA
FLS 1.0 1.0 NA
SLS & ALS 1.0 1.0 1.0
NOTES
1) If the functional load effect reduces the combined load effects,
γF shall be taken as 1/1.1.
2) If the environmental load effect reduces the combined load effects,
γE shall be taken as 1/1.3.
C. Resistance
C 100 Resistance Factors
101 The following resistance factors apply, (see Section
2.C):
÷ safety class factor γ
SC
linked to the actual safety class
and accounts for the failure consequence.
÷ material resistance factor γ
m
to account for material and
resistance uncertainties
÷ a condition factor γ
c
to account for special conditions
specified explicitly at the different limit states where
relevant, see e.g. Table 511.
102 Unless otherwise stated, the resistance factors
applicable to all limit states are specified in Table 53 and
Table 54.
Table 53 Safety class resistance factor γ
SC
Low Normal High
1.04 1.14 1.26
Table 54 Material resistance factor γ
m
ULS & ALS SLS & FLS
1.15 1.0
Guidance note:
For SLS, the set of resistance factors can be defined by the
owner, see G.
For ALS, the set of safety factors depends on the frequency of
occurrence and is to be defined from case to case, see F. In
cases, where the inherent uncertainty related to the accidental
load is negligible and, where a conservative estimate is applied,
the material resistance factor in Table 54 can be reduced to
1.05.
end  of  Guidence  note
C 200 Geometrical Parameters
201 The nominal outside diameter D applies in resistance
calculations for all failure modes.
202 For burst and collapse pressure design checks (i.e. D
200 and D 300) the resistance shall be calculated based on
wall thickness as follows:
Mill pressure test and system pressure test condition
fab nom 1
t t t − · (5.4)
Operational condition
corr fab nom 1
t t t t − − · (5.5)
where:
t
nom
= Nominal (specified) pipe wall thickness
t
fab
= Fabrication (manufacture) negative
tolerance
t
corr
= Corrosion/wear/erosion allowance
203 Resistances for all other limit states related to
extreme loading shall be calculated based on wall thickness
as follows:
Installation/retrieval and system pressure test
nom 2
t t · (5.6)
Otherwise
corr nom 2
t t t − · (5.7)
204
Guidance note:
t
1
is the minimum wall thickness and is relevant for design
checks where failure is likely to occur in connection with a low
capacity. t
2
is used for design checks governed by the external
loading and failure is likely to occur in connection with an
extreme load effect at a location with average thickness.
 end  of  Guidance  note 
205 Variation in pipe wall thickness over the design life
of the riser system shall be considered in longterm fatigue
damage calculations (i.e. inplace, operational condition).
An average representative pipe wall thickness may be
applied in nominal fatigue stress calculations .The following
approximation may be applied for a stationary corrosive
environment:
corr nom 3
t 5 . 0 t t ⋅ − · (5.8)
For fatigue damage calculations prior to permanent
operation (e.g. towout, installation etc) the pipe wall
thickness shall be taken as:
DNVOSF201 Dynamic Risers, October 2010
Page 30 Section 5
DET NORSKE VERITAS
nom 3
t t · (5.9)
C 300 Material Strength
301 The characteristic material strength to be used in the
resistance calculations f
k
is given by:
Tensile circumferential material strength
,
`
.

·
15 . 1
f
, f min f
u
y k
(5.10)
Compressive circumferential material strength
fab y k
f f α ⋅ · (5.11)
Longitudinal material strength
C y k
f f α ⋅ · (5.12)
Where f
y
and f
u
denote the characteristic yield and tensile
strength given in Table 55. Further, α
fab
is a fabrication
factor given by 305 and α
c
is a strain hardening factor given
by 306. Note that α
c
is a function of the pressure among
others.
Table 55 Characteristic yield and tensile strength
Yield stress Tensile strength
( )
U temp , y y
f SMYS f α ⋅ − · ( )
U temp , u u
f SMTS f α ⋅ − ·
Where
SMYS is the Specified Minimum Yield Stress at room
temperature based on the engineering stressstrain
curve.
f
y,temp
is the temperature derating factor for the yield stress;
see 302.
SMTS is the Specified Minimum Tensile Strength at room
temperature based on the engineering stressstrain
curve.
f
u,temp
is the temperature derating factor for the tensile
strength; see 302.
α
U
is the material strength factor, see 304
Guidance note:
For reeling the effect of plastic straining after the pipe mill
shall be evaluated and included in the material property.
 end  of  Guidance  note 
302 The material strength (SMYS, SMTS) is normally
specified at room temperature. Possible influence on the
material properties from the temperature shall be considered
at temperatures above room temperature. This includes:
÷ yield strength, i.e. f
y,temp
÷ tensile strength, i.e. f
u,temp
÷ Young's modulus;
÷ thermal expansion coefficient.
303 Derated material properties at design temperatures
shall be established as input to the design and verified under
manufacture.
Guidance note:
If no other information on derating temperature effects of the
yield strength exists the recommendations for CMn steel, 22Cr
Duplex or 25Cr Duplex stainless steel in Figure 51 below may
be used.
50
140
S
t
r
e
s
s
D
e

R
a
t
i
n
g
(
M
P
a
)
120
100
80
60
40
20
0
160
Temperature C
0 100 150 200
180
20
22Cr
25Cr
CMn
Figure 51 Derating values for yield strength
Likewise, low temperature effects, e.g. during blown down in
gas risers, should be considered when establishing mechanical
and physical material properties.
 end  of  Guidance  note 
304 The material selection may include selection of
supplementary requirement U according to DNV OSF101.
The supplementary requirement ensures increased
confidence in material strength, which is reflected in a
higher material strength factor α
U
, given in Table 56.
Table 56 Material strength factor α
u
Normal Supplementary requirement U
0.96 1.00
Supplementary requirement U has a testing regime which
shall ensure that SMYS is at least 2 standard deviations
below the mean yield strength and that SMTS is at least 3
standard deviations below the mean tensile strength.
Guidance note:
The increased utilisation may be applied for connectors made
of forging and bolts provided an equivalent testing scheme is
adopted
 end  of  Guidance  note 
305 A fabrication factor α
fab
applies to the design
compressive circumferential yield strength for hoop
buckling, local buckling and propagating buckling limit
states. Unless otherwise documented, the fabrication factor
α
fab
in Table 57, applies for pipes manufactured by the
UOE, UO or three roll bending (TRB) or similar cold
deforming processes. Beneficial effect on this reduction
factor due to heat treatment is allowed if documented.
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 31
DET NORSKE VERITAS
Table 57 Fabrication factor α
fab
Compressive strength for welded pipe Tensile strength or
seamless pipe
UOE/ UO/TRB
1.00 0.85 0.925
306 α
c
is a parameter accounting for strain hardening and
wall thinning given by:
( )( )
( )
¹
¹
¹
'
¹
>
−
·
¹
¹
¹
'
¹
>
< < − +
< +
· β
⋅ β + β − · α
else 0
p p for
3
2
) t ( p
p p
q
60 t / D for 0
60 t / D 15 for 45 / t / D 60 q 4 . 0
15 t / D for ) q 4 . 0 (
f
f
) 1 (
e ld
2 b
e ld
h
2
2 2 h
2 h
y
u
c
(5.13)
p
ld
is the local design pressure defined in Section 3, p
e
is the
external pressure and p
b
is the burst resistance given in D
200.
α
c
is not to be taken larger than 1.20. α
c
is for illustration
purpose given in Figure 52 in case of (f
u
/f
y
) = 1.18.
1.00
1.04
1.08
1.12
1.16
1.20
1.24
0 10 20 30 40 50 60 70
D/t
α
c
0
0.2
0.4
0.6
0.8
qh
Figure 52 α
c
versus D/t ratio and pressure ratio q
h
for
(f
u
/f
y
) = 1.18
D. Ultimate Limit State
D 100 General
101 The riser pipe shall be designed against relevant
modes of failure listed in Table 51.
102 This section provides design checks with emphasis
on load controlled conditions. Design principles for
displacement controlled conditions are discussed in D 700.
103 If the design is based on:
÷ load controlled (LC) conditions
÷ design loads based on global riser analysis
÷ linear elastic and ductile materials,
accumulated plastic deformation is considered unlikely and
“shakedown” can automatically be assumed
Guidance note:
A high degree of compatibility with the DNVOSF101
Submarine Pipeline Systems has been attempted where
relevant. In general the same limit states apply for pipeline
systems and dynamic riser systems but the governing failure
modes differ due to different functional requirements between
pipelines and risers.
Pressure and functional loads normally govern wall thickness
sizing for pipelines while extreme environmental loads and
fatigue govern typical dynamic riser design.
The following comments apply to this standard in relation to
DNVOSF101:
− load combination a) in DNVOSF101, section 5.D 300 is
not required for dynamic risers. Further, γ
p
=1.0 herein;
− the additional safety class resistance factors for pressure
containment for compliance with ISO is not required for
dynamic risers. For compliance, see DNV0SF101.
− hoop buckling collapse criterion is formulated in terms of
the minimum (t
1
) rather than nominal (t
2
) thickness;
− the propagating buckling criteria is similar but may be
relaxed if the buckle is allowed to travel a short distance;
− anisotropy is not considered explicitly but the effect is
implicit in the combined loading criteria for internal
overpressure.
In addition a few minor differences exist.
 end  of  Guidance  note 
D 200 Bursting
201 Pipe members subjected to net internal overpressure
shall be designed to satisfy the following condition at all
cross sections:
( )
SC m
1 b
e li
) t ( p
p p
γ ⋅ γ
≤ −
(5.14)
where:
p
li
= Local incidental pressure, see Section 3
p
e
= External pressure
The burst resistance p
b
is given by:
,
`
.

⋅
−
⋅
⋅ ·
15 . 1
f
; f min
t D
t 2
3
2
) t ( p
u
y b
(5.15)
t is a “dummy variable” to be substituted by t
1
or t
2
where
relevant.
202 The local incidental pressure, p
li
is the maximum
expected internal pressure with a low annual exceedence
probability, see Section 3. Normally the incidental surface
pressure, p
inc
is taken 10% higher than the design pressure,
p
d
, i,e.:
d ld li
p 1 . 0 p p ⋅ + · (5.16)
where:
DNVOSF201 Dynamic Risers, October 2010
Page 32 Section 5
DET NORSKE VERITAS
p
ld
= Local internal design pressure, see 3.B 200
203 The burst criterion is valid if the mill pressure test
requirement in DNVOSF101 has been met. If not, a
corresponding decreased utilisation shall be applied.
Guidance note:
The burst criterion is expressed in terms of the resistance for
capped pipe ends. Note that the burst criterion is formulated in
terms of the local incidental pressure rather than a local design
pressure. Hence, the bursting limit state designs explicitly
against the extreme pressure loading condition over the lifetime
in compliance with standard ULS design checks. The allowable
utilisation is however in compliance with recent industry
practice for wellknown riser types.
The nominal thickness is given by:
fab corr 1 nom
t t t t + + ·
when the negative fabrication thickness tolerance is absolute,
t
fab
, and
) t % 1 /( ) t t ( t
fab corr 1 nom
− + ·
when the negative fabrication thickness tolerance is given as a
percentile of the nominal thickness, % t
fab
.
The minimum required wall thickness for a straight pipe
without allowances and tolerances is given by:
( )
1
p p
15 . 1
f
; f min
3
4
D
t
e li SC m
u
y
1
+
− γ γ
,
`
.

⋅
·
 end  of  Guidance  note 
D 300 System Hoop Buckling (Collapse)
301 Pipe members subjected to external overpressure
shall be designed to satisfy the following condition:
( )
m SC
1 c
min e
) t ( p
p p
γ ⋅ γ
≤ −
(5.17)
Where p
min
is a minimum internal pressure.
302 The resistance for external pressure (hoop buckling),
p
c
(t),
is given by:
( ) ( )
t
D
f ) t ( p ) t ( p ) t ( p ) t ( p ) t ( p ) t ( p ) t ( p
0 p el c
2
p
2
c el c
⋅ ⋅ ⋅ ⋅ · − ⋅ −
(5.18)
Solution of the equation above can be found in DNVOS
F101. The elastic collapse pressure (instability) of a pipe is
given by:
2
3
el
1
D
t
E 2
) t ( p
ν −
,
`
.

⋅ ⋅
·
(5.19)
The plastic collapse pressure is given by:
fab y p
f
D
t
2 ) t ( p α ⋅ ⋅ ·
(5.20)
303 The initial departure from circularity of pipe and pipe
ends, i.e., the initial ovality is given by:
D
D D
f
min max
0
−
·
(5.21)
304 The initial ovality shall not be taken less than 0.005
(0.5%). Maximum ovality from fabrication is given in
section G 200. Ovalisation caused during the construction
and installation phase is to be included in the ovality. The
ovalisation due to external pressure or moment in the as
installed position shall not be included.
Guidance note:
p
min
is the local minimum internal pressure taken as the most
unfavourable internal pressure plus static head of the internal
fluid. For installation p
min
equals zero. For installation with
waterfilled pipe, p
min
equals p
e
.
 end  of  Guidance  note 
D 400 Propagating Buckling
401 To ensure that a possible local buckle remains local
and does not lead to successive hoop buckling (collapse) of
neighbouring pipe sections a propagating buckling
(collapse) check is required:
( )
SC m c
pr
min e
p
p p
γ γ γ
≤ −
(5.22)
where
c
γ = 1.0 if no buckle propagation (once initiated) is
allowed. If the buckle is allowed to travel a short distance
(where the neighboring pipe section acts as buckle arrestors)
c
γ may be reduced to 0.9.
The resistance against buckling propagation, p
pr
, is given by:
5 2
2
fab y pr
D
t
f 35 p
,
`
.

⋅ α ⋅ ⋅ ·
(5.23)
402 If the pipe design is sufficient to meet the above
propagation criterion, the system hoop buckling (collapse)
criterion is also met. If conditions are such that propagating
buckles are possible, means to prevent or arrest them should
be considered in the design.
Guidance note:
For a pipe designed to meet the hoop buckling (external
collapse) criteria outlined above, hoop buckling may still be
initiated at a lower pressure by accidental means. Examples of
such means would be impact or excessive bending due to
tensioner failure. Once initiated, such a collapse may form a
propagating buckle that will travel along the pipe until the
external pressure drops below the propagation pressure or until
a change in property arrests the buckle. The consequences of
such a failure should be evaluated.
If buckle arrestors are in pipe sections subjected to fatigue, any
fatigue degradation should be evaluated due to stress
concentration factors.
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 33
DET NORSKE VERITAS
Connectors and riser joints may be considered equivalent to
buckle arrestors, i.e. it may not be necessary to design the riser
for propagating buckling.
 end  of  Guidance  note 
D 500 Combined Loading Criteria
501 Pipe members subjected to bending moment,
effective tension and net internal overpressure shall be
designed to satisfy the following equation:
{ ¦ 1
) t ( p
p p
T
T
) t ( p
p p
1
M
M
2
2 b
e ld
2
k
ed
2
2 b
e ld
k
d
m SC
≤
,
`
.
 −
+
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

+
,
`
.

,
`
.
 −
− ⋅ γ ⋅ γ
(5.24)
where:
M
d
= Design bending moment, see B 100
T
ed
= Design effective tension, see B 100
p
ld
= Local internal design pressure, see 3.B 200
p
e
= Local external pressure
M
k
is the (plastic) bending moment resistance given by:
( )
2
2
2 c y k
t t D f M ⋅ − ⋅ α ⋅ ·
(5.25)
T
k
is the plastic axial force resistance given by:
( )
2 2 c y k
t t D f T ⋅ − ⋅ π ⋅ α ⋅ · (5.26)
p
b
(t
2
) is the burst resistance given by Eq. (5.15).
Guidance note:
The failure modes controlled by this limit state comprise
yielding, gross plastic deformation and wrinkling due to
combined loading.
The design criterion may be viewed as a (plastic) Von Mises
criterion in terms of cross sectional forces and plastic cross
sectional resistance. It is equivalent to the plastic limit bending
moment capacity (including the effect of strain hardening and
wall thinning) for (T
ed
/T
k
) <<1. It reduces to the traditional
wall thickness Von Mises criterion, see e.g. API RP 2RD, for
pressure and effective tension load effects only.
 end  of  Guidance  note 
502 Pipe members subjected to bending moment,
effective tension and net external overpressure shall be
designed to satisfy the following equation:
{ ¦ { ¦ 1
) t ( p
p p
T
T
M
 M 
2
2 c
min e
2
m SC
2
2
k
ed
k
d
2
m SC
≤
,
`
.
 −
γ ⋅ γ +
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

+
,
`
.

γ ⋅ γ
(5.27)
Where the hoop buckling capacity p
c
(t) is given by Eq.
(5.18).
Guidance note:
The failure modes controlled by this semiempirical limit state
is yielding and combined local buckling and hoop buckling due
to combined bending, tension and external overpressure.
System effects should be considered for installation methods
involving many pipe sections being exposed to a similar
loading condition. If detailed information is not available a
condition factor γ
C
=1.05 multiplied with γ
SC
γ
m
apply.
 end  of  Guidance  note –
D 600 Alternative WSD Format
601 As a more easytouse alternative the following
Working Stress Design (WSD) format may be used for the
combined loading check for pipes with D/t ratio less than
30. The present WSD is based on explicit limit states for
combined loading and provides results on the conservative
side compared to the corresponding LRFD limit states.
602 For the WSD format the design load effects equals
the corresponding characteristic load effect, i.e. the load
effect factors and resistance factors equals unity: γ
F
=γ
E
=γ
A
=γ
SC
=γ
m
=1.0. Instead, the basic usage factor shown in
Table 58 apply:
Table 58 Usage factor η for combined
loading
Low Normal High
0.83 0.79 0.75
603 Pipe members subjected to bending moment,
effective tension and net internal overpressure shall be
designed to satisfy the following equation:
2
2
2 b
e ld
2
k
e
2
2 b
e ld
k
) t ( p
p p
T
T
) t ( p
p p
1
M
M
η ≤
,
`
.
 −
+
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

+
,
`
.

,
`
.
 −
− ⋅
(5.28)
where all parameters are defined in D 500.
604 Pipe members subjected to bending moment,
effective tension and net external overpressure shall be
designed to satisfy the following equation:
4
2
2 c
min e
2
2
k
e
k
) t ( p
p p
T
T
M
M
η ≤
,
`
.
 −
+
,
`
.

,
`
.

+
,
`
.

(5.29)
D 700 Displacement Controlled Conditions
701 Loads and load effects may be classified as follows:
÷ Load Controlled conditions (LC or primary), or
÷ Displacement Controlled conditions (DC secondary) or
÷ combined load types.
702 A loadcontrolled condition is one in which the
structural response is primarily governed by the imposed
loads.
703 A displacementcontrolled condition is one in which
the structural response is primarily governed by imposed
geometric displacements.
DNVOSF201 Dynamic Risers, October 2010
Page 34 Section 5
DET NORSKE VERITAS
704 Displacement controlled conditions should be sub
divided into:
÷ conditions with static (functional and pressure) loads;
÷ conditions with dynamic (environmental) loads
705 In static DC loading conditions the following
fundamental design principles apply:
÷ the primary load effect (i.e., LC part of the load effect )
shall fulfil the load controlled criteria in this standard
ignoring the secondary load effects (i.e., DC part of
load effect);
÷ the total (primary and secondary) load effect must be
checked against the strain limits and acceptance criteria
for displacement controlled conditions in DNVOS
F101;
÷ accumulated plastic deformation must be considered.
706 In dynamic DC loading conditions (lowcycle)
fatigue often becomes the limiting condition for extreme
loading conditions. A more rational and fundamental design
principle is to require that inelastic displacements caused by
cyclic loads is not allowed. Hence, the total strain must be
confined to the elastic region.
707 If the bending moment can be assumed secondary a
condition factor γ
c
=0.85 may be multiplied on the bending
moment in D 500 and D 600.
Guidance note:
Examples where bending stress may be considered secondary:
− a riser bent into conformity with a continuous curved
structure such as a reel.
− in areas where the geometric equilibrium shape of the riser
is not influenced by the bending stiffness (i.e. governed by
the geometric stiffness due to the effective tension).
The latter must be documented by analysis with and without
bending stiffness for both static and dynamic loading
conditions.
 end  of  Guidance  note 
708 Displacement controlled conditions must be
documented. Pipe sections and components subjected to
inelastic deformations shall be designed with due
consideration of accumulated plastic deformation
(ratcheting) such as incremental hoop buckling
(accumulated ovality) and plastic (low cycle and ultra low
cycle) fatigue.
E. Fatigue Limit State
E 100 General
101 The riser system shall have adequate safety against
fatigue within the service life of the system. Reference is
made to section 4 and Appendix B for more details with
respect to fatigue design and analysis.
102 All cyclic loading imposed during the entire service
life, which have magnitude and corresponding number of
cycles large enough to cause fatigue damage effects, shall be
taken into account. Temporary phases like transportation,
towing, installation, running and hangoff shall be
considered.
103 All critical sites for anticipated crack initiation for
each unique component along the riser shall be evaluated.
These sites normally include welds and details that causes
stress concentrations.
104 The fatigue assessment methods may be categorised
into:
÷ methods based on SN curves (see E 200);
÷ methods based on fatigue crack propagation;
calculations (see E 300).
105 Normally, the methods based on SN curves are used
during design for fatigue life assessment. Fatigue crack
propagation calculations may be used to estimate fatigue
crack growth life and to establish NDT inspection criteria to
be applied during both fabrication and inservice.
106 If representative fatigue resistance data are not
available, a direct fatigue testing of the actual components
shall be performed with due regard of the chemical
composition of the internal and external environment.
107 The stress to be considered for fatigue damage
accumulation in a riser is the cyclic (i.e., timedependent)
principal stress.
108 The governing cyclic nominal stress component, σ
for pipes is normally a linear combination of the axial and
bending stress given by:
( )
( )
( ) ( )
4
3
4
3
3 3
e
2·t D D
t D M 32
t t D
T
− − ⋅ π
− ⋅ ⋅
+
⋅ − ⋅ π
· σ
(5.30)
109 This combined stress varies around the
circumference of the riser pipe. For cases where the waves
are incident from several different directions, the fatigue
damage must hence be calculated at a number of regularly
spaced points to identify the most critical location.
E 200 Fatigue assessment using SN curves
201 When using the calculation methods based on SN
curves, the following shall be considered:
÷ assessment of shortterm distribution of nominal stress
range ;
÷ selection of appropriate SN curve ;
÷ incorporate thickness correction factor;
÷ determination of stress concentration factor (SCF) not
included in the SN curve, see e.g. DNVRPC203
÷ determination of accumulated fatigue damage D
fat
over
all short term conditions.
202 The fatigue criterion, which shall be satisfied, may be
written:
0 . 1 DFF D
fat
≤ ⋅ (5.31)
where
D
fat
= Accumulated fatigue damage (Palmgren
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 35
DET NORSKE VERITAS
Miner rule)
DFF = Design fatigue factor, see Table 59
Table 59 Design fatigue factors DFF
Safety class
Low Normal High
3.0 6.0 10.0
203 The design SN curve shall be based on the mean
minustwostandard deviations curves for the relevant
experimental data, see DNVRPC203.
E 300 Fatigue assessment by crack propagation
calculations
301 A damage tolerant design approach applies. This
implies that the riser components shall be designed and
inspected so that the maximum expected initial defect size
would not grow to a critical size during service life or time
to first inspection. Crack propagation calculations typically
contain the following main steps:
÷ determination of longterm distribution of nominal
stress range ;
÷ selection of the appropriate crack growth law with
appropriate crack growth parameters. Crack growth
parameters (characteristic resistance) shall be
determined as mean plus 2 standard deviations.
÷ estimation of the initial crack size and geometry and/or
any possible time to crack initiation. Best estimate
initial crack size (mean value) shall be applied. Crack
initiation time is normally neglected for welds ;
÷ determination of cyclic stress in the prospective crack
growth plane. For nonwelded components the mean
stress shall be determined;
÷ determination of final or critical crack size (through the
thickness, unstable fracture/gross plastic deformation) ;
÷ integration of the fatigue crack propagation relation
with respect to the longterm stress range distribution to
determine the fatigue crack growth life.
302 The fatigue crack growth life shall be designed and
inspected to satisfy the following condition:
0 . 1 DFF
N
N
cg
tot
≤ ⋅
(5.32)
where:
N
tot
= total number of applied stress cycles
during service or to inservice inspection
N
cg
= Number of stress cycles necessary to
increase the defect from the initial to the
critical defect size
DFF = Design fatigue factor, see Table 59.
303 The assumed initial defect size, a
i
/2c
i
for surface
defects and 2a
i
/2c
i
for embedded defects, is the expected
value of defects left after fabrication and NDT. The
expected initial defect size (mean value) shall be established
based on an evaluation of the detection capability of the
inspection method, access for inspection during fabrication,
the thickness and geometry of the structure, manufacture
method, surface finish, welding method, full or partial
penetration weld and the number of passes used to complete
the weld.
304 The maximum acceptable initial crack size may be
used to evaluate detection limits of NDT methods for the
actual component.
Guidance note:
For surface cracks starting from the transitions between
weld/base material, a crack depth of 0.1 mm (e.g. due to
undercuts and microcracks at bottom of undercuts) may be
assumed if other documentation about crack depth is not
available. The surface crack depth to total defect length (a
i
/2c
i
)
should be assumed low (less than 1:5) if no other
documentation is available. Light grinding of hot spot areas
should be considered to remove undercuts and increase
reliability of the inspection, see Appendix B.
For single sided girth welds, lack of penetration defects is hard
to detect by NDT. Crack depths in the range of 1 to 2 mm may
be hard to find. Using a reliable welding procedure is important
for such cases, especially for the root pass. Machining off the
root pass is considered to significantly improve the fatigue
quality.
Some codes have reduced life requirements for fatigue crack
growth vs. SN, e.g. a factor of 5 for fatigue crack growth
versus 10 for SN. Note that these codes defines the initial
crack size to be based on the 90 % probability of inspection
level for the applied NDT method and not the mean level as
applied in this standard.
 end  of  Guidance  note 
E 400 Inservice Fatigue Inspections
401 The SN curve approach may be used for screening
purposes to identify the most likely regions where fatigue
cracks may appear during service. Time to first inservice
inspection may be based on crack growth versus time results
with the criteria given in Table 59 in combination with
fabrication/installation records. The inservice inspection
plans after first inspection shall be based on the inspection
results obtained and the plans updated accordingly. For
defects found, fatigue crack calculations to establish residual
life shall be based on the sizing accuracy of the applied
method and the expected value shall be used for fatigue
assessment.
402 Necessary data shall be logged during the life cycle
for documenting and analysis of fatigue status for temporary
risers. The log shall typically include running sequence of
joints, riser configuration, field data (water depth, pressure,
density, etc.), floater data including top tension and the
length of time and seastate for each mode of operation. This
log shall be reviewed regularly to assess the need for fatigue
crack inspections.
403 Inplace NDT or removal of the riser for dry
inspection is considered acceptable means of inspection.
DNVOSF201 Dynamic Risers, October 2010
Page 36 Section 5
DET NORSKE VERITAS
F. Accidental Limit State
F 100 Functional requirements
101 The Accidental Limit State (ALS) is a limit state due
to accidental loads or events. Accidental loads shall be
understood as loads to which the riser may be subjected in
case of abnormal conditions, incorrect operation or technical
failure. Accidental loads typically results from unplanned
occurrences. Normally, the following design checks apply:
÷ resistance against direct accidental load. (Typically
discrete events with an annual frequency of occurrence
less than 10
2
);
÷ ultimate resistance and consequence assessment due to
exceedence of a SLS introduced to define operational
limitations;
÷ postaccidental resistance against environmental loads
(if the resistance is reduced by structural damage caused
by the accidental loads).
102 Relevant failure criteria and accidental loads in terms
of frequency of occurrence and magnitude shall be
determined based on risk analyses and relevant accumulated
experiences. Account shall be taken of other loads that
might reasonably be present at time of the accidental event.
Further, accidental loads shall be determined with due
account of the factors of influence. Such factors may be
personnel qualifications, operational procedures, the
arrangement of the installation, equipment, safety systems
and control procedures.
F 200 Categories of accidental loads
201 Accidental loads may be categorised into (not limited
to):
÷ fires and explosions
÷ impact/collisions, such as:
÷ infrequent riser interference (see H 100)
÷ impact from dropped objects and anchors
÷ impact from floater/floating objects
÷ hook/snag loads, such as :
÷ dragging anchor
÷ failure of support system, such as :
÷ heave compensating system malfunction (loss or
stuck), e.g. tension system or draw works motion
compensator
÷ loss of buoyancy, e.g. air cans for spar units
÷ loss of mooring line, tendon or guidewire
÷ dynamic positioning (DP) failure (driveoff or drift
off)
÷ exceedence of incidental internal overpressure:
÷ loss of pressure safety system
÷ failure of well tubing or packers, etc.
÷ pressure surge
÷ well kill – bullheading
÷ environmental events
÷ earthquake
÷ tsunamis
÷ iceberg
Guidance note:
Environmental load conditions with a 10 000 year return period
as a normal “tail” behaviour in the long term probability
distribution function is implicit in the ULS design criteria and
need not be considered as an accidental (or abnormal) load
condition for risers.
Accidental environmental events should be assessed assuming
1) a return period value with reasonable likelihood of not being
exceeded during the design life (e.g. 200 years) and 2) a rare
intense event (e.g. earthquake) with recurrence interval from
several hundred to a few thousand year.
 end  of  Guidance  note 
F 300 Characteristic accidental load effects
301 Accidental loads and load effects are determined by
the frequency of occurrence and their magnitude. Loads
occurring at the time of an accidental event do not normally
need to be assumed concurrent with an extreme
environmental load condition. However, the damaged
structure resulting from an accidental load event shall be
able to resist relevant pressure and functional loads in an
extreme environmental load condition. Characteristic
accidental load effects and load combinations for different
operating modes are given in Table 510.
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 37
DET NORSKE VERITAS
Table 510 Characteristic accidental load effects and combinations for different operational modes
Load effect category
Limit State Category
Mode of
Operation
Ploads Floads Eloads Aloads
Not Operating Expected value
ALS
Intact
Structure
Operating
Characteristic design
pressure or incidental as
suitable.
Expected value
associated with
the Aloads.
Expected specified
or expected extreme
value associated
with the Aloads.
Value dependent on
measures taken and
magnitude and
probability of
occurrence.
Temporary,
Not Operating
Expected value
ALS
Damaged
Structure
Operating
Characteristic design
pressure or incidental as
suitable.
1
Combined load effect defined with an
annual probability of exceedence = 10
1
Not applicable
NOTE
1) Eloads may be determined on weather forecast if time to repair is short and protective measures can be taken. If the repair period is confirmed to a
season, the probability of exceedance may be relaxed i.e., Eloads may relate to a season rather than a year.
F 400 Design against accidental loads
401 The design against accidental loads may be done by
direct calculation of the effects imposed by the loads on the
structure, or indirectly, by design of the structure as
tolerable to accidents. Example of the latter is tensioner
failure where the tensioner shall provide sufficient integrity
to survive certain environmental scenarios without further
progressive collapse.
402 Design with respect to accidental load must ensure
that the overall failure probability complies with the target
values in Table 25. This probability can be expressed as the
sum of the probability of occurrence of the i’th damaging
event, P
Di
, times the structural failure probability
conditioned on this event, P
fDi
. The requirement is
accordingly expressed as:
∑
≤ ⋅
T , f Di Di  f
P P P
(5.33)
where P
f,T
is the target failure probability according to Table
25. The number of discretisation levels must be large
enough to ensure that the resulting probability is evaluated
with sufficient accuracy.
403 The inherent uncertainty of the frequency and
magnitude of the accidental loads, as well as the
approximate nature of the methods for determination of
accidental load effects, shall be recognised. Sound
engineering judgement and pragmatic evaluations are hence
required.
404 A simplified design check with respect to accidental
load may be performed as shown in Table 511 below
multiplied on appropriate load effect factors selected
according to Table 52 and resistance factors according to
Table 53 and Table 54. The adequacy of simplified design
check must be assessed based on the summation above in
order to verify that the overall failure probability complies
with the target values in Table 25.
Table 511 Simplified Design Check for Accidental
loads
Prob. of
occurrence
Safety Class
Low
Safety Class
Normal
Safety Class
High
>10
2
Accidental loads may be regarded similar to
environmental loads and may be evaluated
similar to ULS design check
10
2 
10
3
To be evaluated on a case by case basis
10
3 
10
4
γ
c
= 1.0 γ
c
= 1.0 γ
c
= 1.0
10
4
10
5
γ
c
= 0.9 γ
c
= 0.9
10
5
10
6
Accidental loads or events γ
c
= 0.8
<10
6
may be disregarded
Guidance note:
Standard industry practice assumes safety factors equal to 1.0
for accidental event with a probability of occurrence equal to
10
4
and survival of the riser is merely related to a conservative
definition of characteristic resistance. In this standard
accidental loads and events are introduced in a more general
context with a link between probability of occurrence and
actual failure consequence. For combined loading, the
simplified design check proposes a total safety factor in the
range 1.11.2. This range is consistent with standard industry
practice interpreted as corresponding to safety class Normal for
accidental loads with a probability of occurrence equal to 10
4
.
The ALS analysis may provide extreme loads for the design of
wellhead and rig equipment, and identify the need for
deliberately introducing weak links in the system. Such weak
links may be required to ensure that unacceptable escalation,
i.e. controlled riser failure above the subsea valve, does not
occur in case of accidents (in particular floater driveoff or
driftoff events or failure of draw works heave compensation
system). When maximum load is calculated in a potentially
weak link, a high characteristic value for the resistance of the
link should be used.
 end  of  Guidance  note –
DNVOSF201 Dynamic Risers, October 2010
Page 38 Section 5
DET NORSKE VERITAS
G. Serviceability Limit State
G 100 General
101 Serviceability limit states are most often associated
with determination of acceptable limitations to normal
operation. In many cases, the Owner will specify
requirements however, the designer must also carry out
evaluations with respect to riser serviceability and identify
relevant SLS criteria for the riser system.
Guidance note:
FMEA, HAZOP and design review meetings are useful
systematic procedures that can lead to identification of SLS
and for reviewing the consequences of setting operating
limitations and of exceeding those limitations.
 end  of  Guidance  note 
102 It is important that all operating limitations and/or
design assumptions are clearly highlighted and implemented
in the operating procedures.
103 Exceeding a SLS shall not lead to failure and an ALS
shall be defined in association with exceedance of SLS. In
addition, the frequency and consequences of events after
exceeding an SLS shall be evaluated. Such events will
typically be controlled by maintenance/inspection routines
and by implementation of early warning or failsafe type
systems in the design.
104 Serviceability limit states for the global riser
behaviour are associated with limitations with regard to
deflections, displacements and rotation of the global riser or
ovalisation of the riser pipe. Some examples are given in the
subsequent sections.
G 200 Ovalisation limit due to bending
201 Risers shall not be subjected to excessive ovalisation
and this shall be documented. In order to prevent premature
local buckling, the flattening due to bending together with
the outofroundness tolerance from fabrication of the pipe
shall be limited to 3.0 %:
03 . 0
D
D D
f
o
min max
0
≤
−
·
(5.34)
202 The requirement may be relaxed if:
÷ a corresponding reduction in moment resistance has
been included;
÷ geometrical restrictions are met, such as pigging and
tool access requirements; and
÷ additional cyclic stresses caused by the ovalisation have
been considered.
203 Ovalisation shall be checked for point loads at any
point along the riser system. Such point loads may arise at
freespan shoulders, artificial supports and support
settlements.
204 Special consideration shall be made of ovalisation
after loading causing plastic strains, e.g. reeling, unreeling
of pipes and riser interference/impact.
G 300 Riser stroke
301 For a top tensioned riser, a tensioner pulls upward on
the top part of the riser in order to limit bending and
maintain constant tension. The tensioner must continue to
pull as the riser and the floater move vertically relative to
each other. The travel of the tensioner is called its 'stroke'.
Riser stroke influences the design requirements for
tensioner, draw works, clearance between surface equipment
and drill floor, length of slick joint, etc.
302 Riser systems shall be designed to have sufficient
stroke such that damages to riser, components and
equipment are avoided.
303 The up and downstroke calculations must include
effects from environmental response, tension, pressure (end
cap effects), temperature, tide, storm surge, swell, makeup
(riser production tolerances), set down/pull down effects and
floater draught. For permanent risers, effects from
subsidence and settlements shall be evaluated.
304 Environmental response includes static and dynamic
stroke. The static stroke is due to current loading and set
down effect due to floater mean offset. The floater mean
offset includes effects from static wind and mean wave drift.
The wave loading introduces relative motions between the
floater and the riser, i.e. dynamic stroke.
305 The most unfavourable fluid density shall be
considered. Additionally, tension changes and length
variations needs to be taken into account.
G 400 Examples
401 Examples of SLS for drilling and workover riser
with subsea BOP are outlined in the following Table 512.
402 Examples when drilling with a surface BOP (e.g.
TLP, SPAR) the riser is part of the well control system and
may not be disconnected and hungoff. Some examples of
how this may influence SLS are summarised in Table 513.
403 Examples for export and import riser serviceability
limits should be set for riser installation and pigging, see
Table 514.
404 Examples for a production riser with a surface tree
the riser is part of the well control system and may not be
disconnected and hungoff. Some SLS examples are given
in Table 515.
405 Other serviceability limits may be determined to limit
the degradation of riser coatings and attachments or for
allowances due to wear and erosion.
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 39
DET NORSKE VERITAS
Table 512 Examples of SLS for drilling and workover with subsea BOP
Function SLS criteria Comment
Drilling with fluid
returns
Limit fatigue on drill string and
wear on wellhead/riser
Usually monitor flexjoint angle and
follow weather forecast and adjust
mooring to minimise joint angle
Guide tools or assemblies
into the well
Excessive angle may result in
getting stuck or not being able to
land the string properly
Due to tight tolerances
Overpull Avoid overloading the wellhead,
BOP and connectors
Overpull may be used to check that a
connector is made up properly or in an
attempt to release a stuck string
Flexjoint or ball joint
Operating limits for
specific operations
Disconnect and Hang off Approaching the resistance of
the wellhead/BOP and
connectors
For a normal hangoff scenario sufficient
time shall be allowed for pulling the
downhole string
Riser stroke Hang off Approaching the resistance of
the tensioner system
Weather is resulting in excessive platform
motion and offset.
Umbilical, choke, kill and
other attachments
BOP and well control Avoid damage Risk and consequences of damage may
govern criteria for interference
Table 513 Examples of SLS for drilling and workover with surface BOP
Function SLS criteria Comment
Drilling with fluid
returns
Limit fatigue on drill string
and wear on wellhead/riser
Usually monitor flexjoint angle or stressjoint
curvature. It is not normally feasible to adjust
moorings
Guide tools or
assemblies into the
well
Excessive angle may result in
getting stuck or not being able
to land the string properly
Due to tight tolerances
Flexjoint or Stressjoint
Operating limits for specific
operations
Overpull Avoid overloading the
wellhead, BOP and connectors
Overpull may be used to check that a
connector is made up properly or in an
attempt to release a stuck string
Riser installation Running and
retrieving the riser
A weather limitation would be
set to avoid riser interference
Usually run on guidewires in close proximity
to other risers
Table 514 Examples of SLS for export and import risers
Function SLS criteria Comment
Riser installation Running and retrieving the
riser
A weather limitation would be
set to avoid riser interference
Usually run on guidewires in close proximity to
other risers
Pigging Inspection or cleaning Pig launching and associated
temporary loading
Table 515 Examples of SLS for production risers with surface tree
Component Function Reason for SLS Comment
Riser installation Running and retrieving the riser A weather limitation would
be set to avoid riser
interference
Usually run on guidewires in close proximity to
other risers
Limit the frequency of bottom
out
The tensioner may be
designed for bottomout
Energy absorption criteria shall be specified Riser stroke
Limit the design requirements
for the jumper from the surface
tree to the topside piping
The tensioner may be
designed for bottomout
Energy absorption criteria shall be specified
H. Special Considerations
H 100 Interference
101 The riser system design shall include evaluation or
analysis of potential interference with other risers, mooring
lines, tendons, hull, the seabed, and with any other
obstruction. Interference shall be considered during all
phases of the riser design life.
102 A feasible design approach may be categorised into:
÷ No Collisions allowed
÷ Collisions allowed
DNVOSF201 Dynamic Risers, October 2010
Page 40 Section 5
DET NORSKE VERITAS
103 A first step is hence to determine whether collisions
are likely to occur or not. If collision occur, it must be
documented that the structural integrity is not endangered,
i.e. the pipe capacity is sufficient for both SLS and ULS
(incl. ALS & FLS) conditions. This requires an assessment
of collision frequency, location, force impulse or relative
riser velocity prior to the impact. Separate local
calculations/analyses will in general be required for
assessment of pipe stresses during impact.
Guidance note:
Owing to the complexity of interference analyses, due balance
between simplified and advanced analyses is recommended to
obtain efficient analyses:
− screening analyses using a simplified approach to identify
critical conditions or configurations;
− detailed analyses of identified critical conditions or
components using stateoftheart interference analyses.
Screening analyses may imply use of
− simplified environmental loads, e.g. current only, simple
profile without directionality;
− simplified Wake Induced Oscillation (WIO) and Vortex
Induced Vibration (VIV) models for current only or
undisturbed flow models for waves;
− simplified onset of collision criteria.
Detailed analyses for criticality assessment of collisions may
include:
− hydrodynamic interaction models;
− global collision models;
− dedicated CFD calculations;
− explicit collision load effect models;
− explicit limitstate design criteria..
 end  of  Guidance  note 
104 Model testing for verification of structural capacities,
hydrodynamic interaction models and global analysis
methodology is recommended.
H 200 Unstable Fracture and Gross Plastic
Deformation
201 Pipe members, including components and girth welds
shall have adequate safety due to unstable fracture for a
representative part or throughwall crack during the service
life of the riser.
202 Defect assessment of crack like defects should
normally be performed in accordance with BS 7910 Level
2A failure assessment diagram Partial safety factors for flaw
size, fracture toughness and yield strength should be as
given in BS 7910, Appendix K, Table K2 while load effect
factors shall be in accordance with B 200.
Guidance note:
The partial factors in Table K2 in BS 7910 annual target
probabilities of 10
3
, 7*10
5
, and 10
5
correspond to those for
safety class Low, Normal and High given in this standard.
 end  of  Guidance  note 
203 Defects assessment at fatigue sensitive locations shall
be additional to fatigue crack evaluations, see E.
Guidance note:
Fatigue failure in the SN curve approach, see E 200, is
normally based on through wall cracks. Where through wall
cracks are applied as failure criteria, it should be ensured that
through wall cracks should not cause unstable fracture.
Normally, brittle fracture in riser systems is avoided by
selection of material with sufficient ductility and Charpy V
notch impact energy and by performing NDT during
fabrication to ensure that only acceptable defects are present in
the riser system after fabrication.
Unstable fracture may occur under unfavourable combinations
of geometry, fracture toughness, crack like welding defects and
stress levels. The risk of unstable fracture increases in general
when the state of “plain strain” is approached at the crack tip.
This occurs in general with large material thickness, low
temperature, high loading rates, high strength material and
deep cracks subjected to bending. Fracture toughness data as
KIC, JIC or CTODC values are necessary to perform defect
assessment.
The failure assessment diagram is a twocriteria failure model
that considers unstable fracture, gross plastic deformations
(plastic limit load), and the interaction between these
mechanisms.
 end  of  Guidance  note 
204 For ALS evaluations, normally no partial load effect
factors are required for load effects, flaw size and toughness,
i.e. all partial safety factor shall be taken as unity. Normally,
the riser pipe is designed based on the principle that plastic
hinges may develop without giving rise to unstable fracture.
In such case, the nominal stress for unstable fracture shall
not be less than the design (yield) stress of the member.
H 300 Global Buckling
301 Global buckling (Euler buckling) implies buckling of
the pipe as a beamcolumn in compression. The procedure is
as for "ordinary" compression members in air using the
concept of effective tension.
302 A negative effective tension may cause a riser to
buckle as a beamcolumn in compression. Distinction shall
be made between loadcontrolled and displacement
controlled buckling. Excessive loadcontrolled buckling
involves total failure and is not accepted while a
displacement controlled buckling may be acceptable if the
postbuckling condition is acceptable.
303 The global buckling resistance for loadcontrolled
condition may be calculated according to recognised
stability criteria in structural design codes, e.g. ISO 13819
2.
304 Displacementcontrolled buckling may be acceptable,
provided it does not result in other failure modes. This
implies that global buckling may be acceptable provided
that:
÷ local buckling criteria are fulfilled in the global post
buckling configuration;
÷ displacement/curvatures/angles of the riser are
acceptable and
÷ cyclic effects are acceptable.
DNVOSF201 Dynamic Risers, October 2010
Section 5 Page 41
DET NORSKE VERITAS
305 Special care shall be given when a small decrease in
top tension of a toptensioned metallic riser could cause
excessive bending moment. In that case, the designer shall
establish a minimum effective tension that gives a margin
above the tension that is predicted to cause excessive
bending moments.
Guidance note:
It is essential that an appropriate tensionedbeam model is used
for the analysis of global buckling. The consequence of a too
small positive effective tension is excessive curvature and
bending moment near the location of minimum effective
tension.
Note that members above the tension joint for top tensioned
risers may be subjected to compressive forces for some riser
types
 end  of  Guidance  note 
DNVOSF201 Dynamic Risers, October 2010
Page 42 Section 6
DET NORSKE VERITAS
SECTION 6 CONNECTORS AND RISER COMPONENTS
Contents
A. General
A 100 Objective
B. Connector Designs
B 100 Functional Requirements
B 200 Design and Qualification Considerations
B 300 Seals
B 400 Local Analysis
C. Documentation
C 100 Documentation
C 200 Operating and maintenance manuals
A. General
A 100 Objective
101 This section gives requirements in relation to design,
analysis and qualification of metal connectors and
components used in riser design. The requirements apply
also to other riser components and at transitions to the pipe
wall thickness. Reference is made to ISO/CD 136287 for
further details on design, analysis and requirements.
Relevant sections include 5.8, 6.5 and 6.8.
102 The aim of the design is to ensure that the connectors
and riser components have adequate structural resistance,
leak tightness and fatigue resistance for all relevant load
cases. Resistance against accidental loads such as fire and
impact shall also be considered when applicable.
Guidance note:
Riser connectors basically provide a means of connecting and
disconnecting riser joints or equipment. The most commonly
used types of riser connector design comprises:
÷ threaded types;
÷ hub type;
÷ dog types, and
÷ bolted flanges designed for facetoface contact.
 end  of  Guidance  note 
B. Connector Designs
B 100 Functional Requirements
101 Riser connectors shall allow for multiple makeup and
breakout in a reliable manner. The connector may permit for
interchangeability between connector halves to allow riser
joints to be run in any sequence.
102 The external profile of all riser components shall not
restrict the passage of equipment like guideframes and
specialised tooling required for riser installation/retrieval,
inspection and maintenance, if applicable.
103 For permanent risers, provisions shall be made on the
riser joint/connector to allow for attachment of an anode
(bracelet). Electrical connection to the riser shall be made by
welding or other qualified method, and shall be made to a
low stress part of the connector. Other methods may be
used.
104 The riser pipe and components shall provide an
internally flush bore to ease running components into a well,
pigging and maintenance operation when applicable.
105 The design shall ensure that any trapped water/fluid
does not interfere with the installation or operation of the
connector.
106 For further functional requirements see ISO/CD
136287 Section 5.8
B 200 Design and Qualification Considerations
201 Connectors shall be designed to sustain the design
loads and deformations arising from makeup/breakout,
external loads applied to the pipe body, thermal gradients
and internal and external pressure loads without exceeding
the connector design resistance. All relevant limit states
must be considered.
202 The connectors should be designed to be at least as
strong as the pipe or weld with respect to strength, fatigue,
leakage and fire resistance.
203 As a minimum, the following loading parameters/
conditions shall be considered and documented by the
manufacturer when designing connectors and components:
÷ makeup loads;
÷ internal and external pressure including test pressure;
÷ bending moments and effective tensions;
÷ cyclic loading;
÷ thermal load effects (trapped fluid/water, dissimilar
metals) and thermal transients;
÷ breakout loads.
204 Issues which may require considerations in ULS and
ALS, include (not limited to):
÷ local buckling;
÷ unstable fracture and excessive yielding;
÷ leak tightness;
÷ thread disengagement.
÷ galling tendency between sliding elements
205 Deformations, deflections and finish damage, which
adversely affects the use, may require consideration in SLS.
206 The FLS capacity shall be verified to ensure that the
connector will not fail due to cyclic loading, see ISO/CD
136287 section 6.5.3.
DNVOSF201 Dynamic Risers, October 2010
Section 6 Page 43
DET NORSKE VERITAS
207 For connectors intended to be used in corrosive
environment, either the connector including components
shall be designed in such a way that an acceptable corrosion
control can be implemented at the joint, or the connector
shall be constructed of, or coated with, a corrosionresistant
material.
208 All riser connectors shall be qualified for the
application based on finite element analysis in combination
with performance qualification testing. Using analytical or
numerical calibration of the qualified connector,
representative connectors of the same type may be designed
by analytical methods (design equations) in combination
with finite element analysis whenever necessary.
209 Connector makeup shall be performed according to
a qualified procedure considering factors, such as friction,
lubrication, etc., in order to reduce the uncertainty in the
preload of the connector and ensure that the preload is
within the design limits.
Guidance note:
It is considered reasonable that the analysis or tests, which
should be carried out on connectors to be used on risers, should
demonstrate fit for purpose of their function. This does not
necessarily mean they have to be as strong and reliable as the
connecting pipe or weld. For static strength, plastic hinge may
preferably develop in the pipe before failure of the connector
occurs in order to increase the ductility in the riser system.
However, the minimum requirements are given above.
In cases where "weak links" are introduced to protect
components against accidental loads, i.e. driveoff, driftoff or
tensioner system failure, a connector with known breaking
resistance may be applied.
 end  of  Guidance  note 
B 300 Seals
301 Connectors shall provide a seal between the mating
segments that is compatible with any fluids that will pass
through the riser. The seal must maintain its integrity under
all external and internal loading conditions. Seal designs are
either integral or nonintegral. Integral seals are built into
the connector and are nonreplaceable. Nonintegral seals
use separate seal elements that can be removed and
replaced.
302 Seal design for connectors and riser components shall
include consideration of external pressure. Seal design shall
also consider operating conditions what may result in
frequent changes in the external loads and internal pressures,
which combined with external pressure results in frequent
pressure reversals on sealing mechanism. All operating
conditions (i.e. commissioning, testing, startup,
temperature, operation, blowdown, etc.) shall be
considered.
303 Seal rings wetted with internal fluid shall include the
same internal corrosion allowance as the connecting pipe
and be of compatible material. Alternatively, seals and
sealing surface shall be corrosion resistant in the actual
environment.
304 The seal and the connector including any bolts and
preload shall be considered together as a system to
determine the sealing performance. The effect of sealing
performance by the connector includes effects such as
torque of pin/box connectors and bolt resistance and
preload.
305 Metaltometal seals are preferred as the primary
seals on riser connectors. For permanent risers where metal
tometal seals are not utilised, redundant seals (primary plus
backup) should be provided.
306 Seals for riser connectors should be static, i.e.
sealing should take place between surfaces which have little
or no movement relative to each another.
307 Connectors exposed to cyclic loading shall utilise
nonloadcarrying seals in order to maintain high reliable
against leakage with time.
308 Seals of high reliability should be used to confine
flammable fluids, fluids under high pressure and corrosive
fluids. Seals must be selected with consideration to the
required service life, the service exposure in terms of
chemical aggressiveness and temperature as well as pressure
and relative displacements that need to be accommodated.
Guidance note:
All seals are sensitive to damage during handling, installation
and reassembly. A single seal therefore may have modest
reliability. To enhance the reliability, a double seal may be
provided. To achieve redundancy, the two seals should be of a
different design without common failure modes.
 end  of  Guidance  note 
B 400 Local Analysis
401 Local FE analysis should be performed for
connectors and structural components, including landing
blocks, taper joints, tension joints, flex/ball joints, slick
joints, complex riser joint cross sections (multiple pipes).
Loads and boundary conditions for use in local analysis
shall be obtained from the global analysis procedure.
Guidance on FE analysis of connectors and riser
components is given in ISO/CD 136287 section 6.8.
402 The most unfavourable combination of specified
tolerances shall be used in connection with FE analysis for
strength, leakage and fatigue (SCF's),
C. Documentation
C 100 Documentation
101 The documentation for the connector shall as a
minimum comply with the requirements of ISO/CD 136287
section 6.8.15.
C 200 Operating and maintenance manuals
201 The documentation for the connector shall as a
minimum comply with the requirements of ISO/CD 136287
section 6.8.16.
DNVOSF201 Dynamic Risers, October 2010
Page 44 Section 7
DET NORSKE VERITAS
SECTION 7 MATERIALS
Contents
A. General
A 100 Objective
A 200 Application
A 300 Material Selection
B. Additional Requirements
B 100 General
B 200 Long term properties
A. General
A 100 Objective
101 This section specifies the requirements for
materials, manufacture, fabrication and documentation of
riser pipe, components, equipment and structural items in
the riser system, with regard to the characteristic properties
of materials which shall be obtained after heat treatment,
expansion, final shaping and assembly. The requirements
are relevant both for pressure containing and for load
carrying parts.
A 200 Application
201 The requirements in this section are applicable for
metallic risers of the following materials:
÷ carbon Manganese steel;
÷ clad/lined steel; and
÷ corrosion resistant alloys (CRA) including ferritic
austenitic (duplex) steel, austenitic stainless steels,
martensitic stainless steels (“13% Cr”), other stainless
steels and nickel based alloys.
202 This standard applies to risers fabricated from
linepipe material meeting internationally recognised codes
for materials, manufacturing, coatings, fabrication and
NDT methods and procedures in general with the
exceptions given in Table 71 and in part B of this section.
203 The additional considerations in Table 71 could be
met by additional evaluations and/or specifications to the
applied codes or by applying the material, welding and
NDT requirement in DNVOSF101 and in part B of this
section.
204 The design utilisation in this standard depends on
the material quality and level of control, see Section 5. If a
higher utilisation is used, the principles and requirements
in DNVOSF101, Submarine Pipeline Systems,
supplementary requirement U, shall be applied for all
metallic materials included in this section.
A 300 Material Selection
301 The materials selected shall be suitable for the
intended use during the entire service life. The materials
for use in the riser system shall have the dimensions and
mechanical properties, such as strength, ductility,
toughness, corrosion and wear resistance, necessary to
comply with the assumptions made in the design.
302 Materials for riser systems shall be selected with
due consideration of the internal fluid, external
environment, loads, temperatures (maximum and
minimum), service life, temporary/permanent operations,
inspection/ replacement possibilities and possible failure
modes during the intended use. The selection of materials
shall ensure compatibility of all components in the riser
system. All elastomers and other nonmetallic materials
shall have documented compatibility with all fluids to
which they could be exposed including pressure and
temperature cycles.
303 All materials liable to corrode shall be protected
against corrosion. Special attention should be given to
local complex geometry, welds, areas that are difficult to
inspect/repair, consequences of corrosion damage, and
possibilities for electrolytic corrosion.
304 Requirements for corrosion allowance shall comply
with DNVOSF101. Special consideration shall be given
to the splash zone.
Guidance note:
The external corrosion allowance in the splash zone for CMn
steel is usually taken as 68 mm.
 end  of  Guidance  note 
305 All sliding surfaces shall be designed with
sufficient additional thickness against wear and tear.
Special attention should be given to the following where
applicable: clamped supports, sliding supports, slick joints,
dynamic seals, ball joints and telescopic joints.
306 The possibility for “sour” service conditions shall
be evaluated for all riser components, which can be
exposed to fluids with H
2
S during the lifetime of operation
of the riser.
307 The quality of the materials used shall be
tested/documented. Requirements to testing and control,
i.e. mechanical and corrosion testing, nondestructive
testing, dimensional and weight verification, shall be
determined during design, manufacture and fabrication,
based on the consequence with respect to failure and
experience.
B. Additional Requirements
B 100 General
101 Risers shall be made in seamless or longitudinally
welded pipes.
102 The riser components shall be forged/extruded
rather than cast whenever a favourable grain flow pattern,
DNVOSF201 Dynamic Risers, October 2010
Section 7 Page 45
DET NORSKE VERITAS
a maximum degree of homogeneity and the absence of
internal flaws are of importance.
103 Accumulated plastic strain, ε
p
, resulting from
installation and operation shall be treated in accordance
with the principles of DNVOSF101, sec 5 D1000.
104 However, the accumulated plastic strain limits of
0.3% and 2.0% applies only to the DNVOSF101 linepipe
specification. Equivalent criteria have to be developed for
other materials based on the fracture properties, welding
and NDT applied.
Guidance note:
Treatment of accumulated strains in accordance with DNV
OSF101:
− Requirements and guidelines to performance of ECA at
accumulated strains ε
p
≥ 0.3% are given in DNVOS
F101, sec 5, D1100 and sec 12.
− Supplementary requirement P in DNVOSF101 shall
apply for riser pipes with accumulated strain ε
p
≥ 2%.
 end  of  Guidance  note 
105 Reduction of area Z of cast and forged CMn fine
grain and low alloy steel shall be ≥ 35%. For heavy wall
components with SMYS above 420 MPa a higher ductility
level may be required. Requirement for ductility in the
through thickness direction shall be considered.
106 Limitations on SMYS on parts exposed to cathodic
protection shall be according with DNVRPB401.
107 Generic base polymer(s) ASTM D1418, physical
property requirements, storage and age control
requirements shall be defined for nonmetallic pressure
containing parts.
B 200 Long term properties
General
201 The long term material properties with regard to
fatigue and corrosion shall be documented. Special
consideration shall be given as to whether regular
inspection intervals or replacements can be applied (as for
temporary risers used for drilling, completion / workover)
or if inspection only is possible by means of remote
control equipment (as for permanent risers used for export,
import, production, injection).
Fatigue properties
202 Adequate fatigue life of base metal and weldments
shall be verified by fatigue analyses that are based on
fatigue testing (SN fatigue or fatigue crack growth
testing) or existing fatigue data.
Guidance note:
− when test results in terms of existing fatigue data are used
as basis for fatigue analyses, the tests shall have been
conducted on materials with expected fatigue properties
equal to the chosen material and in a representative
internal/external environment (including corrosion
protection if relevant);
− selection of SN curves shall match the weld detail and
quality;
− where sufficient and relevant test data are not available,
further testing shall be conducted.
 end  of  Guidance  note 
203 It is strongly recommended to specify tight
dimensional requirements at pipe ends for SCR’s in order
to reduce the stress concentration factors associated with
the girth welds. This can be obtained by the use of
supplementary requirement D in with DNVOSF101.
204 NDT of longitudinal welds shall include 100%
control for transverse imperfections, and be in accordance
with NDT Level I in DNVOSF101 or similar.
205 Weldments and other components with high fatigue
loads shall be identified, and extended NDT of these shall
be considered. Extended NDT can take place in the form
of spot checks performed by other qualified operator.
Corrosion
206 For temporary risers manufactured from CMn
steel, reduction in wall thickness due to internal corrosion
shall be evaluated. An evaluation shall take into
consideration the material properties, internal environment
as well as the maintenance and inspection procedures that
shall be applied. Effects of corrosion shall be accounted
for with a minimum of 1mm allowance unless it can be
documented that a corrosion allowance can be eliminated.
207 The external surface for temporary risers shall be
protected by a suitable coating system in addition to
routine coating repair and preservation of damaged
coating.
208 Special considerations shall be given to riser pipes
to be used for fluids containing hydrogen sulphide and
defined as “sour service” according to NACE Standard
MR0175. This can be obtained by the use of
supplementary requirement S in DNVOSF101.
Wear
209 Wear resistance shall be considered, particularly for
drilling risers or other wear exposed components.
Adequate wear resistance shall be verified by analyses and
/ or testing. Manufacturing process, machining and
fabrication shall also be considered.
DNVOSF201 Dynamic Risers, October 2010
Page 46 Section 7
DET NORSKE VERITAS
Table 71 Additional Considerations
Resistance Content DNV
OSF101
Sec 6. D
Recognised codes
Additional considerations
1
L
o
a
d
e
f
f
e
c
t
c
a
l
c
u
l
a
t
i
o
n
s
C
o
l
l
a
p
s
e
L
o
c
a
l
b
u
c
k
l
i
n
g
P
r
e
s
s
u
r
e
C
o
n
t
a
i
n
m
e
n
t
S
t
r
a
i
n
l
i
m
i
t
s
2
(
0
.
3
%
&
2
%
)
I
n
c
r
e
a
s
e
d
u
t
i
l
i
s
a
t
i
o
n
F
a
t
i
g
u
e
3
S
o
u
r
s
e
r
v
i
c
e
F
r
a
c
t
u
r
e
a
r
r
e
s
t
“Recognised” linepipe code X X X
Ovality
4
X
Mill test
5
X
I
n
c
l
u
d
e
d
Fracture properties,
Welding and NDT
6
X
Suppl. P Ductility
10
X (X)
Suppl. U Statistics
7
X
Suppl. D Dimensional requirements
3
X
Suppl. S Sour service X
Suppl F Fracture arrest X
NDT level I
8
NDT (X) X
High strength steel (yield stress > 555)
9
X X (X)
1 The “additional considerations” shall constitute input to an evaluation regarding the highlighted topic. Such an evaluation shall end up with resulting
specifications or guidance as required.
2 See B 103 of this section.
3 See B 202 and 203 of this section.
4 The moment capacity formulation is valid for (Dmax – Dmin)/D less than 3%, ref. DNVOSF101 Sec.5 D800
5 Mill test requirement in accordance with DNVOSF101 Sec.6 E1100 (hoop stress to be at least 96% of SMYS)
6 Criteria to be based on a fracture mechanic assessment
7 To document that SMYS is at least 2 standard deviations below the mean yield stress and that SMTS is at least 3 standard deviations below the mean
ultimate strength “
8 See B 204 and 205 of this section
9 DNVOSF101 is limited to yield stress less than 555. The effect of “other” stressstrain curves for high strength steel shall be evaluated if relevant
10 Testing, strain hardening
DNVOSF201 Dynamic Risers, October 2010
Section 8 Page 47
DET NORSKE VERITAS
SECTION 8 DOCUMENTATION AND VERIFICATION
Contents
A. General
A 100 Objective
B. Documentation
B 100 Design
B 200 Design basis
B 300 Design analysis
B 400 Manufacture and fabrication
B 500 Installation and Operation
B 600 DFI Résumé
B 700 Filing of documentation
C. Verification
C 100 General requirements
C 200 Verification during the design phase
C 300 Verification during the fabrication phase
A. General
A 100 Objective
101 This section gives requirements for documentation
and verification of riser systems during design, fabrication,
installation and operational phases.
B. Documentation
B 100 Design
101 Design documentation shall, as far as practicable, be
concise, nonvoluminous, and should include all relevant
information for all relevant phases of the lifetime of the riser
system. The design documentation shall be presented in
such a form that it is readily applicable for design review
and third party verification.
102 Documentation shall be available to the purchaser or
the purchaser's agents. Submittals and/or approval
procedures shall be agreed between the purchaser and the
supplier. Documents that are considered proprietary and
confidential shall be available for review.
B 200 Design basis
201 A design basis document shall be established in the
initial stages of the design process. The design basis
document normally include
÷ information supplied by the owner;
÷ procedures for riser system and component analysis
including analysis models and applied computer
programmes;
÷ all applicable load cases, limit states and safety classes
for all relevant temporary and operating design
conditions.
202 A summary of those items normally to be included in
the design basis document is included in Appendix F.
B 300 Design analysis
301 The design analysis documentation shall be self
contained and selfexplanatory setting forth in full detail,
including, but not limiting to the following items:
÷ a summary including design check key results and
illustrations in figures;
÷ explanation of notations and abbreviations;
÷ introduction including the objective of the document
and a brief description of the riser system;
÷ design basis if not included in a separate document, see
B 200;
÷ calculation input data, including material details,
assumptions for calculations and details of the computer
programs
÷ reference number of the standard/guideline/textbook
including the reference number for the formulae;
÷ full traceability of the calculations performed;
÷ wall thickness selection including minimum thickness,
tolerances, corrosion, wastage and other allowances
where applicable;
÷ graphs for the geometric model, including boundary
conditions;
÷ key results presented in a clear and concise manner (i.e.
utilisation ratios along the riser) and evaluation of the
results in the light of the limit states and assumptions
made in the analysis wrt. procedure/methods;
÷ relevant component and interface design loads,
including sources and assumptions;
÷ assumptions made with respect to treatment, inspection
and maintenance of the riser system in service.
302 Drawings shall be provided for the fabrication and
construction of the riser system, including but not limited to:
÷ floater layout drawings with risers;
÷ riser fabrication drawings; and
÷ drawings of the corrosion protection system.
B 400 Manufacture and fabrication
401 The following information shall be prepared prior to
start of or during manufacture of pipes, components,
equipment, structural and other fabricated items:
÷ material and manufacturing specifications;
÷ Manufacturing Procedure Specification (MPS);
÷ Quality Plans;
÷ welding procedure specifications/qualification records
if relevant;
÷ NDT procedures;
÷ manufacturing/fabrication procedures; and
÷ manufacturer's/fabricator's quality system manual.
DNVOSF201 Dynamic Risers, October 2010
Page 48 Section 8
DET NORSKE VERITAS
402 All relevant documentation shall be submitted to
owner, including but not limited to:
÷ fabrication procedures, incl. test requirements and
acceptance criteria, personnel qualification records, etc.;
÷ material certificates for e.g. pipes, piping components,
riser clamps, bolts, anodes, seal rings;
÷ fabrication procedure qualification reports including
welding procedure qualification records;
÷ test records (visual, NDT, tests on samples,
dimensional, heat treatment, pressure testing, FAT,etc.);
÷ necessary asbuilt drawings;
÷ complete statistics of chemical composition, mechanical
properties and dimensions for the quantity delivered;
÷ coating and corrosion protection data sheets; and
÷ all nonconformances identified during manufacture and
fabrication, and repairs performed
B 500 Installation and Operation
501 Installation and Operational requirements shall be
documented in a Riser Installation and Operation Manual(s).
The manual(s), which should be prepared jointly by the
designer and the owner, defines how to safely install,
operate and maintain the riser and its component systems.
502 The following information shall be prepared prior to
start of installation:
÷ Failure Mode Effect Analysis (FMECA) and HAZOP
studies;
÷ installation and testing specifications and drawings;
÷ installation Manuals;
÷ operational procedures for e.g. handling, running,
operation, emergency disconnect, hangoff;
÷ contingency procedures; and
÷ contractor Quality System manual.
503 The Riser Installation and Operational Manual
should contain as a minimum the following information:
÷ stepbystep procedure for handling, transportation,
running/retrieving, operating, preservation and storage
of the riser system;
÷ operating limits for each mode of operation;
÷ inspection and maintenance procedures for each
component;
÷ manufacturers drawings of the riser system components
outlining critical dimensions, weights and part numbers
of various components;
÷ recommended spare parts list.
B 600 DFI Résumé
601 A DFI Résumé shall be prepared for riser systems
including equipment and components. It shall contain all
relevant data and documentation used for:
÷ the design, fabrication and installation phase
÷ operation of the riser system and
÷ preparation of plans for periodic inspections of the riser
system
602 Documentation referred to in the DFI Résumé shall
be kept for the lifetime of the riser system and shall be
easily retrievable at any time.
603 The main objectives of the DFI résumé are to ensure
that only necessary information is kept available, to
facilitate the safe, effective and rational operation, and
maintenance and modifications of the riser system and input
to the preparation of plans for periodic inspection
604 The purpose of the DFI résumé is to:
÷ provide a reference key to the detail technical
documentation;
÷ provide a system description for the riser system ;
÷ provide a summary of all design, fabrication and
installation including, responsibility, requirements,
verification activities, deviations, detail design, follow
on engineering, design basis data, and critical design
areas with references to underlying detailed
documentation;
÷ provide recommendations, requirements and sufficient
information for the operation, inservice inspection,
integrity evaluation, maintenance activities and
modification or requalification throughout the entire
lifetime of the installation.
605 The DFI résumé is a historical document. Any
changes to the riser system after startup will be a part of
operation history and shall be reflected in a condition
résumé. The DFI résumé is therefore not supposed to be
updated based on events/changes made in the operation
phase.
B 700 Filing of documentation
701 Maintenance of complete files of all relevant
documentation during the life of the riser system is the
responsibility of the owner.
702 The engineering and asbuilt files shall, as a
minimum, comprise the documentation from design,
fabrication, installation and commissioning.
703 The engineering documentation shall be filed by the
Owner or by the engineering contractor for a minimum of
10 years. Design basis and key data for the riser system shall
by filed for the lifetime of the system. This includes
documentation from design to startup and also
documentation from possible major repair or reconstruction
of the riser system.
704 Files to be kept from the operational and
maintenance phases of the riser system shall, as a minimum,
include final inservice inspection reports from startup,
periodical and special inspections, condition monitoring
records, and final reports of maintenance and repair.
DNVOSF201 Dynamic Risers, October 2010
Section 8 Page 49
DET NORSKE VERITAS
C. Verification
C 100 General requirements
101 Compliance with provisions contained in relevant
national and international regulations or decisions made
pursuant to such regulations, shall be verified.
102 The extent of the verification and the verification
method in the various phases shall be assessed. The
consequences of any failure or defects that may occur during
construction of the riser system and its anticipated use shall
receive particular attention in this assessment.
103 The verification shall confirm whether the riser
system satisfies the requirements for the specific location
and method of installation and operation, taking into
consideration the design, including material selection and
corrosion protection, and the analysis methods used.
104 There shall be organisational independence between
those who carry out the design work, and those who verify
it.
105 Independent analyses shall to the extent practicable
possible be performed using different software as applied in
design.
106 Verification work and findings shall be documented.
C 200 Verification during the design phase
201 Verification of design should include checking of the
following items:
÷ that specifications are in compliance with the applicable
rules and regulations etc;
÷ appropriate personnel qualifications and organisation of
the design;
÷ calculations of loads and load effects;
÷ that accidental loads are in compliance with the results
from the risk analyses ;
÷ the usefulness of computer software, and that the
programmes are adequately tested and documented.
This is of particular importance when programmes are
used in dealing with new problems, constructions or in
case of new/modified software;
÷ independent calculations should be performed of the
riser system including riser components of significance
to the overall safety. The calculations should be
sufficiently accurate and extensive to demonstrate
clearly that the dimensions are adequate;
÷ that measuring requirements are complied with, e.g. for
environmental data;
÷ that deviations during fabrication and installation are
assessed and if necessary corrected;
÷ that drawings are in accordance with calculations and
specifications ;
÷ that corrosion, wear and erosion protection measures
are adequate;
÷ that the design of important structural details are
adequate.
C 300 Verification during the fabrication phase
301 Verification during fabrication should include the
following items to check that:
÷ the specifications are in accordance with public
regulations/provisions and safety requirements ;
÷ satisfactory work instructions and procedures are
prepared;
÷ personnel qualifications are in accordance with the
requirements;
÷ the methods and equipment of suppliers and at the
fabrication site are satisfactory with regard to control of
dimensions and quality of riser pipe, components and
materials;
÷ dimensions including assembly tolerances, NDT
detection limits, materials, surface protection and work
performance are in accordance with the basic
assumptions made during design;
÷ deviation procedures are adequate during fabrication
÷ the transportation and storage of materials and
fabricated assemblies are adequate.
DNVOSF201 Dynamic Risers, October 2010
Page 50 Section 9
DET NORSKE VERITAS
SECTION 9 OPERATION, MAINTENANCE AND REASSESSMENT
Contents
A. General
A 100 Objective
B. Inservice Inspection, Replacement and
Monitoring
B 100 General
B 200 Riser Inspection
B 300 Riser monitoring
B 400 Guidelines for inspection intervals
B 500 Condition Summary
C. Reassessment
C 100 General
C 200 Ultimate Strength
C 300 Extended Service life
C 400 Material Properties
C 500 Dimensions and Corrosion Allowance
C 600 Cracked Pipes and Components
A. General
A 100 Objective
101 The objective of this section is to provide
requirements for operation and inservice inspections. This
section also provides general guidance on structural integrity
assessment of risers to demonstrate fitness for purpose in
case deviations from design appear during operation.
B. Inservice Inspection, Replacement
and Monitoring
B 100 General
101 Risers shall be operated, maintained and inspected to
maintain an acceptable safety level throughout the service
life of the riser. They should also be inspected after
potentially damaging incidents and to confirm that any
repairs have been properly performed. Inspections relating
to areas such as the following may be necessary for risers
and riser components:
÷ overloaded/permanently deformed riser string
components;
÷ fatigue cracking (e.g. girth welds, connectors, anode
attachment welds);
÷ leaks (loosening of mechanical connectors, seal ring
damage);
÷ damage, e.g. dents, scratches, loosened or heavily
distorted coating;
÷ internal and external wear;
÷ internal and external corrosion, see DNVOSF101,
section 10.
÷ anticorrosion/abrasion coatings;
÷ cathodic protection;
÷ marine growth;
÷ soil conditions at seabed, e.g. touch down point;
102 Risers should be visually examined for factors such
as external damage, pipe distortion, excessive marine
growth, external corrosion, general pipe configuration and
sliding of buoyancy modules and/or ballast. Defects should
be documented with respect to type, size and location. The
influence of defects on structural or pressure integrity
should be assessed.
B 200 Riser Inspection
201 The inspection philosophy should be an integral part
of the design. Criticality of components and ease of
inspection should be considered early to ensure that
provisions are made for adequate inspection.
202 The designer should ensure that necessary inspection
methods or replacement procedures are available and are
scheduled and described in adequate detail as part of the
operating and maintenance documentation for the facility.
203 Parts that are damaged repaired or particularly
exposed and where failure will incur serious consequences
shall be subject to particular attention in the planning of in
service inspection and maintenance.
204 Risers to be inspected for fatigue cracks should be
inspected in accordance with the principles given in section
5.E 300.
Guidance note:
Equipment consumables such as seals, lubrication, periodically
disconnected components and paint should generally be
inspected or replaced on a scheduled basis. Moreover, the
equipment should be designed to facilitate these maintenance
operations. Manufacturer supplied data should include
recommended maintenance operations and intervals.
 end  of  Guidance  note 
205 The maximum interval between inspections should
be based on the component's predicted time to failure
divided by a safety factor. The safety factor should account
for uncertainties in timetofailure predictions, risks of
failure and ease of inspection. The designer should also
consider the time required for repairs or replacement when
determining maximum inspection intervals. Inspection
intervals should be developed for each mode of failure such
as fatigue, abrasion, wear, ageing and corrosion.
206 If the maximum inspection interval is longer than the
intended service life, inspection is not expected to be
necessary and need not be included in the operation and
maintenance documents. However, if during operation the
intended service life is extended beyond the original
maximum inspection interval of a component, then the
component should be inspected and refurbished if necessary
or replaced.
DNVOSF201 Dynamic Risers, October 2010
Section 9 Page 51
DET NORSKE VERITAS
B 300 Riser monitoring
301 The riser's internal and external operating condition
should be monitored to reveal whether design conditions
have been exceeded. This monitoring should include the
recording of riser response and tension (if relevant) as well
as the composition, pressure and temperature of the riser
contents. Wall thickness measurements by internal means,
e.g. pigs and by external means at selected reference points
should be considered.
Guidance note:
A riser monitoring system is not mandatory, but it is useful for
setting and maintaining precise tension, for monitoring riser
dynamics and for design verification. The riser monitoring
system can also be applied in connection with active floater
positioning for reduction of stresses, top/bottom flexjoint etc
of e.g. drilling risers. The system can also be used to record
and estimate riser fatigue damage.
 end  of  Guidance  note 
B 400 Guidelines for inspection intervals
401 The following factors should be taken into account
when determining inspection intervals:
÷ safety class;
÷ specific intervals based on criteria discussed elsewhere
in this section;
÷ present condition and service history, e.g., age, results
of previous inspections, changes in design operating or
loading conditions or prior damage and repairs;
÷ redundancy;
÷ riser type and location, e.g., deep water or new design
with few long term operating examples.
402 The intervals given in the Table 91 should not be
exceeded unless experience or engineering analysis justifies
longer intervals. In such cases, justification for changing
guideline inspection intervals, based on the factors listed in
this section, should be documented and retained by the
owner.
Table 91 Guideline for inspection intervals
Component Inspection type Interval
Above water
components
Visual 1 year
Below water
components
Visual 35 years
All components NDT As needed
Cathodic protection Visual or
Potential Survey
35 years
Areas of known or
suspected damage
As appropriate After exposure to
design event
Components
retrieved to surface
As recommended
by manufacturer
After disconnect
B 500 Condition Summary
501 The DFI Résumé is a historic document, which is
completed at the start of the operational phase.
502 Modifications, change of use, damage or
detrimentation of the riser system shall be included in an
updated condition summary. The condition summary shall
provide user groups with an overall picture of the actual
condition and functioning of the riser system during the
operational phase. This document should be updated
annually. Verification activities should be considered also
for the operational phase.
C. Reassessment
C 100 General
101 An existing riser shall undergo an integrity
assessment to demonstrate fitness for purpose if one or more
of the following conditions exists:
÷ extension of service life beyond the originally
calculated design life;
÷ damage or deterioration to a riser component;
÷ change of use that violates the original design or
previous integrity assessment basis;
÷ departure form the original basis of design, e.g. by
÷ change in environmental data or relocation;
÷ change in floater;
÷ change in internal fluid;
÷ change in top tension for TTR’s.
102 Assessment of existing risers should be based on the
most recent information of the riser. Load data should be
revised according to latest metocean data and the current
layout of the riser.
103 In case of change of use, repair, modifications,
damage or detrimentation of the riser system, measures shall
be implemented to maintain an acceptable safety level.
C 200 Ultimate Strength
201 The ultimate strength of damaged members should be
evaluated by using a rational, justifiable engineering
approach, e.g. DNV RPF101 may be applied.
202 The riser pipe or riser component must have
sufficient ductility to develop the failure mechanism in
question and large inelastic displacements or fractures due
to repeated yielding must not occur. Local buckling or other
nonlinear instabilities must be considered in the calculation.
C 300 Extended Service life
301 Extended service life may be based on results from
performed inspections throughout the prior service life.
Such an evaluation should be based on:
÷ reliability of inspection method(s) used;
÷ elapsed time from last inspection performed and/or
inspection/repair history.
Guidance note:
In some situations, even where cracks are not found it should
be considered to perform a light grinding at the hot spot areas
DNVOSF201 Dynamic Risers, October 2010
Page 52 Section 9
DET NORSKE VERITAS
of the riser systems to remove undercuts and increase the
reliability of the inspection.
Detected cracks may be ground and inspected again, to
document that they are removed. The remaining life of such a
repair should be assessed in each case.
 end  of  Guidance  note 
C 400 Material Properties
401 . Material properties may be revised from design
values to ‘as built’ values based on material certificates. The
yield and tensile strength may be taken as the minimum
guaranteed yield and tensile strength given in material
certificates.
402 Alternatively, material tests may be used to establish
the characteristic ‘as built’ yield strength. Due consideration
must be given to the inherent variability in the data. The
determination of characteristic values shall be in accordance
with the evaluation procedure given in ENV 1993 11,
Annex Y.
C 500 Dimensions and Corrosion Allowance
501 Strength assessment shall be based on 'as built'
dimensions, reduced for corrosion allowance.
502 For unprotected or cathodically protected steel, the
section thickness and the expected corrosion may be
updated based on the measured values. Section thickness for
use in the strength assessment may be calculated from the
measured section thickness combined with the expected
corrosion in the remaining lifetime, based on the observed
rate of corrosion.
Guidance note:
DNV RPF101 Corroded Pipelines gives guidance on
assessment of pipes corrosion/erosion defects including local
wall thinning due to fabrication tolerances and grind repairs.
 end  of  Guidance note 
C 600 Cracked Pipes and Components
601 Pipes and components containing cracks should be
repaired/replace as soon as possible.
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 53
DET NORSKE VERITAS
APPENDIX A GLOBAL ANALYSIS
Contents
A. General
A 100 Objective
B. Physical Properties of riser systems
B 100 General
B 200 Top tensioned risers
B 300 Compliant riser configurations
B 400 Nonlinearities
C. Global riser system analysis
C 100 Purpose of global analysis
C 200 General modelling/analysis considerations
C 300 Static finite element analysis
C 400 Finite element eigenvalue analysis
C 500 Dynamic finite element analysis
D. Combined floater/slender structure analysis
D 100 General
D 200 Coupled system analysis
D 300 Efficient analysis strategies considering coupling
effects
D 400 Coupled floater motion analysis
D 500 Decoupled floater motion analysis
E. Hydrodynamic loading on slender structures
E 100 General
E 200 Morison equation for circular crosssections
E 300 Morison equation for double symmetric cross
sections
E 400 Principles for selection of hydrodynamic
coefficients
F. Marine growth
G. Hydrostatic pressure loading
H. Internal fluid flow
H 100 General
H 200 Steady flow
H 300 Accelerated uniform flow
H 400 Slug flow
I. Forced Floater Motions
J. Hydrodynamic loading in moonpool
J 100 General
J 200 Moonpol kinematics
J 300 Hydrodynamic coefficients
K. Structural damping
K 100 Global Rayleigh damping model
K 200 Local Rayleigh damping models
L. References
L 100 Standards, Guidelines and Handbooks
L 200 Technical references
A. General
A 100 Objective
101 The objective of this Appendix is to give guidance on
global riser system analysis referred to present technical
level of tailormade FE computer codes for static and
dynamic analysis of slender structures.
A general generic presentation is used as the basic methods
of analysis can be applied to a wide range of riser systems.
Comments related to recommended procedures for specific
riser systems are addressed whenever appropriate. The focus
will be on the following essential issues:
÷ general overview of global system behaviour and
important nonlinearities;
÷ general overview of analysis techniques with emphasis
on treatment of nonlinearities;
÷ overview of important load models (e.g. effective
tension, hydrodynamic loading, internal fluid flow,
structural damping etc);
÷ guidance to global analysis to provide consistency with
the guideline requirements, and
÷ stateoftheart review of recent developments
regarding analysis techniques of particular interest for
deep water riser systems (e.g. coupled floater/slender
structure analysis).
102 The overall intention with the present document is to
support practical implementation of the LRFD and WSD
design formats and provide background information for
selection of adequate method of analysis.
103 The document should not be regarded as a self
contained document on analysis but rather as an introduction
to basic principles and more detailed guidance on selected
important topics. Functional description and extended use of
references is applied to describe well established
procedures, general techniques and accepted modelling
practice (guidelines, handbooks and technical papers)
104 In particular, API RP 2RD should be consulted for a
more detailed technical description as well as modelling
guidance of special components such as tensioner, flexjoint,
stressjoint, balljoint etc (mainly sections A.2, A.6.2,
A.6.4.2, A.6.3.5 and A.6.5). Reference is also made to API
RP 2RD, sections A.6.4.4.1 and A.6.5.1 for modelling of
multiple tubular crosssections in global analyses.
DNVOSF201 Dynamic Risers, October 2010
Page 54 Appendix A
DET NORSKE VERITAS
B. Physical Properties of riser
systems
B 100 General
101 The purpose of this section is to give a brief
overview of characteristic physical properties and governing
nonlinearities of riser systems. Such information is crucial
when selecting analysis strategy to describe the static and
dynamic behaviour when the system is exposed to
environmental loading due to current, waves and floater
motions.
102 The main functional requirements to marine risers is
to provide for transfer of fluids and gas between seafloor
and a floater, as well as allow for transportation of various
well operation tools. Risers are therefore commonly grouped
into the following categories, reflecting the area of
application:
÷ drilling risers;
÷ workover/completion risers;
÷ export/injection risers;
÷ production risers.
These categories differ with respect to typical dimensions,
crosssectional composition, types of operation, functional
requirements and design load conditions.
103 Risers will typically be operated from a floater. A
main concern in selection of the global riser configuration is
how floater motions should be absorbed by the riser system.
It is therefore convenient to distinguish between top
tensioned  and compliant risers to reflect the principle
applied for absorption of floater motions. Characteristic
properties of these main riser categories are discussed
separately below.
104 There is also a significant potential for hybrid riser
configurations, combining the properties of tensioned and
compliant risers in an efficient way, examples are given by
e.g. Espinasse et al (1989)
B 200 Top tensioned risers
201 Vertical risers supported by a top tension in
combination with boundary conditions that allows for
relative riser/floater motions in vertical direction are denoted
top tensioned risers. Furthermore, the riser is constrained to
follow the horizontal floater motion at one or several
locations. The intended (idealised) behaviour is that the
applied top tension should maintain a constant target value
regardless of the floater motion. Hence, the effective tension
distribution along the riser is mainly governed by functional
loading due to the applied top tension and the effective
weight of the riser. The relative riser/floater motion in
vertical direction is commonly denoted stroke. Applied top
tension and stroke capacity are the essential design
parameters governing the mechanical behaviour as well as
the application range. Top tensioned risers are applicable for
all functional purposes as mentioned above and will hence
represent an attractive alternative for floaters with rather
small heave motion (e.g. TLP, Spar platforms, deep draft
floaters (DDF) and semisubmersibles).
202 Top tensioned risers operated from TLP’s and semi
submersibles are equipped with a separate (hydraulic) heave
compensation system to account for the floater motions and
at the same time maintain a constant target value for the
applied top tension. Bending moments are mainly induced
by horizontal floater motions and transverse loading due to
current and wave action. A pronounced peak in the bending
moment distribution is normally seen close to the wave
zone.
203 An alternative solution is used for Spar platforms
where the top tension is obtained from buoyancy modules
attached along the upper part of the riser inside the
moonpool. Several supports are introduced along the riser
system to constrain the riser motion in transverse hull
direction. There are no constraints (except from friction
forces) in longitudinal direction allowing the hull to move
relative to the riser system. Bending moments in risers
operated from a shell Spar are mainly due to the resulting
horizontal hull motion as well as hydrodynamic loading
from the entrapped water in the moonpool. Pronounced
peaks in the bending moment distribution are normally
found at the support locations.
204 The static and dynamic behaviour of top tensioned
risers is largely governed by the applied top tension. The
effective weight of the riser system defines the lower
limitation for the applied top tension to avoid compressive
effective tension in the riser at static position. A significant
higher top tension must however be applied to account for
imperfect tensioner arrangements and allow for redundancy
in case of partial loss of top tension. Increased top tension
can also be applied to reduce the probability of collision in
riser arrays and limit the mean angles in bottom of the risers.
The applied top tension is commonly specified in terms of
increase relative to the effective weight of the riser system,
also denoted as overpull. The required overpull is system
dependent with a typical range of 30%60%.
205 Steel pipes have traditionally been applied for
conventional water depths. Titanium and composite pipes
are suggested for deep water applications in order to keep
the top tension requirement at an acceptable level. Steel
risers with buoyancy modules attached can alternatively be
applied for deep water.
206 The crosssectional composition depends on the
functional applications. Export, import and low pressure
drilling risers are normally single tubular risers. Multitube
crosssections may typically be found in high pressure
drilling and workover risers as well as production risers.
207 A taper joint, flexjoint or balljoint is applied to
reduce bending stresses at termination to the seafloor. A
flexjoint or balljoint may be applied to reduce bending
stresses at termination to the floater. A taper joint may also
be included at the keel of Spar and other deep draught
floaters.
B 300 Compliant riser configurations
301 Compliant riser configurations are designed to absorb
floater motions by change of geometry, without use of heave
compensation systems. Compliant risers are mainly applied
as production, export and injection risers. The required
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 55
DET NORSKE VERITAS
system flexibility is for conventional water depths normally
obtained by arranging nonbonded flexible pipes in one of
the ‘classical’ compliant riser configurations: Steep S, Lazy
S, Steep Wave, Lazy Wave, Pliant Wave or Free Hanging
(catenary).
302 In deepwater, it is however also possible to arrange
metallic pipes in compliant riser configurations. Free
Hanging Risers in steel have been installed in the Gulf of
Mexico (Phifer et al 1994), and Lazy Wave configurations
in steel and titanium have been proposed as deepwater riser
systems for TLP and Semisubmersibles. A Lazy Wave
configuration with increased horizontal extension termed
Long Wave is proposed for application of metallic risers for
a deep water Floating Production Ship and Offloading
facility (FPSO) for North sea conditions (Karunakaran et al
1996). In such applications it may also be desirable to apply
prebend pipe sections to reduce the dynamic curvature at
critical locations along the riser (i.e. hog and sag bends).
Single pipe crosssections are typically applied for
compliant riser configurations
303 Compliant riser systems will in general experience
significantly larger static and dynamic excursions when
compared to top tensioned risers. The floater motion
characteristics will in many situations be decisive for the
dynamic tension and moment variation along the riser (e.g.
TLPs, Semisubmersibles, Ships). Environmental load
effects will consequently also be of greater concern for
compliant configurations. Critical locations on compliant
risers are typically the wave zone, hog and sag bends, touch
down area at seafloor and at the terminations to rigid
structures.
304 Titanium may offer several benefits relative to steel
for some of these configurations. This is due to a low
modulus of elasticity (half that of steel) implying a higher
degree of flexibility. Furthermore, the yield stress is
typically higher than for steel and the specific weight is
much lower (about half the steel weight).
305 Termination to rigid structures is an essential design
issue for compliant riser configurations. Possible solutions
are carefully designed bend stiffener, ball joint or flex joint.
The primary design requirement is to limit bending
curvature and pipe stresses, the secondary design
requirement is to minimise forces on the supporting
structure.
B 400 Nonlinearities
401 A basic understanding of important nonlinearities of
riser systems is of vital importance for system modelling as
well as for selection of adequate global analysis approach.
Nonlinearities will also be decisive for the statistical
response characteristics for systems exposed to irregular
loading. An essential issue is how nonlinear properties of
the riser system and hydrodynamic loading mechanisms
transform the wave frequency Gaussian excitation (i.e.
waves and 1st order floater motion) into nonGaussian
system responses. Important nonlinearities that always
should be carefully considered can be summarised as:
1) Geometric stiffness (i.e. contribution from effective
tension to transverse stiffness). Tension variation is
hence a nonlinear effect for risers;
2) Hydrodynamic loading. Nonlinearities are introduced
by the quadratic drag term in the Morison equation
expressed by the relative structurefluid velocity and by
integration of hydrodynamic loading to actual surface
elevation;
3) Large rotations in 3D space;
4) Material nonlinearities, and
5) Contact problems in terms of seafloor contact (varying
location of touch down point and friction forces) and
hull/slender structure contact.
402 The relative importance of these nonlinearities is
strongly system and excitation dependent. Nonlinearities
due to item 1) and 2) will, at least to some extent, always be
present. Item 3) is relevant for twoaxial bending due to in
plane as well as out of plane excitation, while 4) and 5) are
more system specific nonlinear effects. It should be noted
that external hydrostatic pressure is not considered to be a
nonlinear effect as hydrostatic pressures normally will be
handled by the effective tension / effective weight concept
(Sparks 1984) in computer programs tailor made for slender
structure analysis (e.g. Engseth et. al 1988, O’Brien et. al.
1988).
C. Global riser system analysis
C 100 Purpose of global analysis
101 The purpose of global riser system analyses is to
describe the overall static and dynamic structural behaviour
by exposing the system to a stationary environmental
loading condition. A global cross sectional description in
terms of resulting force/displacement relations (axial force
versus axial elongation, bending moment versus curvature
and torsion moment versus twist angle) is applied in such
analyses. Relevant global response quantities can be
grouped into the following main categories:
÷ resulting crosssectional forces (effective tension,
bending moments, torsional moment);
÷ global riser deflections (curvature, elongation, angular
orientation);
÷ global riser position (coordinates, translations, distance
to other structures, position of touch down point on
seafloor, etc), and
÷ support forces at termination to rigid structures (
resulting force and moments).
These response quantities are given directly as output from
global riser analyses. It should be noted that the frequency
content in all response quantities can be WF or combined
WF and LF depending on the analysis strategy applied in the
global response analysis.
102 Subsequent detailed cross sectional analysis to
determine local stresses and strains can be performed using
resulting cross sectional forces from the global analysis as
boundary conditions and considering possible
DNVOSF201 Dynamic Risers, October 2010
Page 56 Appendix A
DET NORSKE VERITAS
external/internal pressure loading. Detailed component
analysis can also be performed by application of resulting
forces and deformations obtained from global analyses as
boundary conditions in local quasistatic analyses (e.g. flex
joint, taper joint and bend stiffener analysis).
103 Procedures for evaluation of LRFD capacity checks
for combined loading are addressed separately in Appendix
C
C 200 General modelling/analysis considerations
201 A Finite Element (FE) approach is normally
considered for global riser system analysis. The most
important features required for adequate modelling and
analysis of deep water systems in general can be
summarised as:
÷ 3D formulation allowing for unlimited translations and
rotations;
÷ conventional small strain slender beam and bar
elements including material and geometric stiffness
and allowing for nonlinear material properties;
÷ seafloor/riser and hull/riser contact formulations;
÷ adequate structural damping formulation;
÷ hydrodynamic loading according to the Morison
equation expressed by the relative water/structure
velocity and acceleration;
÷ regular and irregular loading due to waves and floater
motions;
÷ current modelling;
÷ special features allowing for efficient modelling of
components such as swivels, hinges, buoyancy
modules, clump weights, flexjoints etc;
÷ nonlinear static analysis, and
÷ nonlinear time domain dynamic analysis.
202 The computational efforts of nonlinear time domain
dynamic analysis considering a detailed global riser
response model can be substantial. This is in particular the
case for irregular analyses where rather long simulations in
general are required to estimate extreme responses with
sufficient statistical confidence. It is therefore desirable to
apply simplified analysis strategies as a supplement to the
general advanced approach in order to achieve more
efficient computer analyses.
203 The basic strategy to obtain efficient analyses is to
use a simplified response model and/or use of simplified
timesaving FE analysis methodologies such as 2D
formulations and linearized time and frequency domain
dynamic analyses, see C 500 for a description. Other
strategies such as use of special designed quasistatic
analysis for bend stiffeners (e.g. Sødahl and Larsen 1989)
and simulation of critical events identified by a simplified
approach (e.g. Passano 1995) can also be applied to gain
computational efficiency. The latter approach can be highly
beneficial when the wanted results are relatively few
extreme responses of a complex system exposed to irregular
excitation.
204 Model uncertainties will always be present in
numerical simulations of marine structures. Deviations from
the unknown ‘true’ response will depend on method of
analysis as well as the response model. Simplifications
introduced to achieve analyses that are more efficient will in
most cases lead to an increased model uncertainty. A
judgement regarding tradeoff between computational
efficiency and model uncertainty will therefore always be
involved when strategies for cost effective analysis are
decided. Issues that often must be considered in the decision
process are briefly described in the following:
÷ the acceptable accuracy is dependent on the purpose of
the analysis, i.e. the required accuracy of the analysis.
This ‘target accuracy’ is dependent on the purpose for
which the results will be used.
÷ the acceptable loss in accuracy by introduction of a
simplified approach must be seen in relation to other
uncertainties involved. (e.g. uncertainties related to
modelling of environmental loading, floater motions,
cross sectional properties, tension control etc.) The
standard approach often applied in practical analyses is
based on engineering judgement and experience
possibly supported by some simple parametric studies.
÷ numerous simplified analyses will normally produce
more information regarding overall static and dynamic
system behaviour when compared to a reduced number
of advanced analyses. With limited computer resources
available this should always be kept in mind when
deciding the analysis strategy. Different conclusions
may be drawn depending on scope of work (feasibility,
preliminary design, detailed design, final verification).
Simple methods allowing for a broad analysis scope is
attractive in the early analysis stage while more
specialised advanced analyses of identified critical
conditions are of more interest in final stages.
÷ the simulation length used in stochastic analyses is
crucial to obtain sufficient confidence of extreme
response estimates. It can be shown that the statistical
uncertainty roughly will be reduced proportional to the
square root of the simulation length. The benefit from
long simplified simulations versus reduced simulation
length using a more advanced tool must be considered
carefully when available computer resources are
limited.
C 300 Static finite element analysis
301 The purpose of the static analysis to establish the
static equilibrium configuration due to static loading
(weight, buoyancy, top tension, current) for given locations
of riser terminations to rigid structures (e.g. terminations to
seafloor and floater). Static analysis is always the first step
in global riser analysis and defines the starting point for
subsequent eigenvalue and dynamic analyses. Static riser
analyses are normally performed using a nonlinear FE
approach. Following standard FE terminology, it is
convenient to distinguish between the following basic
loading components:
1. volume forces (weight and buoyancy):
2. specified forces (e.g. applied top tension);
3. prescribed displacements (displacement of terminal
points from stressfree to specified positions), and
4. displacement dependant forces (current loading).
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 57
DET NORSKE VERITAS
302 Each of these load components is in a standard FE
approach applied in one or several load increments starting
from an initial stressfree configuration (i.e. user defined
reference configuration defining the state of no stress) to
obtain the static riser configuration. Static equilibrium is
ensured by equilibrium iteration at each load increment.
Guidance note:
The load components are for compliant riser configurations
often applied one by one in the order 134 (2 is irrelevant for
compliant riser configurations) or alternatively 1 followed by
simultaneous application of 3 and 4.
The situation is somewhat different for top tensioned risers
where the effective weight of the riser system is carried by the
applied top tension. This requires that 2 is applied before or
simultaneous with 1 to avoid instability problems if the riser is
modelled to be free to translate vertically at upper end.
These examples illustrates that application order of the defined
load components can be decisive for the efficiency and stability
of the static solution procedure. Load application order should
therefor be considered carefully to avoid instability problems.
 end  of  Guidance  note 
303 Some computer programs tailor made for slender
structure analysis offers an alternative strategy utilising the
catenary solution as starting point for FE analysis (e.g.
Engseth et al 1988, O’Brien and McNamara 1989). The
basic idea behind this approach is that the overall influence
from bending stiffness is moderate for compliant riser
configuration, which means that the FE and catenary
solutions are close. In this approach, application of volume
forces (1) and prescribed displacements (3) is replaced by
one equilibrium iteration starting from the catenary solution
established by e.g. the procedure described by Peyrot and
Goulois (1979). This combined use of catenary/FE analysis
gives a significant reduction in computation time for
compliant riser systems and, perhaps most important, helps
to avoid instability problems often encountered during
application of prescribed displacements for such systems.
C 400 Finite element eigenvalue analysis
401 Eigenvalue analysis is used to determine the
eigenfrequencies and eigenmodes of the riser system. The
analysis represent a fundamental check of the dynamic
properties of the riser system and should always be
considered as the first step in the dynamic system analysis.
Eigenvalue analysis is of particular interest in the early stage
design of deep water tensioned risers operated from Tension
leg and Spar platforms to avoid unwanted resonance
dynamics.
C 500 Dynamic finite element analysis
501 Risers, stationkeeping system and floater comprise
an integrated dynamic system responding to environmental
loading from wind, waves and current in a complex way.
Coupled staticand dynamic analysis of the complete system
is in general required to establish the floater motions of deep
water systems in terms of mean position and combined
wave and low frequency motions. In such analyses, it will
normally be sufficient to apply a rather crude slender
structure model, see D for an introduction to coupled
analysis
502 Global riser system analyses are, however, normally
performed considering forced excitation due to wave
frequency (WF) floater motions as well as direct wave and
current loading. The WF floater motions are computed in
the frequency domain. A representative mean floater
position accounting for average environmental forces as
well as low frequency (LF) motions is usually applied in
riser system analysis. It should however be noted that the
described approach only is applicable to riser systems that
do not respond dynamically to LF floater motions.
Combined WF and LF forced vessel motions should be
considered in riser analysis if riser dynamics is significantly
influenced by low frequency excitation.
503 Treatment of nonlinearities is the distinguishing
feature among available analysis techniques. Based on the
identified nonlinearities, it is obvious that the response
characteristics of riser systems in general are nonGaussian.
Time domain analysis is consequently the primary method
of analysis, especially concerning prediction of extreme
response.
504 Commonly used dynamic FE analysis techniques,
treatment of nonlinearities and main area of application are
summarised in the following.
÷ nonlinear time domain analysis based on step by step
numerical integration of the incremental dynamic
equilibrium equations. A NewtonRaphson type of
equilibrium iteration is applied at each time step. The
nonlinear approach will give an adequate description of
all nonlinear effects and will consequently give a good
representation of a possible nonGaussian response.
Nonlinear simulations will typically be needed for
systems undergoing large displacements, rotations or
tension variations or in situations where description of
variable touch down location or material nonlinearities
are important;
÷ linearized time domain analysis based on linearization
of the dynamic equilibrium equations with regard to
stiffness, damping and inertia forces at static
equilibrium position, (i.e. structural linearization). This
means that the system stiffness, damping and mass
matrices are kept constant throughout the analysis and
that system displacement vector can be found by a
simple back substitution at each time step. Nonlinear
hydrodynamic loading according to the Morison
equation is, however, still included. The linearized
approach is far more efficient than nonlinear analysis
and is hence an attractive alternative when
hydrodynamic loading is the major nonlinear
contributor. A typical application is analysis of
tensioned risers with moderate transverse excursions.
÷ frequency domain analysis based on linearization of
stiffness, damping, inertia, and external forces at
static equilibrium position (i.e. structural and load
linearization). Stochastic linearization for combined
wave/current loading is required for irregular analysis.
Frequency domain analysis will always give a Gaussian
DNVOSF201 Dynamic Risers, October 2010
Page 58 Appendix A
DET NORSKE VERITAS
response and is therefore in general not recommended
for extreme response prediction. The main application
area is fatigue calculations and longterm response
statistics to identify design conditions to be applied in
time domain analyses. The computation time is small
when compared to time domain analyses.
These techniques can give significantly different results
depending on the actual system characteristics, see e.g.
Rooney et al (1990) for application examples for a top
tensioned TLP production/injection riser.
D. Combined floater/slender
structure analysis
D 100 General
101 Floater, risers and mooring lines comprise an
integrated dynamic system responding to environmental
loading due to wind, waves and current in a complex way.
Current loading and damping due to the slender structures
(i.e. risers, tethers and mooring lines) may significantly
influence the low frequency floater motions of deep water
mooring systems. Consistent treatment of these coupling
effects is decisive for adequate prediction of floater motions
as well as slender structure response for some deep water
concepts such as slack and taut moored FPSO’s (Ormberg
et al 1998) and Spar platforms (Colby et al 1999). This
section is dedicated to discussion of analysis methodologies
for the complete system involving fully coupled system
analysis, coupled floater motion analysis as well as
traditional decoupled floater motion analysis. Floater
motion analysis and detailed slender structure analysis are
carried out separately in the two latter approaches to achieve
computations that are more efficient.
For a more comprehensive discussion of techniques for
analysis of the complete system, see for instance Ormberg et
al (1997), Ormberg and Larsen (1997), Ormberg et al (1998)
and Colby et al (2000).
102 The discussion of coupled analysis is mainly focused
on application to nonlinear systems (e.g nonlinar restoring
characteristics from slender structures) with significant LF
floater motions. The intended application area is typically
coupled analysis of FPSO’s and Spar platforms. The LF
motions of these systems may be significantly influenced by
slender structure coupling effects.
D 200 Coupled system analysis
201 All relevant coupling effects can be consistently
represented using a fully coupled analysis where the floater
force model is introduced in a detailed Finite Element (FE)
model of the complete slender structure system including all
mooring lines and risers. Nonlinear timedomain analysis
considering irregular wave frequency (WF) and low
frequency (LF) environmental loading is generally required
to give an adequate representation of the coupled
floater/slender structure dynamics on nonlinear systems. It
should be noted that this approach yields dynamic
equilibrium between the forces acting on the floater and
slender structure response at every time instant. It will
therefore be no need for assessment of the low frequency
damping from the slender structure, as this contribution is
automatically included in the slender structure response. The
output from such analyses will be floater motions as well as
a detailed slender structure response description (e.g. tension
in mooring lines as well as tension, moment, shear,
curvature and displacement in risers).
202 The computational efforts required for coupled
system analysis considering a detailed model of the slender
structure including all mooring lines and risers are
substantial and should therefore mainly be considered as a
tool for final verification purposes.
203 Coupled floater motion analysis in combination with
subsequent slender structure analysis as discussed in the
sections D 300 – D 400 is generally recommended to
achieve a more efficient and flexible analysis scheme. By
careful modelling, this approach is capable of predicting
floater motions and detailed slender structure response with
same precision as the complete coupled system analysis.
D 300 Efficient analysis strategies considering
coupling effects
301 Several strategies can be proposed to achieve
computational efficiency. All strategies have in common
that the floater motion and slender structure analyses are
carried out separately. The first step is always a floater
motion analysis. Computed floater motions are then applied
as loading in terms of forced boundary displacements in
subsequent slender structure analysis. Risers and critically
loaded mooring lines are analysed one by one in the slender
structure analyses contributing to computational flexibility
as well as a significant reduction in computation time.
302 The most direct way to proceed is to apply time
series of combined WF and LF floater motions computed by
the floater motion analysis as boundary conditions in the
slender structure analyses as shown in Figure A1 (branch
a). This approach will also capture possible LF slender
structure dynamics as well as influence from LF response
(possibly quasistatic) on the WF response. Such effects
may be of importance for some deepwater mooring line and
riser designs.
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 59
DET NORSKE VERITAS
Vessel Motion Analysis
Slender Structure
Analysis
LF & WF vessel motions
Simplified slender
structure model
Advanced vessel
mode
l
Advanced slender structure
model (Model and analyse
selected risers and mooring
lines one by one)
Vessel WFmotion
transferfunction
(
RAO
)
Establish
"
representative" offset
(mean and LF)
Select vessel motion
representation
a b
LF & WF
vessel motions
WF slender structure
response
Slender Structure
Analysis
LF and WF slender
structure response
Figure A1 Vessel motion analysis
303 Traditional assumptions can alternatively be applied
considering WF floater motions as dynamic excitation while
LF floater motions are accounted for by an additional offset,
see Figure A1 (branch b). The slender structure is
consequently assumed to respond quasistatically to LF
floater motions.
304 For guidance regarding calculation of representative
floater offset, reference is made to e.g. section 6.2.2 in API
RP 2SK.
305 Treatment of coupling effects in the floater motion
analysis is decisive for the validity of this approach. Floater
motions can be simulated using a coupled – or decoupled
approach as described in sections D 400 and D 500,
respectively.
D 400 Coupled floater motion analysis
401 The primary purpose of coupled floater motion
analysis is to give a good description of floater motions,
detailed slender structure response is secondary. It can
therefore be proposed to apply a rather crude slender
structure FE model (crude mesh, no bending stiffness etc.)
in the coupled analysis still catching the main coupling
effects (restoring, damping, and mass). The numerical
solution technique as well as floater force model is however
identical to the approach applied in coupled system analysis
as described above.
402 This approach gives a significant reduction in
computation time due to a reduced number of degrees of
freedom in the coupled analyses. Case studies of deep water
mooring systems verify that the floater motions can be
accurately predicted with acceptable computation time using
a simplified modelling of the slender structures (Ormberg et
al 1998).
D 500 Decoupled floater motion analysis
The purpose of this approach is to compute rigid body
floater motions considering static, low frequency and wave
frequency environmental loading. LF motions are computed
by stepwise numerical integration in time domain, while
WF motions normally are computed in frequency domain.
501 The floater force model is identical to the model
applied in coupled analyses as described above. The slender
structures are represented in a simplified way in terms of
static restoring force characteristics and a constant LF
viscous damping. The restoring force characteristics may
include effects from current loading on the slender
structures. If not, it is possible to account for current loading
on the slender structures in terms of an equivalent constant
force acting on the floater. The computation time is small
when compared to the coupled floater motion analysis.
502 Assessment of the LF damping for the actual
environmental condition is crucial for the LF floater motion
analysis. This information can be extracted from model
tests of the complete system or from coupled floater motion
analysis as discussed in 303 It has been experienced that
time histories covering roughly 2025 LF motion cycles is
required to obtain adequate damping estimates, see e.g.
Ormberg et al (1998) for further details. It should however
be emphasised that the damping contribution from the
slender structures for some systems is sensitive to the
environmental excitation. The damping estimate should
therefore preferably be based on the same environmental
condition as considered in the decoupled floater motion
analysis.
503 Efficient analysis and consistent treatment of
coupling effects can hence be achieved by splitting the
floater motion analysis in a rather ‘short’ coupled floater
motion simulation and ‘long’ decoupled floater motion
simulation. The LF damping estimated from the coupled
floater motion analysis is applied in the decoupled floater
motion analysis to obtain consistent treatment of coupling
effects. Furthermore, the computational efficient decoupled
approach allows for long simulations to achieve the required
statistical confidence.
E. Hydrodynamic loading on slender
structures
E 100 General
101 The hydrodynamic loading on slender structures can
be expressed by the Morison equation in terms of the
DNVOSF201 Dynamic Risers, October 2010
Page 60 Appendix A
DET NORSKE VERITAS
relative fluidstructure velocities and accelerations. The
fluid and structural velocity and acceleration vectors are in a
FE approach known in the global reference frame. The fluid
velocities and acceleration vectors can be found by
considering relevant contributions from wave kinematics
(regular or irregular, undisturbed or disturbed), current
(constant velocity or velocity and acceleration) or moonpool
kinematics. The latter calls for special attention and is
discussed separately below.
102 Hydrodynamic loading in normal and tangential pipe
directions is normally computed independently according to
the socalled crossflow (or independence) principle. This
approach requires that the fluid and structural acceleration
vectors are decomposed in the instantaneous normal and
tangential pipe directions.
103 Formulation of the normal load component is
dependent on the actual shape of the crosssection. The
Morison equation is therefore discussed separately for
circular and doublesymmetric crosssections, which cover
most situations of practical interest.
104 Hydrodynamic loading according to the Morison
formulation is a major source to nonlinearities in the
response characteristics of slender structures. Hence,
treatment of the Morison type of loading is an essential issue
when selecting method of analysis, see C. It is however also
most important to keep in mind that the eigenmodes and
eigenvalues of the system are influenced by the added mass
term in the Morison equation. Added mass contributions
should therefore be carefully evaluated as a part of
eigenvalue analysis in order to do an adequate assessment of
the governing eigenmodes and eigenperiods of the system
(e.g. added mass of aircans for Spar riser systems).
E 200 Morison equation for circular cross
sections
The Morison equation for circular crosssections can be
expressed as:
t
t
M
2
b
t
t
M
2
b
t t t t h
t
D t
n
n
M
2
b
n
n
M
2
b
n n n n h
n
D n
r ) 1 C (
4
D
v C
4
D
) r v (  r v  D C
2
1
f
r ) 1 C (
4
D
v C
4
D
) r v (  r v  D C
2
1
f
& & & & &
& & & & &
−
π
ρ −
π
ρ + − − ρ ·
−
π
ρ −
π
ρ + − − ρ ·
(A.1)
where:
f
n
Force per unit length in normal direction
f
t
Force per unit length in tangential direction
ρ Water density
D
b
Buoyancy diameter (i.e. equivalent diameter
for description of resulting buoyancy on a
general riser cross section)
D
h
Hydrodynamic diameter
n n
v , v & Fluid velocity and acceleration in normal
direction
n n
r , r & & & Structural velocity and acceleration in normal
direction.
n
M
n
D
C , C
Drag and inertia coefficients in normal
direction
t t
v , v & Fluid velocity and acceleration in tangential
direction
t t
r , r & & & Structural velocity and acceleration in
tangential direction.
t
M
t
D
C , C
Drag and inertia coefficients in tangential
direction
201 In addition, C
A
= C
M
– 1 is defined as the added mass
coefficient. The two first terms are in FE implementations
included in the external load vector while the latter term
(added mass term) is included in the mass matrix. Hence, it
is important to observe the eigenmodes and eigenperiods of
the system will be influenced by the added mass term.
202 The described formulation is applicable to bare
circular pipes as well as equivalent circular pipe models
(e.g. equally spaced circular cylindrical buoyancy elements
on a circular pipe as illustrated in Figure A2). The cross
flow principle is in the latter case somewhat questionable,
though, and special attention should be paid to the selection
of hydrodynamic coefficients, see the SINTEF handbook
(1992), section A.4.2.2.
Figure A2 Pipe with attached buoyancy elements
E 300 Morison equation for double symmetric
crosssections
301 The Morison equation can be extended to double
symmetric cross sections by decomposing the normal
velocities and accelerations in direction of the local cross
sectional symmetry axes. These local symmetry axes are
denoted y and z in the following and are shown in Figure
A3.
y
z
D
hz
D
hy
Figure A3 Definition of local cross sectional axes
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 61
DET NORSKE VERITAS
302 The force components in tangential and normal y
and zdirections can be expressed as:
t
t
M b t
t
M b t t t t ht
t
D t
nz
nz
M b nz
nz
M b nz nz nz nz hz
nz
D nz
ny
ny
M b ny
ny
M b ny ny ny ny hy
ny
D ny
r ) 1 C ( A v C A ) r v (  r v  D C
2
1
f
r ) 1 C ( A v C A ) r v (  r v  D C
2
1
f
r ) 1 C ( A v C A ) r v (  r v  D C
2
1
f
& & & & &
& & & & &
& & & & &
− ρ − ρ + − − ρ ·
− ρ − ρ + − − ρ ·
− ρ − ρ + − − ρ ·
(A.2)
where:
f
ny
, f
nz
Force per unit length in normal y – and z
directions
f
t
Force per unit length in tangential direction
ρ Water density
A
b
Crosssectional buoyancy area
D
hy
,D
hz
Hydrodynamic diameter (i.e. projected area)
in normal yand zdirections
D
ht
Hydrodynamic reference diameter in
tangential direction
ny ny
v , v & Fluid velocity and acceleration in normal y
direction
nz nz
v , v & Fluid velocity and acceleration in normal z
direction
t t
v , v & Fluid velocity and acceleration in tangential
direction
ny ny
r , r & & & Structural velocity and acceleration in
normal ydirection
nz nz
r , r & & & Structural velocity and acceleration in
normal zdirection
t t
r , r & & & Structural velocity and acceleration in
tangential direction
ny
M
ny
D
C , C
Drag and inertia coefficients in normal y
direction
nz
M
nz
D
C , C
Drag and inertia coefficients in normal z
direction
t
M
t
D
C , C
Drag and inertia coefficients in tangential
direction
303 In addition, the inviscid moment per unit length
about the longitudinal axis (i.e. tangential direction) can be
expressed as
) r v )( r v )( C C ( A m M
nz nz ny ny
nz
M
ny
M b 66 t
& &
&
− − − ρ + Ω − ·
(A.3)
where m
66
is the added moment (see Table 4.3 p.145 in
Newman 1977) and Ω is the angular velocity. The last term
is the Munkmoment.
304 The described model is applicable for modelling of
hydrodynamic loading on more complex crosssections such
as pipe bundles, combined pipeumbilical cross sections etc.
The force resultant will in general not follow the direction of
the velocity. Note that this model is not applicable for
rotational symmetric cross sections, as the correct Morison
formulation will not be retrieved.
The inviscid force and moment for a general crosssection is
discussed by Newman (1977) p 139
E 400 Principles for selection of hydrodynamic
coefficients
401 The hydrodynamic coefficients are dependent on a
number of parameters:
÷ body shape;
÷ Reynolds number Re = UD/ ν, where U is the free
stream velocity, D is the diameter and ν is the kinematic
viscosity;
÷ Keulegan Carpenter number KC = U
M
T/D, where U
M
is
the free stream velocity amplitude of the oscillatory
flow and T is the period of oscillation;
÷ roughness ratio k/D, where k is the characteristic
dimension of the roughness on the body;
÷ reduced velocity U/f
n
D, where f
n
is the natural
frequency of the riser;
÷ relative current number U
c
/U
M
, where U
c
is the current
velocity and U
M
is the velocity of the oscillatory
motion.
402 An extensive discussion on the dependence of the
hydrodynamic coefficients on several of these parameters
can be found in Sarpkaya and Isaacson (1981) and Sumer
and Fredsøe (1997). Their presentation is concentrated on
circular cylinders with different Re, KC and roughness ratio.
Nevertheless, it can be difficult to decide the coefficients
based on the abovementioned criteria, for instance due to
varying flow conditions or lack of information. For
practical purposes, it is often sufficient to use a simplified
consideration to select the coefficients in Morison’s
equation. As a first approximation, use values for steady
flow. For circular bare pipes natural choices are C
M
= 2 and
C
D
= 0.7 – 1.0. A more detailed discussion is found in the
SINTEF handbook (1992)
For riser with buoyancy elements, reference is made to the
Sintef handbook (1992), section A4.2.2. The tangential
forces can be equally important to the normal forces.
403 Focus should always be given to selecting
hydrodynamic coefficients slightly on the conservative side.
In such considerations, it is important to distinguish between
areas where the drag term act as excitation (e.g. wave zone)
and areas where the drag term act as damping (e.g. parts of
the riser system insignificantly influenced by wave loading).
Slightly high/low values should be selected for areas where
the drag term act as excitation/damping, respectively. A
sharp distinction between areas with excitation or damping
is not always possible. Therefore, a sensitivity study should
always be performed to support rational conservative
assumptions when a high level of confidence is required.
DNVOSF201 Dynamic Risers, October 2010
Page 62 Appendix A
DET NORSKE VERITAS
F. Marine growth
101 Marine growth on slender structures will influence
the loading in terms of increased mass, diameter and
hydrodynamic loading.
102 Site dependent data for marine growth are normally
specified in terms of density, roughness and depth variation
of thickness. The marine growth characteristics are basically
governed by the biological and oceanographic conditions at
the actual site. The relative density of marine growth is in
the range of 11.4 depending on the type of organisms.
103 The thickness of marine growth to be included in
design analyses will in addition be dependent on operational
measures (e.g. regular cleaning, use of anti fouling coating)
as well as structural behaviour (e.g. less marine growth is
normally considered for slender structures with significant
dynamic displacements).
104 Field measurements at the actual location combined
with field experience regarding the extent of marine growth
on similar structures will therefore be the reference for
specification marine growth to be considered in design
analyses
105 In FE analyses, it is recommended to increase mass,
buoyancy diameter and drag diameter according to the
specified depth variation of marine growth. In addition, the
hydrodynamic coefficients should be assessed with basis in
the roughness specified for the marine growth.
G. Hydrostatic pressure loading
101 Loading due to external and internal pressure acting
on a pipe section is normally treated in terms of the effective
weight/tension concept (Sparks 1984):
g A g A g m W
P A P A T T
o o i i p e
o o i i w e
ρ − ρ + ·
+ − · (A.4)
Where :
T
e
Effective Tension
T
w
True wall tension (i.e. axial stress resultant found
by integrating axial stress over the crosssection)
A
o
,A
i
External and internal cross sectional areas
W
e
Effective weight (i.e. submerged weight of pipe
including content)
M
p
Mass of pipe
ρ
o
,ρ
i
External and internal fluid densities
G Acceleration of gravity
P
0
,P
i
External and Internal pressure
102 Hydrostatic pressure is acting normal to
instantaneous orientation of the pipe and can hence be
classified as a nonconservative loading (follower load). The
main advantage of the effective weight/tension formulation
is that loading due to hydrostatic pressure is represented by
vertical conservative forces (i.e. the nonconservative
pressure force model is replaced by an equivalent
conservative volume force model). This is of vital
importance for efficiency and stability in computer
implementations as pressure integration over the deformed
pipe geometry is avoided.
103 The physical significance of the effective tension can
be summarised as:
÷ the geometric stiffness is governed by the effective
tension. This means that the effective tension is the
overall governing stiffness parameter for the vast
majority of slender structures;
÷ global buckling and stability is governed by the
effective tension;
÷ the effective tension formulation is applicable to any
general shaped volume stiff cross section (i.e. cross
sectional volume is not influenced by the hydrostatic
pressure).
104 The effective tension formulation is also directly
applicable to multipipe crosssections (i.e. pipes inside
pipes) by summation of effective tension and effective
weight contributions from all pipes, see e.g. Skomedal
(1990) for further details.
H. Internal fluid flow
H 100 General
101 Loading due to internal fluid flow is normally
included in terms of the hydrostatic pressure and mass
contribution in global riser analyses.
102 The hydrostatic pressure is treated by the effective
weight/effective tension concept as described in section G
for static and dynamic analyses. In dynamic analyses, the
mass of the internal fluid is included in the effective mass
defined as the crosssectional mass including content m
e
=
m
p
+ ρ
i
A
i
. Hence, the effective mass is consistent with the
effective weight (i.e. the effective weight corresponds to the
submerged weight of the effective mass). This model is
formally correct for static and dynamic analyses of risers
conveying a hydrostatic internal fluid (i.e. no fluid flow
through the pipe is considered).
103 Additional loading (centrifugal and coriolis forces) is
however introduced due to the internal fluid flow through
the pipe. The significance of the additional loading is briefly
discussed in the following for three fluid flow categories:
steady flow, accelerated uniform flow and slug flow.
H 200 Steady flow
201 Steady flow corresponds to e.g. normal production
with a homogenous fluid flow with constant velocity
through the pipe. The effect of a steady flow on a static riser
configuration can be summarised as (Fylling et al 1986,
Patel and Seyed 1989):
÷ the effective tension is not affected by the steady flow;
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 63
DET NORSKE VERITAS
÷ the steady flow causes an increase in the true wall
tension and a corresponding axial elongation, and
÷ the static configuration is only affected by the axial
elongation caused by the steady flow. This effect is
negligible for compliant riser configurations, but may
have some significance for calculation of the vertical
upper end position of top tensioned risers.
202 The effective tension including the effect of steady
internal fluid flow can be expressed by:
2
i i i o o i i w e
v A P A P A T T ρ − + − ·
(A.5)
Where ν
i
is the steady internal fluid velocity. Steady flow is
hence seen to modify the wall tension in a similar manner to
internal pressure, i.e., by increasing the wall tension and
leaving the effective tension unchanged. It should be
observed that the increase in wall tension
2
i i i W
v A T ρ · ∆ is
independent of curvature and flow direction.
203 The steady fluid flow will introduce additional
loading on a curved pipe exposed to dynamic excitation by
e.g. forced support displacements. Model tests of a
submerged Ushaped flexible hose considering a range of
fluid velocity and support motion combinations have been
conducted to investigate this effect (Sintef 1992). It was
concluded that dynamic response was insignificantly
influenced by the steady fluid flow, indicating that this
loading type is of less importance for riser systems.
H 300 Accelerated uniform flow
301 Sudden stop or start of the internal fluid flow will
cause an accelerated uniform flow. This flow pattern will
result in a transient inplane excitation of the riser system.
The global response to this excitation can be predicted by
nonlinear time domain analysis using the load model as
described for slug flow conditions in H 400.
H 400 Slug flow
401 Slug flow is characterised by an alternating flow of
liquid slugs and gas pockets. “Steady state” slug flow can be
classified as either hydrodynamic slugging or terrain
induced slugging governed by the site specific elevation
profile of the flowline. Severe slug flow is typically related
to the latter condition. Transient slug conditions can in
addition occur during startup, operational changes of flow
rate and pigging operations. Slug volumes seen during
pigging are usually the largest slug volumes seen during
normal operation. For a more detailed discussion of slug
flow characteristics as well as review of numerical
simulation techniques to predict the slug flow, see e.g.
Burke and Kashou (1995).
402 Slug flow can hence be considered as a time
dependent variation of internal flow velocity and density at
any location along the riser. Implementation of an adequate
load model due to such flow conditions is needed for
prediction of the global riser response due to slug flow by
nonlinear time domain analyses. For such applications, it is
convenient to parameterise the slug flow in terms of
velocity, length and density of each individual slug as well
as the slug frequency defining the time interval between
successive slugs. All these quantities may in general be
considered as stochastic variables. Furthermore, the
velocity, density and length will in general change
(according to deterministic considerations) as the slug
passes through the riser. All relevant data characterising the
slug flow is input to the global response analysis and is
typically established by numerical multiflow simulations
and/or laboratory measurements supported by field
experiences.
403 The load model accounting for the slug flow must in
general include contributions due to mass, weight,
centrifugal force and the coriolis force as the slugs travel
through the pipe (the latter term is often considered less
important and consequently omitted). Furthermore, it is
assumed that the slug flow and riser motions can be treated
independently (i.e. the slug flow through the pipe is not
influenced by the motion of the riser). For further details and
examples of practical application see e.g. Fylling et al
(1986), Patel and Seyed (1989) and Sanderson et al (1999)
404 Dynamic effects due to slug flow is normally most
pronounced in areas along the riser with high curvature due
to the centrifugal load component (e.g. hog/sag bend, close
to supports, touch down area). Passage of large slugs
through a gas riser may also introduce significant quasistatic
change in riser configuration due to change in effective
weight by the slug mass. In addition, slug frequencies close
to governing eigenfrequencies of the riser system should be
considered carefully. Response due to slug flow is expected
to be most pronounced for deep water compliant gas riser
configurations (e.g. metallic lazy wave configurations).
Possible slug flow excitation should therefore always be
carefully evaluated for deepwater compliant riser
configurations.
I. Forced Floater Motions
101 Forced floater motions are defined as displacements
imposed on the riser due to motions of the surface floater.
These forced displacements may be introduced at several
elevations on the riser depending on type of floater (e.g
Semi, TLP, Spar, Ship).
102 Riser support motions may be obtained in several
ways, e.g using time series from model testing, or
coupled/decoupled analyses, or frequency domain results.
The most appropriate method (time/frequency domain) has
to be selected, depending on riser system and floater being
considered.
103 The frequency domain solution implies selecting an
appropriate quasistatic offset (mean + slowly varying) and
heel/tilt (e.g. Spar) and superimposing the wave frequency
(WF) motions. WF floater motions are usually given as
RAO's (response amplitude and phase angle). Special care
has to be exercised when transferring the RAO's from e.g a
motion analysis program to a purpose made riser analysis
program.
104 If LF displacements/rotations are of importance for
determination of resulting riser responses the optimum
DNVOSF201 Dynamic Risers, October 2010
Page 64 Appendix A
DET NORSKE VERITAS
solution is to perform time domain analyses (WF and LF
included). Simplified frequency domain analyses can
however be performed to assess the LF riser responses/
importance.
105 High frequency (HF) motions (e.g. ringing and
springing for TLP's) will usually not be of concern for the
riser system. This has to be evaluated on a case by case
situation. Typically, high ringing accelerations of a TLP
may have to be evaluated in case of heavy riser components
(trees) at deck level.
106 The presence of the floater gives rise to changes in
the fluid kinematics (velocity, acceleration and direction).
This disturbance is most easily determined by use of
radiation/ diffraction analysis programs. The outputs from
such programs are RAO’s for disturbed kinematics (velocity
components) consistent with the floater motion RAO’s. For
large volume floaters and risers located close to e.g
columns/pontoons, this disturbance must be determined and
accounted for in design.
J. Hydrodynamic loading in
moonpool
J 100 General
101 Treatment of hydrodynamic loading on slender
structures in the moonpool may be of vital importance for
some floater concepts (e.g. Deep draft floaters and shell
Spar platforms)
J 200 Moonpol kinematics
201 Kinematics of the entrapped water in the moonpool
area can in principle be treated in the same way as the
disturbed wave kinematics described in section I (i.e. in
terms of transfer functions for moonpool kinematics
consistent with the hull motion transfer functions). This
approach requires that the entrapped water is included in the
hydrodynamic model used to compute the floater motion
characteristics. Such calculations will however require a
very careful modelling to achieve a realistic picture in case
of complicated moonpool geometry and/or multiple risers in
the moonpool. Special attention should be focused on
possible resonant modes of the entrapped water.
202 Due to the complexity of the problem, it is often
desirable to apply a simplified model in practical
calculations.
A simplified model for the moonpool kinematics can be
obtained by assuming that the entrapped water follows the
hull motion rigidly. Fluid velocity and acceleration
components can then be found at any location in the
moonpool by straightforward transformations of the hull
motions (i.e. translations and rotations at a specified location
on the floater, centre of gravity is typically selected as
motion reference point). This formulation is applicable for
FD as well as TD analysis. The latter approach allows for
consistent treatment of moonpool kinematics due to
simultaneous WF and LF floater motions.
It is recommended that models of moonpool kinematics are
verified against model tests.
J 300 Hydrodynamic coefficients
301 Uncertainty is connected to the hydrodynamic
coefficients applicable inside a moonpool. Assuming that
the entrapped water follows the hull motion rigidly, the
hydrodynamic loading in the normal direction can be
expressed as:
) r v )( 1 C (
4
D
v
4
D
) r v (  r v  D C
2
1
f
n H
n
M
2
b
H
2
b
n H n H h
n
D n
& & &
& &
− −
π
ρ +
π
ρ +
− − ρ ·
(A.6)
where
H H
v , v & are the hull velocity and acceleration
components normal to the riser.
302 The riser motions relative to the moonpool are to a
large extent governed by how the riser is supported inside
the moonpool. The riser motion in transverse moonpool
direction will typically be constrained at several locations
along the riser (e.g. shell Spar). The excitation force is hence
not very sensitive to the C
D
and C
M
values due to the small
relative motion between the fluid and the riser (see equation
above). The “Froude Krylov”term, i.e. the inertia
contribution due to fluid acceleration, is governing for the
excitation forces. Accurate assessment of the drag and
inertia coefficients may therefore not be necessary for
adequate modelling of hydrodynamic loading for this
support condition. The C
M
value will however influence the
eigenvalues of the riser system.
303 Sensitivity studies considering riser dynamics and
eigenvalues should always be performed to support
decisions regarding choice of hydrodynamic coefficients to
yield conservative response estimates. It should also be
observed that the drag term will act as damping. A slightly
low value is recommended as a conservative estimate.
K. Structural damping
K 100 Global Rayleigh damping model
101 The Rayleigh damping model commonly adopted for
description of structural damping:
K M C
2 1
α + α · (A.7)
This means that the global damping matrix (C) is found as a
linear combination of the global mass (M) and global
stiffness (K) matrices. α
1
and α
2
are denoted the mass and
stiffness proportional damping coefficients, respectively.
The motivation for adopting this model is mainly due to
computational conveniences. One of the advantages is that
the damping matrix is orthogonal with respect to the
eigenvectors of the system, which allows for a simple
expression of the modal damping ratio ξ (i.e. damping
relative to critical) at angular frequency ω:
DNVOSF201 Dynamic Risers, October 2010
Appendix A Page 65
DET NORSKE VERITAS
,
`
.

ω α +
ω
α
· ζ
2
1
2
1
(A.8)
102 In practical calculations, α
1
and α
2
can be selected to
give required modal damping ratios ξ
i
and ξ
j
at specified
angular frequencies ω
i
and ω
j
, respectively. Explicit
expressions are given in standard textbooks on structural
dynamics, e.g. Clough and Penzien (1975).
103 The proportional damping model applies to all global
degrees of freedom and will hence give the same damping
level in all parts of the structure. It should also be observed
that the mass proportional damping would give damping due
to rigid body motions. The mass proportional damping is
therefore normally neglected for compliant structures
undergoing large rigid body motions (Sintef 1992). For a
stiffness proportional damping model, the structural
damping at angular frequency ω is given by:
2
2
ω
α · ξ
(A.9)
104 The stiffness proportional damping is seen to
increase linearly with the frequency. In practical
applications, α
2
is selected to give a realistic damping level
at the dominating dynamic frequency. For combined WF
and LF response, α
2
is selected to give a realistic damping at
the dominating wave frequency (typically peak period in the
wave spectrum). The proposed model will hence give
damping in the wave frequency range (i.e. dynamic
frequency range) and for all practical purposes no damping
in the low frequency range (i.e. quasistatic frequency
range). This is considered to be a realistic model, as
damping is not considered to be of significance for
prediction of the quasistatic low frequency response.
105 The Rayleigh damping model is applicable to all
relevant global analysis strategies (nonlinear and linearized
time domain and frequency domain analyses). However,
application in nonlinear time domain analyses call for
special attention. The global mass and stiffness matrices will
in a nonlinear time domain analysis be a function of the
instantaneous nodal positions and rotations. The following
implementations of the Rayleigh damping model can hence
be applied:
÷ updated formulation which means that the Rayleigh
damping matrix is based on instantaneous mass and
stiffness matrices;
÷ constant formulation which means that the Rayleigh
damping matrix is based on mass and stiffness matrices
at static equilibrium position (i.e. the damping matrix is
kept constant throughout the analysis).
106 The latter approach is in general recommended due
to better performance regarding numerical stability and
computational efficiency. This is in particular the case for
low effective tension problems using a stiffness proportional
damping model. For such problems, damping is typically
needed to stabilise the solution when geometric stiffness is
lost due to low dynamic effective tension. The updated
stiffness proportional formulation would hence give a low
damping when most needed to stabilise the solution.
K 200 Local Rayleigh damping models
201 The global Rayleigh damping model, as discussed in
the previous section, applies to all global degrees of freedom
and is hence limited to specification of an overall realistic
structural damping level. However, for some applications it
is desirable to have more flexible damping models. The
following extensions of the Rayleigh model have been
proposed:
÷ local spatial damping to facilitate specification of
different damping levels in different parts of the
structure. In practical implementation, this can be
achieved by specification of damping coefficients for
element subsets before assembly of global system
matrices;
÷ specification of different damping levels in different
local deformation modes (i.e. axial, bending and
torsional deformation modes). In practical
implementation, this can be achieved by specification of
local damping coefficients related to axial, bending, and
torsional deformations in local element system before
transformation and assembly of global system matrices.
For further details, see Bech et al (1992).
202 The local models can be applied in combination and
used in updated as well as constant damping formulations as
discussed in the previous section. The main drawback with
the local damping models is that the orthogonality with
respect to the system eigenvectors is lost. The simple closed
form expression for modal damping presented above is
therefore formally not applicable for specification of the
damping level.
L. References
L 100 Standards, Guidelines and Handbooks
API RP 2RD “Design of Risers for Floating Production
Systems (FPSs) and Tensionleg Platforms (TLPs)” First
Edition, June 1998.
API RP 2SK “ Recommended Practice for Design and
Analysis of Station Keeping Systems for Floating
structures” Second Edition, December 1997.
SINTEF (1992) “FPS 2000 Flexible Pipes and Risers:
Handbook on Design and Operation of Flexible pipes “,
Sintef report STF70 A92006, ISBN no 8259572664.
L 200 Technical references
Bech A, Skallerud B, Sødahl N (1992) “Structural Damping
in Design Analyses of Flexible Risers” Marinflex’92,
London, November 1992.
Burke N E, Kashou S F (1995) “Slug Sizing/Slug Volume
Prediction, State of the art review and simulation” OTC
paper no 7444.
Colby C, Sødahl N, Katla E, Okkenhaug S (2000)
“Coupling Effects for a Deepwater Spar”, OTC 12083
DNVOSF201 Dynamic Risers, October 2010
Page 66 Appendix A
DET NORSKE VERITAS
Clough R W, Penzien J (1975) : “ Dynamics of Structures”
McGrawHill 1975.
Engseth, A, Bech, A, Larsen, C M (1988) “Efficient Method
for Analysis of Flexible Risers” Proc. BOSS 1988.
Espinasse, P P, Schnittker, D A and Sturdevant, L A (1989):
“Design and Application of a Dynamic Flexible Pipe Array
for Placid's GC 29 FPS” Paper No 6164, Proc. OTC,
Houston.
Fylling I J , Sødahl N, Bech A (1986) “ On the Effects of
Slug Flow on the Response of Flexible Risers and
Submerged Hoses” IBC Technical Services, London 1986.
Karunakaran, D, Nordsve, N T, Olufsen A (1996): “An
Efficient Metal Riser Configuration for Ship and Semi
Based Production Systems” Proc. ISOPE 1996, Los
Angeles.
Newman J N (1977) “Marine Hydrodynamics”, Cambridge:
The MIT press.
O’Brien, P J, McNamara, J F, (1989) “Significant
Characteristics of Threedimensional Flexible Risers
Analysis” Engineering Structures, Vol 11,1989.
Ormberg H, Fylling I. J., Larsen K., Sødahl N. (1997):
“Coupled Analysis of Vessel Motions and Mooring and
Riser System Dynamics” Proc. OMAE 1997.
Ormberg H., Larsen K. (1997): “Coupled Analysis of
Floater Motion and Mooring Dynamics for a Turret Moored
Tanker” Proc. BOSS 1997.
Ormberg H, Sødahl N, Steinkjer O (1998) “ Efficient
Analysis of Mooring Systems using Decoupled and
Coupled Analysis” OMAE 98.
Passano E (1995) “Reducing Simulation Lengths in Extreme
Response Estimation of a Flexible Riser” Proc. OMAE
1995.
Patel M H, Seyed F B (1989) “Internal flowinduced
behaviour of flexible risers” Engineering Structures, 1989,
Vol 11, No 4.
Peyrot, A.H., Goulouis, A.M (1979) “Analysis of Cable
Structures”, Computers and Structures, Vol 10, 1979
Phifer, E.H., Kopp,F., Swanson, R.C., Allen, D.W. and
Langner, C.G.(1994): ”Design and Installation of Auger
Steel Catenary Risers.” Paper No. 7620, Proc. OTC,
Houston.
Rooney P, Engebretsen K, Pettersen D J (1990): “TLP Rigid
Riser, a Case Study” OTC 6435, Offshore Technology
Conference, Houston.
Sanderson N, Moros T, Fairhurst P,Harrison G (1999):
“Design and Practical Considerations of Steel Catenary
Riser and Pipeline Systems” Proc. of 2
nd
workshop on
subsea pipelines, Rio de Janeiro.
Sarpkaya T , Isaacson M (1981) “Mechanics of Wave
Forces on Offshore Structures” Van Nostrand Reinhold
Company 1981.
Skomedal, E (1990) “Buckling Calculation of Offshore
Platform Conductors” 1st European Offshore Mechanics
Symposium, Trondheim, 1990.
Sparks, C P (1984): “The Influence of Tension, Pressure
and Weight on Pipe and Riser Deformations and Stresses.”
Journal of Energy Resources Technology, ASME,Vol 106.
Sumer, B M, Fredsøe, J (1997) “Hydrodynamics around
cylindrical structures”. World Scientific Publishing.
Sødahl N., Larsen C.M (1989) “ Design Procedure for
Bending Stiffeners in Flexible Riser Systems” PRADS 89.
DNVOSF201 Dynamic Risers, October 2010
Appendix B Page 67
DET NORSKE VERITAS
APPENDIX B FATIGUE ANALYSIS
Contents
A. General
A 100 Objective
A 200 Application
A 300 Fatigue design
A 400 Methods for fatigue damage assessment
B. Fatigue analysis procedures
B 100 General
B 200 Basic fatigue damage methodology
B 300 Global fatigue analysis procedures
C. Narrow Band Fatigue Damage Assessment
C 100 General
C 200 Narrow Band Gaussian Fatigue Damage
C 300 Narrow Band NonGaussian Fatigue damage
D. Wide band Fatigue Damage Assessment
D 100 General
D 200 Cycle counting
D 300 Semiempirical Solutions
D 400 Analytical Solutions for Bimodal Spectra
E. Fatigue Capacity SN Curves
E 100 General
A. General
A 100 Objective
101 This objective of this Appendix is to support section
5.E on fatigue assessment of risers subjected to repeated
fluctuations and provide details on fatigue analyses methods
recommended in Section 4.C 200. Extensive references are
given to the DNV RPC203, Fatigue Strength Analysis.
A 200 Application
201 The assessment procedure assumes that the riser has
been designed in accordance with all other requirements in
this standard.
A 300 Fatigue design
301 In general, the fatigue life of a component can be
broken down into two phases: Crack initiation and
propagation. In the case of unwelded components (e.g.,
seamless pipes and machined components), the crack
initiation period represents the bulk of the total fatigue life.
This is particularly noticeable at high fatigue lives where the
fatigue crack initiation period may exceed 95% of the
fatigue life. In the case of machined components, once a
fatigue crack has grown to a detectable size, the component
is virtually at the end of its useful life and will normally be
withdrawn from service if repair is not possible.
302 In the case of welded joints, weld toe/root
discontinuities are generally present. These behave as pre
existing cracks. Consequently, the bulk of the fatigue life of
a welded joint can be attributed to fatigue crack propagation.
303 The difference in the crack initiation phase of parent
material and welded joints has significant effects on overall
fatigue performance. In general, the fatigue strength of an
unwelded component increase with material tensile strength
due to the increased initiation life associated with higher
strength materials. In the case of welded joints however, the
fatigue strength is relatively unaffected by material tensile
strength because the bulk of the fatigue life of a welded joint
is spent in the propagation phase. Although crack
propagation rates can change from one material to another
and from one environment to another, there is no consistent
trend with regard to tensile strength.
A 400 Methods for fatigue damage assessment
401 A typical sequence in fatigue design of a riser is
shown in Table B1.
Table B1 Summary of a typical fatigue assessment
procedure
Task Comment
Define fatigue
loading.
Based on operating limitations
including WF, LF and possible VIV
load effects.
Identify locations to
be assessed.
Structural discontinuities, joints (girth
pipe welds, connectors, bolts), anode
attachment welds, repairs, etc.
Global riser fatigue
analysis.
Calculate shortterm nominal stress
range distribution at each identified
location.
Local joint stress
analysis.
Determination of the hotspot SCF from
parametric equations or detailed finite
element analysis.
Identify fatigue
strength data.
SNcurve depend on environment,
construction detail and fabrication
among others.
Identify thickness
correction factor
Apply thickness correction factor to
compute resulting fatigue stresses
Fatigue analyses Calculate accumulated fatigue damage
from weighted shortterm fatigue
damage.
Further actions if
too short fatigue
life.
Improve fatigue capacity using.:
÷ more refined stress analysis
÷ fracture mechanics analysis.
÷ change detail geometry
÷ change system design.
÷ weld profiling or grinding
÷ improved inspection /replacement
programme
For a general introduction to methodology for fatigue
damage assessment, reference is made to DNV RPC203
Sec 1.3.
DNVOSF201 Dynamic Risers, October 2010
Page 68 Appendix B
DET NORSKE VERITAS
B. Fatigue analysis procedures
B 100 General
101 Three different contributions to fatigue damage
should be addressed: The waveinduced, the lowfrequency
and the vortexinduced stress cycles. The former two are
addressed in the following, while the latter is described in
Appendix E.
102 A general approach for calculation of wave and low
frequency fatigue damage contributions is based on
application of the following procedure:
÷ the wave environment scatter diagram is subdivided
into a number of representative blocks;
÷ within each block, a single seastate is selected to
represent all the seastates within the block. The
probabilities of occurrence for all seastates within the
block are lumped to the selected seastate;
÷ the fatigue damage is computed for each selected short
term seastate for all the blocks;
÷ the weighted fatigue damage accumulation from all sea
states can be expressed as:
∑
·
·
s
N
1 i
i i L
P D D
(B.1)
Where
D
L
Longterm fatigue damage
N
s
Number of discrete sea states in the wave
scatter diagram
P
i Sea state probability. Normally parameterised
in terms of significant wave height, peak
period and wave direction, i.e. P(H
s
,T
p
,θ
)
i
D
i
Short term fatigue damage
B 200 Basic fatigue damage methodology
201 The basic fatigue capacity is given in terms of SN
curves expressing the number of stress cycles to failure, N,
for a given constant stress range, S:
m
S a N
−
·
(B.2)
or equivalently:
) S log( m ) a log( ) N log( − ·
(B.3)
Where a and m are empirical constants established by
experiments.
202 The stress range to be applied in fatigue damage
calculations is found by application of a stress concentration
factor as well as a thickness correction factor to the nominal
stress range :
k
ref
3
0
t
t
SCF S S
,
`
.

⋅ ⋅ ·
(B.4)
Where:
S
0
Nominal stress range
SCF Stress concentration factor
k
ref
3
t
t
,
`
.

Thickness correction factor
The thickness correction factor applies for pipes with a wall
thickness t
3
greater than a reference wall thickness, t
ref
=25mm. The thickness exponent, k, is a function of the
actual structural design and hence also related to SN curve,
see DNV RPC203 Sec 2. and section E for further details.
203 Bilinear (twoslope) SN curves in loglog scale are
also frequently applied for representation of the
experimental fatigue capacity data, i.e.
¹
'
¹
≤ ⋅
> ⋅
·
−
−
sw
m
2
sw
m
1
S S S a
S S S a
N
2
1 (B.5)
m
1
and m
2
are fatigue exponents (the inverse slope of the bi
linear SN curve) and
1
a and
2
a are characteristic fatigue
strength constant defined as the meanminustwostandard
deviation curve. S
sw
is the stress at intersection of the two
SNcurves given by:
,
`
.
 −
·
1
sw 1
m
) N log( ) a log(
sw
10 S
(B.6)
N
sw
is the number of cycles for which change in slope
appears. Log(N
sw
) is typically 67. For further details
reference is made to DNV RPC203.
1
10
100
1000
1.E+03 1 E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 1 E+10
No of cycles, N
S
t
r
e
s
s
R
a
n
g
e
,
S
NSW
SSW
(a1;m1)
(a2;m2)
Figure B1 Basic definitions for twoslope SNcurves
204 The MinerPalmgren rule is adopted for
accumulation of fatigue damage from stress cycles with
variable range:
∑
·
i i
i
) S ( N
) S ( n
D
(B.7)
Where n(S
i
) is the number of stress cycles with range S
i
and
N(S
i
) is the number of stress cycles to failure as expressed
by B.3.
The expected fatigue damage per unit time can for a linear
SN curve in loglog scale be expressed as:
DNVOSF201 Dynamic Risers, October 2010
Appendix B Page 69
DET NORSKE VERITAS
[ ]
∫
∞
· ·
0
m 0
S
m 0
S E
a
f
ds ) s ( f s
a
f
D
(B.8)
Where f
0
is the mean number of stress cycles per unit time
and f
S
(s) is the probability density function (PDF) for the
stress cycles. The expected fatigue damage is hence directly
related to the mth order moment, E[S
m
] (or µ
m
) of the stress
cycle PDF. For a bilinear SNcurve in loglog scale the
corresponding expression becomes:
∫ ∫
∞
+ ·
sw
sw
1 2
S
0 S
S
m
1
0
S
m
2
0
ds ) s ( f s
a
f
ds ) s ( f s
a
f
D
(B.9)
Equation (B.8) and (B.9) constitutes the basic formulation
for assessment of the shortterm fatigue damage in each
stationary environmental condition as expressed by (B.1).
(B.8) and (B.9) can also be applied to compute the long
term fatigue damage directly from the longterm distribution
of stress cycles. For an introduction to methodology for
establishment of longterm response distributions, reference
is made to Annex C.
B 300 Global fatigue analysis procedures
301 The basis for fatigue damage calculations is global
load effect analyses to establish the stress cycle distributions
in a number of stationary shortterm environmental
conditions. The general principles for selection of analysis
methodology and verification of simulation model as
outlined in Annex A and Annex D respectively should be
adhered to.
302 The shortterm fatigue conditions should be selected
carefully to give an adequate representation of the stress
cycles for the lifetime of the riser system. The selection
must be based on a thorough physical knowledge regarding
static and dynamic behaviour of the riser system with
special attention to FE modelling, hydrodynamic loading,
resonance dynamics and floater motion characteristics.
Sensitivity studies should be performed to support rational
conservative assumptions regarding identified uncertain
parameters (e.g. soil properties for fatigue analysis in the
touchdown area of SCR’s)
303 Fatigue analysis will normally involve global load
effect analyses in a number of low to moderate seastates.
This is because the main contribution to the total fatigue
damage in most cases comes from low to moderate sea
states with high probability of occurrence rather than a few
extreme seastates. Compared to extreme response analysis,
the degree of nonlinearity involved is generally smaller.
Adequate results can hence be obtained by use of linearized
time domain or frequency domain analyses in many cases.
However, any use of simplified analysis methodology shall
be verified against nonlinear time domain analyses.
304 The fatigue damage will generally have contributions
from wave frequency (WF) as well as low frequency (LF)
stress cycles. The WF floater motions as well as direct wave
loading on the riser govern WF fatigue damage, while the
LF floater motions govern LF fatigue damage. The relative
importance of WF and LF fatigue damage is strongly system
dependent and will in addition vary significantly with the
location along the riser. It is always recommended to do an
assessment of the relative contributions from WF and LF
stress cycles to the fatigue damage to support rational
decisions regarding choice of method of analysis. LF fatigue
damage may be disregarded if it is documented by proper
analyses that the LF fatigue damage is negligible when
compared to WF fatigue damage.
305 Adequate fatigue life shall be documented for all
parts of the riser system. Examples of critical areas wrt.
fatigue damage of metallic risers are given in the following:
÷ the areas close to upper/lower termination of top
tensioned risers will normally experience significant
dynamic bending stress variation. Fatigue close to upper
termination is normally governed by WF stress cycles
while LF response may be of significance close to the
seafloor termination. Accurate modelling of boundary
conditions and stiffness properties is required (e.g.
taper joints, stiffness characteristics of flexjoints etc);
÷ the splash zone is normally a critical area for top
tensioned as well as compliant riser configurations
mainly due to WF bending stress cycles. Description of
wave loading up to actual wave elevation is of vital
importance for accurate prediction of fatigue damage
Due regard should also be given to possible
disturbances in the wave kinematics caused by the
presence of the floater. Time domain analyses
supported by sensitivity studies to confirm adequacy of
load model is recommended (i.e. results are sensitive to
mesh size as well as wave kinematics);
÷ seafloor touchdown area is a critical area for steel
catenary risers and other proposed compliant riser
configurations. Soil properties, mesh size and mean
floater position are important for prediction of fatigue
damage. Time domain analyses are generally
recommended together with sensitivity studies to
support rational conservative assumptions regarding
soil properties. The adequacy of the mesh applied in the
touchdown area should also be confirmed by sensitivity
studies, and
÷ considerations regarding resonance dynamics and
combined WF and LF fatigue damage are of special
importance for spar risers (in particular for integral air
can solutions). Critical locations are typically close to
riser supports in the hull area. Special attention should
be given to possible LF stress cycles at the keel joint.
C. Narrow Band Fatigue Damage
Assessment
C 100 General
101 The basic assumption in narrowband fatigue damage
estimation is that the stress cycles (S) can be determined
directly from the stress maxima (S
a
). Each cycle’s range is
assumed to be twice the value of the corresponding value of
the local stress maximum, yielding:
DNVOSF201 Dynamic Risers, October 2010
Appendix B Page 70
DET NORSKE VERITAS
a
S 2 S ⋅ = (B.10)
Furthermore, the number of stress cycles per unit time is
given directly by the zero crossing frequency, f
0
of the stress
response process.
C 200 Narrow Band Gaussian Fatigue Damage
201 If the stress response process is assumed to be narrow
banded and Gaussian, the distribution of local stress
maxima, S
a
, is defined by a Rayleigh probability density as:
( )
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
σ
−
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
σ
=
2
2
a
2
a
a S
2
s
exp
s
s f
(B.11)
where s
a
is the local stress maximum and σ is the standard
deviation of the stress response process.
202 For a linear SNcurve (in loglog scale) the fatigue
damage per unit time can be expressed as:
( ) ⎟
⎠
⎞
⎜
⎝
⎛
+ Γ σ = 1
2
m
2 2
a
f
D
m
0
(B.12)
where Γ(⋅) is the gamma function given by
∫
∞
− ϕ −
= ϕ Γ
0
1 t
dt t e ) (
(B.13)
203 For an SN curve which is bilinear on loglog scale,
the fatigue damage becomes
( )
( )
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
σ
⎟
⎠
⎞
⎜
⎝
⎛
+ ⋅
σ ⋅
+
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
σ
⎟
⎠
⎞
⎜
⎝
⎛
+
σ ⋅
=
2
sw 2
2
2
m
0
2
sw 1
1
1
m
0
2 2
S
;
2
m
1 G
a
2 2 f
2 2
S
;
2
m
1 G
a
2 2 f
D
2
1
(B.14)
where G
1
and G
2
are the complementary incomplete Gamma
function and incomplete Gamma function, respectively
∫
∫
− −
∞
− −
=
=
x
t
x
t
dt t e x G
dt t e x G
0
1
2
1
1
) , (
) , (
ϕ
ϕ
ϕ
ϕ
(B.15)
and where ) log(
1
a and m
1
are the intercept and inverse
slope of the right leg of the bilinear SN curve, ) log(
2
a and
m
2
are the intercept and inverse slope of the left leg of the
bilinear SN curve, and S
SW
is the stress range at the knee of
the bilinear SN curve.
204 The fatigue damage is hence directly expressed by
the standard deviation and zerocrossing frequency of the
stress response process This formulation is of special
convenience for frequency domain analyses where results
from the global analyses are expressed in terms of the auto
spectral density, S
σσ
(ω), of the stress response process.
The standard deviation, σ and zero crossing frequency f
0
are
hence given as:
0
λ σ =
(B.16)
0
2
0
2
1
λ
λ
π
= f
(B.17)
where λ
n
is the n
th
order spectral moment of the response,
given by
∫
∞
=
0
) ( ω ω ω λ
σσ
d S
n
n
(B.18)
C 300 Narrow Band NonGaussian Fatigue
damage
301 For time domain analyses, the twoparameter
Weibull distribution model is frequently employed as a
generalisation of the Rayleigh distribution for the local
maxima (i.e., for NonGaussian stressresponse processes).
The Weibull probability density function is given by:
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
⎟
⎠
⎞
⎜
⎝
⎛
− =
− −
β
β β
α
β α
a
a a S
s
s s f exp ) (
1
(B.19)
Note that the Rayleigh distribution in (B.11) is obtained for
β=2 and σ = α 2
The Weibull distribution may be fitted to the shortterm (or
longterm) distribution of the local maxima. The Weibull
distribution parameters (α: scale, β: shape) are linked to the
statistical moments σ μ ˆ , ˆ for the local maxima as follows:
2
1
1
2
1 ˆ
1
1 ˆ
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
β
+ Γ −
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
β
+ Γ α = σ
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
β
+ Γ α = μ
(B.20)
These equations can be used to establish moment estimates
of the distribution parameters with basis in sample estimates
σ μ ˆ , ˆ from time domain simulations.
302 The fatigue damage per unit time in the general case
of a bilinear SNcurve can then be expressed analytically as
follows:
( )
( )
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
⎟
⎠
⎞
⎜
⎝
⎛
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ ⋅
⋅
+
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
⎟
⎠
⎞
⎜
⎝
⎛
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+
⋅
=
β
β
α β
α
α β
α
2
; 1
2
2
; 1
2
2
2
2
0
1
1
1
0
2
1
sw
m
sw
m
S m
G
a
f
S m
G
a
f
D
(B.21)
DNVOSF201 Dynamic Risers, October 2010
Appendix B Page 71
DET NORSKE VERITAS
D. Wide band Fatigue Damage
Assessment
D 100 General
101 For marine risers, the stress response is normally
neither narrowbanded nor completely widebanded. In a
wideband response a strict relationship between the stress
cycles and stress maxima and minima do not exist. For this
reason the distribution of stress cycles can not be evaluated
accurately from the distribution of stress maxima. The
following procedures exist to describe fatigue damage for a
wide band process:
⎯ cycle counting algorithms,
⎯ semiempirical solutions, or
⎯ simplified analytical solutions
102 Wide band fatigue assessment is of special
importance for fatigue assessment of combined WF/LF
stress response. It is in general applicable to results from
time domain analyses but can also be applied in connection
with frequency domain analyses through a transformation of
frequency domain results to time domain (by e.g. FFT
simulation)
D 200 Cycle counting
201 The fatigue damage may be obtained by counting the
stress cycles in the actual or simulated stress time histories.
Specials purpose counting algorithms have been developed
with techniques applicable to nonGaussian stress time
histories. The recommend method is the Rain Flow
Counting (RFC) method.
202 The RFC method provides an estimate of the stress
probability density function (i.e. sample estimate of f
S
(S)
and of the average number of stress cycles per unit time.).
For a linear SN curve, (B.8) can subsequently be applied
for estimation of fatigue damage in each stationary short
term condition. Extension to more general SN curves (e.g.
bilinear) is straightforward.
203 The response process due to combined wave and
low frequency excitation is generally widebanded. Time
domain simulation and cyclecounting procedures will
accordingly be relevant.
204 Cycle counting methods represent time domain
estimates of fatigue damage. Statistical uncertainties will
therefore always be present in the estimates. Sensitivity
studies should therefore be conducted to document that
adequate fatigue damage estimates have been obtained. This
is of special importance for combined WF/LF stress time
histories or in cases with SNcurves with large (inverse)
slope (i.e. large ‘m’).
D 300 Semiempirical Solutions
301 A number of semiempirical expressions have been
proposed in the literature to correct the narrow band fatigue
damage calculation for the effects of a broad bandwidth. An
often used approach is based on the assumption that the true
damage D
RFC
(i.e. using a rain flow counting technique) can
be established from a corrected narrowband result:
RFC NB RFC
D D κ = (B.22)
where D
NB
is the narrow banded Gaussian fatigue damage
given by C 200 and κ
RFC
is a correction factor. Wirshing and
Light, see e.g. Barltrop & Adams proposed the following
expression:
323 . 2 m 587 . 1 b
m 033 . 0 926 . 0 a
where
) 1 )( a 1 ( a ) m (
b
RFC
− =
− =
ε − − + = κ
(B.23)
where ε is the bandwidth parameter defined by (note that
ε=1 for a broad banded process and ε=0 for a narrow banded
process):
4 0
2
2
1
λ λ
λ
ε − =
(B.24)
302 As a promising alternative, Dirlik, see e.g. Barltrop
& Adams proposed an empirical closed form expression for
the stress probability density function.
D 400 Analytical Solutions for Bimodal Spectra
401 Accurate analytical solutions to fatigue damage
estimates can be obtained for wellseparated bimodal stress
spectra (e.g. a process with a combination of low frequency
and wave frequency Gaussian component). Reference is
given to Jiao & Moan (1990), where a correction function
on a form similar to (B.22) have been derived by analytical
means assuming two independent narrowbanded Gaussian
process.
402 In case the process may be assumed to be composed
of two independent Gaussian stress response processes an
upper bound on the estimated fatigue damage can be
obtained by adding the variances of the contributions
directly. The zerocrossing frequency may be expressed as a
combination of the respective zerocrossing frequencies
based on expressions for the sum of two independent
Gaussian processes.
E. Fatigue Capacity SN Curves
E 100 General
101 The fatigue design is based on use of SN curves
obtained from fatigues testing. For practical fatigue design,
welded joints are divided into several classes, each with a
corresponding SN curve. Fatigue capacity data for joint
classifications of relevance for risers are given in Table B2.
The joint classifications apply to typical joints/details for
risers subjected to cyclic bending moment and tension. For
further details, reference is made to DNV RPC203, Sec 2.3
and Appendix 1.
102 If fatigue data does not exist for the material, detail
and environment under consideration, SN curves should
either be developed by testing, use of fracture mechanics
assessment or by use of lower bound SN curves. Special
DNVOSF201 Dynamic Risers, October 2010
Page 72 Appendix B
DET NORSKE VERITAS
care shall be taken for chemical environments not covered
by DNV RPC203.
103 A stress concentration factor (SCF) applies to
account for possible stress magnification due to imperfect
geometry of two adjacent joints (e.g. due to fabrication
tolerances and installation procedures). The SCF may be
calculated by detailed FE analyses or by closed form
expressions for the actual structural detail. The following
closed form expression applies for welded riser joints /1/:
( )
5 . 0
3
3
) t / D ( exp
t
e 3
1 SCF
−
− + =
(B.25)
where e is the representative eccentricity due to geometrical
imperfections to be applied in fatigue design. Assessment of
the representative eccentricity shall be based on detailed
information regarding production tolerances and
installation/welding procedures supported by rational
conservative assumptions as appropriate for the actual
design.
DNVOSF201 Dynamic Risers, October 2010
Appendix B Page 73
DET NORSKE VERITAS
Table B2 SN curves for Risers
Description
Welding Geometry and hot
spot
Tolerance
requirement
2)
SN curve;
according to
DNV RPC203
Thickness
exponent
k
SCF
e ≤ min(0.1t
3
, 3 mm)
F1 0.00 1.0 Single side
e > min(0.1t
3
, 3 mm)
e ≤ min(0.1t
3
,5.4 mm)
F3 0.00 1.0
e ≤ min(0.1t
3
, 2 mm)
F 0.00 1.0 Single side
on backing
e > min(0.1t
3
, 2 mm)
e ≤ min(0.15t
3
, 4 mm)
F1 0.00 1.0
Single side
e ≤ min(0.15t
3
, 4 mm)
D 0.15
1)
Eq. (E.1)
Double side
e ≤ min(0.15t
3
, 4 mm)
D 0.15
1)
Eq. (E.1)
Seamless pipe B1 0.00 1.0
Machined
components
B1 0.00 FEM
analysis
Automatic
longitudinal
seam welded
pipes
B2 0.00 1.0
Steel bolts
and threaded
joints in
tension
F1 (coldrolled)
W3 (cut threads) 0.40
1)
1.0
NOTE 1 The thickness penalty applies only for thickness greater than 25 mm. No benefit can be taken for
sections thinner than 25 mm. For bolts, the reference thickness is the stress diameter.
NOTE 2 For girth weld eccentricities greater than 0.15t
3
or 4 mm, whichever is the smaller, special
considerations apply, e.g. engineering critical assessment.
F. References
Barltrop & Adams “Dynamics of Fixed Offshore
Structures”, Third Edition, Butterworth & Heinemann.
Jiao, G. & Moan, T. “Probabilistic Analysis of Fatigue due
to Gaussian Load Processes”, Probabilistic Engineering
Mechanics, 1990, Vol. 5, No.2.
DNVOSF201 Dynamic Risers, October 2010
Page 74 Appendix C
DET NORSKE VERITAS
APPENDIX C ASSESSMENT OF EXTREME LOAD EFFECT FOR
COMBINED LOADING
Contents
A. General
A 100 Objective
B. Design principles
B 100 General
B 200 Design based on environmental statistics
B 300 Design based on response statistics
C. Implementation of the LRFD design format
C 100 General
C 200 Generalised load effect
C 300 Shortterm acceptance criteria
C 400 Long term acceptance criteria
C 500 ULS Analysis Procedure
C 600 Post processing procedures
C 700 Computer implementation
D. Implementation of the WSD design format
D 100 General
D 200 Implementation in design analyses
E. Shortterm extreme load effect estimation
E 100 General
E 200 Envelope statistics
E 300 Extreme response estimation
E 400 Statistical uncertainty and simulation planning
F. Longterm load effect statistics
F 100 General
F 200 Response surface approach
G. References
G 100 Standards, Guidelines and Handbooks
G 200 Technical references
A. General
A 100 Objective
101 The objective of this Appendix is to provide an
introduction to practical implementation of LRFD and WSD
design checks for combined loading based on the generalised
load effect formulation introduced in Section 3. The main
focus is on consistent implementation of the LRFD design
format.
102 Two fundamentally different methods can be applied
for assessment of the characteristic load effects:
÷ Based on environmental statistics
÷ Based on response statistics
The purpose of this document is to give an outline of these
strategies with emphasis on the computational efforts
involved in practical applications as well as inherent
limitations.
B. Design principles
B 100 General
101 Riser systems in general are highly nonlinear
structures due to nonlinearities introduced by hydrodynamic
loading, geometric stiffness, large rotations in 3D space and
possible material nonlinearities as well as seafloor contact.
The relative importance of these nonlinearities is strongly
system and excitation dependent. Time domain finite
element analysis therefore constitutes the primary method of
analysis for slender structures. For a more detailed
discussion, reference is made to Appendix A.
B 200 Design based on environmental statistics
201 It has traditionally been common practice to adopt the
extreme response found by exposing the system to multiple
stationary design environmental conditions as the
characteristic extreme response. Each design condition is
hence described in terms of a limited number of
environmental parameters (e.g. Hs, Tp etc) and a given
duration (e.g. 36 hours). Different combinations of wind,
wave and current yielding the same return period for the
combined environmental condition are typically applied.
Examples of relevant combinations to obtain 100years
design conditions are given in Table C1. The ‘associated’
return periods must be assessed with basis in environmental
statistics for the actual location.
Table C1 Examples of typical 100years design
environmental conditions
Case Wind
[year]
Waves
[year]
Current
[year]
A 100 Associated Associated
B Associated 100 Associated
C Associated Associated 100
202 Wind loading is indirectly included in the global riser
system analysis as an important contributor to mean floater
position and low frequent floater (LF) motions. Waves and
wave frequent (WF) floater motions are included as dynamic
loading in the global riser system analysis while LF floater
motions normally are included as a representative static
offset. The offset accounting for LF motions is additional to
the mean floater offset governed by mean environmental
loading. Reference is given to Appendix A for further
discussion of analysis strategies. For guidance regarding
calculation of representative floater offset, reference is made
to e.g. section 6.2.2 in API RP 2SK.
DNVOSF201 Dynamic Risers, October 2010
Appendix C Page 75
DET NORSKE VERITAS
203 Either regular or irregular wave loading is considered
in the global response analyses. The former is denoted design
wave approach while the latter is denoted a design storm
approach.
204 The most severe directional combination of wind,
waves and current consistent with the environmental
conditions at the actual site should be applied for permanent
installations. This will in most cases imply that waves,
current and floater offset are (conservatively) assumed to act
in the same direction. Analyses are performed for assumed
critical directions (typically ‘near, far and cross’ conditions)
and final characteristic response is identified as the most
unfavourable from the analyses.
205 A wave period variation shall in addition be
performed to identify the most unfavourable loading
condition. At least 3 different periods covering a realistic
variation range (e.g. 90% confidence interval) should be
considered. Alternatively, simultaneous variation of wave
height and period (e.g. Hs, Tp) as described by environmental
contours can be applied for more consistent identification of
critical conditions.
206 The main problem related to design criteria based on
environmental statistics is that the return period for the
characteristic load effect is unknown due to the nonlinear
dynamic behaviour of most riser systems. This will in general
lead to an inconsistent safety level for different design
concepts and failure modes. Acceptable results can however
be expected for quasistatic systems with moderate
nonlinearities. A verification of design criteria should be
performed in the following situations:
÷ New concepts
÷ Systems with significant nonlinear response
characteristics
÷ Dynamically sensitive systems
The verification should be based on longterm extreme load
effect assessment as discussed in section F for critical
conditions.
B 300 Design based on response statistics
301 Consistent assessment of the Dyear load effect will in
general require a probabilistic response description due to the
longterm environmental action on the riser system. The load
effect with a return period of Dyears, denoted x
D
, can
formally be found from the longterm load effect distribution
as:
D
D X
N
x F
1
1 ) ( − ·
(C.1)
where :
N
D
 total number of load effect maxima during D years.
F
X
(x)  longterm peak distribution of the (generalised) load
effect
302 The main challenge related to this approach is to
establish the longterm load effect distribution due to the non
linear dynamic behaviour experienced for most riser systems.
Guidance to possible computational strategies is further
outlined in F. The described procedures have been applied for
assessment of design loads for riser systems in research
activities but are yet not established as standard industry
design practise. However, design based on response statistics
is in general the recommended procedure and should be
considered whenever practical possible for consistent
assessment of characteristic load effects (especially for
verification purposes when shortcomings in the traditional
approach based on environmental statistics have been
identified). This is of particular importance for ULS
conditions which normally are associated with the most
pronounced nonlinear response characteristics.
C. Implementation of the LRFD
design format
C 100 General
This section gives an introduction to consistent
implementation of LRFD capacity checks for combined
loading considering global time domain analysis. Main focus
is placed on implementation of design equations for ULS
conditions as this is the most general approach. Relevant
simplifications in case of SLS and ALS conditions are briefly
discussed. Acceptance criteria are established for design
based on environmental statistics (short term approach) as
well design based on response statistics (long term approach).
Statistical techniques for extreme load effect estimation for a
short and log term design approach are discussed separately
in E and F respectively.
C 200 Generalised load effect
201 Consistent treatment moment/tension correlation is a
key issue for efficient capacity checks for combined loading.
For this purpose it is convenient to consider generalised
loading expressed by the following generic equation:
) , , p ), t ( T ), t ( M ( g ) t ( g
k ed d
Λ ∆ · R
(C.2)
Where g(t) is the generalised load effect (or utilisation
function) at a specific location on the riser and M
d
, T
ed
denote
design values for bending moment and effective tension,
respectively. Furthermore, ∆p denote the local differential
pressure,
k
R is a vector of crosssectional capacities and Λ is
a vector of safety factors (i.e. material, safety class and
condition factors). The importance of this formulation is that
the combined time dependent action of bending moment and
tension is transformed into a scalar process expressed by the
generalised load effect.
C 300 Shortterm acceptance criteria
301 The code checks for combined loading in a stationary
design condition is hence reduced to extreme value prediction
of the generalised load effect, i.e.
DNVOSF201 Dynamic Risers, October 2010
Page 76 Appendix C
DET NORSKE VERITAS
1
max
≤ g (C.3)
Where
max
g is a representative extreme value of ) (t g . The
maximum value of ) (t g applies in a design wave approach
(excluding the startup transient) , while statistical extreme
value prediction is required in a design storm approach, see
below
302 The standard framework for response processing of
results from time domain analyses can therefore be directly
applied for code checks. This will typically include
application of response envelopes in case of regular wave
analysis and statistical extreme value prediction in case of
irregular wave analysis
303 Statistical estimation of the expected extreme value
(or most probable extreme value) for a given duration (e.g. 3
6 hours) is hence required in case of irregular analyses. It
should however be noted that g(t) always will be a non
Gaussian response process. This is because the bending
moment components and effective tension normally are non
Gaussian response processes and because the limit state
function defines a nonlinear transformation of these time
series. Expected extremes of nonGaussian time histories are
in practical applications normally estimated from a
parametric probabilistic model (e.g. Weibull) fitted to the
simulated realisation of individual response peaks (i.e. peaks
of g(t)). Reference is made to E for a further discussion
304 This approach will automatically account for the
correlation between effective tension and bending moment
components and is hence capable of optimal design (i.e.
allow for maximum utilisation).
305 Conservative estimates always could be obtained by
separate estimation of design values for effective tension and
resulting bending moment disregarding correlation effects
which formally may be expressed as:
( ) 1 , , p , T , M g
k
max
ed
max
d
≤ Λ ∆ R
(C.4)
where indices
max
indicate extreme values. This approach may
yield acceptable result when the design is driven by one
dominating dynamic component (typically bending moment
for top tensioned risers with well functioning heave
compensation system).
C 400 Long term acceptance criteria
Consistent extreme load effect estimate for combined loading
can be found as a percentile in the longterm distribution of
the generalised load effect. The acceptance criterion can
hence be expressed as:
. 1 x
D
≤ (C.5)
Where x
D
is the percentile in the long term (generalised) load
effect distribution corresponding to a return period of D
years. Techniques for establishment of the longterm load
effect distribution are discussed separately in F.
C 500 ULS Analysis Procedure
501 Separation of global response into components due to
functional and environmental loading is an additional key
issue for ULS analyses, which require due consideration of
analysis strategy as well as response post processing.
502 The basic force output from global time domain
analyses are simultaneous time series of bending moments
and effective tension. These response quantities contain
contributions due to functional as well as environmental
loading. Separation of these quantities into components
requires that the static configuration due to functional loading
is determined separately. The following analysis sequence
can be applied:
1) Static analysis  functional loading.
The purpose of the 1st step in the analysis sequence is to
establish the static equilibrium configuration due to
functional loading (i.e. effective weight and nominal floater
position). The analysis is typically started from an initial
stressfree configuration with incremental application of
functional loading to reach the final solution. The static force
output is two axial bending moments and effective tension
due to functional loading:
[ ]
eF
zF yF F
T
M , M M ·
r
(C.6)
2) Static analysis  environmental loading.
This analysis is restarted from 1) considering additional
loading due to steady current and mean floater offset due to
environmental actions.
3) Dynamic time domain analysis  environmental loading.
This analysis is restarted from 2) considering additional
relevant dynamic environmental loading on the system (e.g.
loading due to wave action and floater motions, possible slug
flow etc) The force output is simultaneous time histories of
two axial bending moment and effective tension:
[ ]
) t ( T
) t ( M ), t ( M ) t ( M
e
z y
·
r
(C.7)
503 The referred global response quantities are assumed to
contain the total response, i.e. dynamic components from
environmental loading as well as static components due to
functional and environmental loading. This is in accordance
with the storage and output conventions applied in the
majority of tailor made computer codes for slender structure
analysis
In fact, this analysis sequence is convenient for application
of static and dynamic loading and is used in the vast majority
of design analyses. The distinction between static and
dynamic environmental loading is always a key issue that
must be evaluated carefully in view of the actual concept
(e.g. static vs. dynamic current and LF floater motions). The
only additional effort needed from the analyst is hence
separate storage and treatment of the static response due to
functional loading.
DNVOSF201 Dynamic Risers, October 2010
Appendix C Page 77
DET NORSKE VERITAS
C 600 Post processing procedures
601 The post processing to compute the generalised load
effect based on output from the ULS analysis procedure
described in the previous section can be summarised in the
following steps:
1. Establish response components due to environmental
loading:
eF e eE
F E
T ) t ( T ) t ( T
M ) t ( M ) t ( M
− ·
− ·
r r r
(C.8)
2. Establish design values
( ) ( )
) t ( T T ) t ( T
) t ( M M ) t ( M M
) t ( M M ) t ( M
eE E eF F ed
2
zE E zF F
2
yE E yF F
E E F F d
γ + γ ·
γ + γ + γ + γ ·
γ + γ ·
r r
(C.9)
3. Establish time history of the generalised load effect
( ) , , p ), t ( T ), t ( M g ) t ( g
k ed d
Λ ∆ · R
(C.10)
602 SLS and ALS LRFD capacity checks can be based
directly on time series for resulting moment and effective
tension given as output from the global analyses. Consistent
treatment of correlation requires that steps 2) and 3) in the
post processing procedure discussed in the previous section is
considered.
C 700 Computer implementation
701 The key to efficient LRFD capacity checks for
combined loading is a computer implementation of the
procedures described in the previous sections. The main
technical features needed in to perform capacity checks in a
stationary design condition can be summarised as:
÷ Separation of global load effects into E and F
components
÷ Generate time series of the generalised load effect
÷ Processing results from regular/irregular dynamic
analysis
÷ Analyse several E,F safety factor combinations
÷ Evaluate utilisation by nonGaussian extreme value
statistics
÷ Evaluate statistical confidence in extremes
÷ Evaluate contribution from P,F,E loads
÷ Efficient communication with FE global analysis
program
÷ Graphical presentation of results as a function of location
along the riser.
Guidance note:
The following combinations of partial coefficients need to be
checked for LRFD ULS conditions
f
γ
e
γ
1.1 1.3
1.1 0.77
0.91 1.3
 end  of  Guidance  note 
702 A computer program with the described functionality
is capable of performing all relevant capacity checks for
combined loading automatically with a minimum of input
from the analyst. Application examples are presented by
Sødahl et al (2000)
D. Implementation of the WSD design
format
D 100 General
101 Practical implementation of the WSD design format
for combined loading is simpler when compared to the LRFD
ULS design checks because no separation of the load effect
into F and E components is required. Implementation of the
WSD design format is hence similar to the LRFD SLS and
ALS design checks as discussed in section C 602. A brief
introduction to implementation of WSD design checks is
however given in the following for completeness.
D 200 Implementation in design analyses
201 The generalised load effect for the WSD design
format for combined loading can be expressed as:
) , , p ), t ( T ), t ( M ( g ) t ( g
k e
η ∆ · R (C.11)
Where M, T
e
, p ∆ and η denote bending moment, effective
tension, local differential pressure and usage factor,
respectively. The generalised load effect can hence be
computed directly from the effective tension and bending
moment components given as output from the global
analyses. The resulting bending moment is computed as:
) t ( M ) t ( M ) t ( M
2
y
2
x
+ ·
(D.12)
202 Evaluation of acceptance criteria based on the
generalised load effect is identical as outlined in sections C
300 and C 400 for the LRFD approach
E. Shortterm extreme load effect
estimation
E 100 General
101 For a design storm approach, the extreme load effect
can be estimated as the expected or most probable largest
response peak for the specified duration of the design
DNVOSF201 Dynamic Risers, October 2010
Page 78 Appendix C
DET NORSKE VERITAS
condition. This approach applies to SLS, ULS, and ALS
design conditions.
102 Output from irregular time domain analyses using the
analysis/post processing procedure as described in section B
are time traces describing one realisation of the (generalised)
load effect. The probabilistic distribution of the load effect
process is in general is nonGaussian. Furthermore, the
duration of the simulated time record will in many situations
be shorter than the specified duration of the design condition
due to practical limitations related calculation time on the
computer. Thus, extrapolation is often involved in practical
estimation of characteristic extreme load effect. The process
of obtaining extreme load effect estimates from time domain
analyses will typically involve the following issues :
÷ Envelope statistics.
÷ Estimation of extreme values from nonGaussian load
effect time series
÷ Estimation of simulation length required to obtain
extreme load effect estimates with sufficient statistical
confidence
E 200 Envelope statistics
201 Response envelope is defined as extreme response
values (minimum and/or maximum) attained during the time
domain simulation as a function of location along the
structure. This concept is very useful to establish design
values in case of deterministic loading (e.g.. regular wave
loading).
202 A more careful interpretation is however needed for
application of the envelope concept in for application in a
design storm approach. Envelopes from irregular time
domain analyses will represent realisations of the extreme
load effect for the duration considered in the time domain
simulation. For prediction of characteristic extreme response
it is hence required that the simulation time must be identical
to the duration of the design condition (e.g. 3 – 6 hours).
Furthermore, the extremes predicted in this way will have
low statistical confidence as they only represent the extreme
load effect found for one realisation.
203 Improved statistical confidence can be achieved by
considering the average envelope found by averaging over
several realisations. The average envelope will hence
represent expected extreme load effect along the structure,
which is the wanted output from shortterm global response
analysis. The statistical uncertainty can be expressed in terms
of the standard deviation of the estimated expected extreme
value,
T
σ :
R
E
T
N
σ
· σ
(C.13)
where N
R
is the number of realisations and σ
E
is the standard
deviation of the extreme response estimated from all
realisations. This “brut force” approach will yield unbiased
extreme response estimates at any location along the
structure, but is in most situations too time consuming to be
of practical use. It can however be applied for simple systems
and for verification of more sophisticated statistical methods
for prediction of extreme response based on one realisation.
E 300 Extreme response estimation
301 The main steps involved in statistical processing of
stochastic load effect time histories to produce characteristic
extreme load effect can be summarised as:
÷ Select probabilistic distribution model (i.e. parametric
probabilistic model for individual response peaks or
extreme peak for a given duration)
÷ Estimate parameters in the selected model based on the
available response time history realisation
÷ Accept/reject the selected model (e.g. by use of
engineering judgement or formal statistical hypothesis
test)
÷ Compute estimate of characteristic extreme load effect
based on the fitted model (i.e. percentile in fitted peak
distribution or expected extreme peak value)
÷ Quantify statistical uncertainty of the estimated
characteristic extreme load effect.
302 The major challenge is often related to selection of an
adequate probabilistic distribution model for the individual
peaks of the load effect process. Special attention must be
placed on description of the upper tail of the distribution,
which is of vital importance for estimation of extreme values.
The choice of distribution model is complicated by the fact
that the nonGaussian response characteristic in general is
strongly system and excitation dependent. A significant
variation of the nonGaussian response characteristics must in
addition be foreseen along the riser. The choice of a proper
distribution model will hence depend on the riser system,
excitation level as well as location along riser. In most
practical applications, the choice of probabilistic model is (at
least to some extent) empirical, based on previous experience
and physical knowledge of the dynamic behaviour of the
actual riser system of concern (see also Appendix A for a
discussion of governing nonlinearities). Simple parametric
models (e.g. Normal, Rayleigh, Weibull, Exponential) are
frequently applied.
303 The selected parametric model is fitted to the
simulated peak sample using an appropriate statistical
estimation technique (e.g. method of moments, probability
weighted moments, maximum likelihood, regression, etc),
For a more detailed discussion, reference is made to
statistical textbooks, e.g. Bury (1975). A problem often
encountered in practical applications is that fitted parametric
model fails to describe the ‘true’ upper tail behaviour
resulting in biased extreme value prediction. Special
estimation techniques (tail fitting techniques) have been
designed to improve the fit in the upper tail region at the
expense of a somewhat increased statistical uncertainty
(Sødahl and Larsen 1992).
304 Mathematical arguments in terms of limiting
asymptotic distributions can in addition be applied to
establish models for extreme peaks within a specified time
window (e.g. Gumbel extreme value distribution, see Gumbel
DNVOSF201 Dynamic Risers, October 2010
Appendix C Page 79
DET NORSKE VERITAS
1958) and for peak excesses over high thresholds (e.g.
generalised Pareto distribution, see Davison and Smith
1995).
E 400 Statistical uncertainty and simulation
planning
401 A fundamental problem related to estimation of
characteristic extreme load effect is that statistical
uncertainties are introduced because estimates are based on
simulated time series realisations of finite lengths. Different
realisations will consequently give different estimates of the
extreme load effect. The estimation variability can be
expressed in terms of the probabilistic distribution of the
applied estimator, commonly denoted the sampling
distribution. The sampling distribution can hence be applied
to express the confidence of the estimated characteristic
extreme response as a function of simulation length for each
particular estimator of concern. This information can be
applied directly in practical planning of computer simulations
to estimate the simulation length needed to give estimates of
characteristic extreme response with a specified confidence
(Sødahl and Larsen 1992). The sampling distribution will
also be the basis for selecting the most efficient estimator
among several possible candidates.
402 The exact sampling distribution is in general very
difficult to establish for finite samples and is in practical
calculations normally approximated by the Gaussian
distribution. This assumption is justified by theoretical results
showing that the sampling distribution of most estimators of
practical interest will approach the Gaussian distribution
asymptotically as a function of sample size (see e.g. Cramer
(1971) for moment based estimators). The sampling
distribution can consequently be completely described by the
mean value and variance of the estimator. Approximate
techniques (e.g. asymptotic expressions) or numerical
simulation techniques (e.g. bootstrap estimation) assuming a
sample of independent stochastic variables is normally used
to establish the variance. The independence assumption is
normally an acceptable approximation for the peak sample.
403 For moment based estimators (i.e. estimators that can
be expressed as a function of sample moments) the following
relation between simulation length t
S
and standard deviation
of the estimator σ
T
can be established by asymptotic
approximations:
S
T
t
c
· σ
(C.14)
where c is a (unknown) constant.
404 The following procedure can be applied for practical
planning of simulations to obtain a target confidence
specified in terms of the standard deviation
c
T
σ :
÷ Perform time domain analysis with initial duration
i
S
t
÷ Estimate extreme response and associated standard
deviation of the estimate
i
T
σ based on the initial
duration
i
S
t
÷ If target confidence is not obtained (i.e.
i
T
σ >
c
T
σ ) the
increased simulation length
c
S
t needed to fulfil the
confidence requirement can be estimated as:
2
c
T
i
T i
S
c
S
t t
,
`
.

σ
σ
≥
(C.15)
An important consequence of this equation is that an increase
of the simulation length with a factor of 4 is required to
reduce the standard deviation with a factor of 2.
F. Longterm load effect statistics
F 100 General
101 The longterm load effect distribution is a result of the
combined wind, wave and current action on the coupled
floater/slender structure system i.e. a probabilistic description
of the response from the longterm environmental action. The
longterm environmental loading process can be divided into
time intervals with stationary conditions, denoted shortterm
conditions. It is further assumed that each shortterm
condition can be completely described by a limited number of
environmental parameters (Waves will for example typically
be described by significant wave height, peak period,
spreading, mean direction etc). The longterm response
distribution can hence formally be expressed as:
m m m M
M
M
M
d ) ( f )  x ( F ) ( w ) x ( F
 X X ∫
·
(C.16)
where :
F
X
(x) Longterm distribution of load effect peaks
M Vector of parameters describing shortterm
environmental conditions
w(M) Weight function accounting for variation in
load effect mean level crossing frequency
F
XM
(x m) Short term distribution of load effect peaks for
a stationary environmental condition (i.e.
conditional on M)
f
M
(m) Distribution of environmental parameters
102 The main challenge related to this approach is to
establish the shortterm load effect distribution F
XM
as
nonlinear irregular time domain analysis in general will be
required to give an adequate description of the response
process.
103 Discrete approximations to this general formulation
form the basis for approximate techniques for assessment of
the longterm load effect distribution in practical
applications. These methods have in common that
simplifications are introduced in the longterm load effect
description to enable practical computations. Simplifications
are typically based on rational conservative assumptions
regarding system behaviour with respect to e.g.
DNVOSF201 Dynamic Risers, October 2010
Page 80 Appendix C
DET NORSKE VERITAS
environmental directionality, wave/current correlation, floater
position, operation of the system etc.
Guidance note:
The relative importance of waves, current and floater motions to
the response of deepwater riser systems is strongly system
specific. Floater type, station keeping system, riser configuration
and boundary conditions will determine how the external
loading is transformed into deformations and internal reaction
forces in the riser. A significant variation in response
characteristics along the riser must also be anticipated. Waves
and floater motions will always be crucial for the response in the
upper part of the riser. The situation is more complex in lower
parts of deepwater riser systems. Floater offset and current are
expected to be governing for the global response in lower parts
of tensioned risers. Wave induced floater motions will normally
be of some importance all along compliant riser configurations.
Possible simplifications and conservative assumptions
introduced to ease the design process of deepwater risers must
therefore always be evaluated very carefully for each riser
concept of concern. As an example, the commonly applied
discrete formulation for environmental statistics described in
terms of a H
s
T
p
wave scatter diagram can be expressed as:
) T , H ( P ) T , H  x ( F ) T , H ( w ) x ( F
i p s i p
N
1 i
s
s
X i p s X ∑
·
·
(C.17)
Where:
N Number of discrete sea states in the wave
scatter diagram
w(H
s
,T
p
)
i
Weight factors accounting for variation in
level crossing frequency
P(H
s
,T
p
)
i
Sea state probability
F
X
(x) Longterm distribution of load effect maxima
p s
s
X
T H x F ,  (
Short term distribution of load effect maxima
 end  of  Guidance  note 
F 200 Response surface approach
201 The response surface approach can be thought of as a
direct numerical approximation to the general formulation as
defined by Eq. (C.13). The computational efforts involved in
such techniques are establishment F
XM
by global load effect
analyses in a limited number of carefully selected stationary
environmental conditions considering irregular wave
excitation. Interpolation/extrapolation techniques can then
subsequently be applied to establish F
XM
for all relevant
environmental conditions required for assessment of the long
term load effect distribution. See e.g. Farnes & Moan (1993)
for a flexible riser application example and Sødahl et al
(2000) for a SCR application example.
202 The response surface approach can be formulated in
the following steps:
1. Select a limited number of basic ‘representative’
seastates (i.e. combined environmental conditions).
These seastates will serve as ‘interpolation points’ and
should hence be selected very carefully
2. Perform global response analysis for the basic seastates
considering irregular time domain analysis.
3. Establish probabilistic models for the shortterm load
effect distributions for all basic seastates. Fitting a
parametric model (e.g. Weibull distribution) to the
simulated sample of load effect peaks is a typical
procedure.
4. Establish shortterm distributions for all relevant
seastates by interpolation/extrapolation techniques using
results obtained by analysis of the basic seastates as
interpolation points.
5. Establish longterm response distribution by use of the
discrete approximation to the general formulation
defined in F 101.
203 The response surface will hence enable computation
of the longterm load effect distribution considering a
possible nonGaussian shortterm load effect characteristics.
For practical application, it is however crucial that acceptable
precision can be obtained by use of relatively few basic
seastates (e.g. 5 or less).
G. References
G 100 Standards, Guidelines and Handbooks
API RP 2SK “ Recommended Practice for Design and
Analysis of Station Keeping Systems for Floating structures”
Second Edition, December 1997.
G 200 Technical references
Bury K V (1975) “ Statistical Models in Applied Science”,
John Wiley, 1975.
Cramer H (1971):” Mathematical Methods of Statistics”
Princeton University Press, 1971.
Davison A C and Smith R L (1990): “ Models for
Exceedances Over High Thresholds” Journal of the Royal
Statistical Society Series B, 52, pp 393442.
Farnes K. A. and Moan T. (1993): ” Extreme Response of a
Flexible Riser System using a Complete LongTerm
Approach”,Proc. ISOPE,1993.
Gumbel E J (1958) “ Statistics of Extremes”, Columbia
University Press , 1958..
Larsen C M and Olufsen A (1992): ” Extreme Response
Estimation of Flexible Risers by Use of Long Term
Statistics” Proc. ISOPE 1992..
Sødahl N, Larsen C M (1992): “Methods for Estimation of
Extreme Response of Flexible Risers” Proc. ISOPE 1992.
Sødahl N, Mørk K, Kirkemo F, Leira B, Igland R (2000):
“Design and Analysis of Metallic Risers. Global analysis
Procedures” OMAE 2000
DNVOSF201 Dynamic Risers, October 2010
Appendix D Page 81
DET NORSKE VERITAS
APPENDIX D VERIFICATION OF GLOBAL ANALYSIS MODEL
Contents
A. General
A 100 Objective
A 200 Introduction
B. Verification of theoretical models
C. Verification of numerical procedures
C 200 Spatial discretisation
C 300 Frequency discretisation
C 400 Time discretisation
D. References
A. General
A 100 Objective
101 The purpose of this Appendix is to give an
introduction to principles for verification of the computer
model applied in global static and dynamic finite element
analysis, ref. Section 4.
A 200 Introduction
201 The computer model of a riser system represents two
fundamentally different types of approximations to the
physical system:
÷ Theoretical models;
÷ Numerical approximations;
202 The theoretical models represent the fundamental
assumptions in terms of idealised models for the physical
system. Examples of theoretical idealisations are
environmental models (e.g. wave spectrum, Airy wave
kinematics etc), load models (e.g. Morison equation, soil
model etc) and models for structural behaviour (e.g. global
crosssectional models, Rayleigh structural damping,
solution strategy etc.).
203 Furthermore, numerical approximations of the
theoretical models are needed to facilitate computer
solution. The numerical approximation will typically
involve spatial discretisation of the structure into a finite
number of elements as well as time and/or frequency
discretisation of the dynamic loading.
204 Hence, the key issue involved in verification of the
computer model is to ensure that the theoretical models and
numerical approximations represent the real physical
behaviour of the riser system. As discussed in Appendix A
C 200 the required accuracy is closely linked to the purpose
of the analyses (e.g. feasibility studies, early design, detailed
design, and final verification)
B. Verification of theoretical models
101 Global analyses should in general be performed with
welldocumented and verified computer codes for analysis
of slender structures. Furthermore, accumulated experience
expressed in terms of recommended practice for modelling
and analysis should always be consulted.
102 However, it is crucial to have a basic physical
understanding of the applicability and limitations in
commonly used theoretical models. This is of particular
importance for a critical assessment of modelling and
analysis of new concepts and to ensure that adequate results
are obtained when simplified modelling and analysis
strategies are applied.
103 Any use of simplified analysis strategies will in
general require benchmark validation by comparison to
more advanced analysis procedures. Examples of typical
situations are given in the following:
÷ Dynamic analyses should be considered to verify
quasistatic assumptions;
÷ Linearized time domain analyses should be validated by
nonlinear time domain analyses;
÷ Frequency domain analyses should be validated against
time domain analyses;
÷ Verification of combined use of global and local
quasistatic response models by comparison to a
complete response model (e.g. quasistatic model for
bend stiffener response);
÷ Floater/slender structure coupling effects should always
be assessed by coupled analysis and/or model tests for
deepwater mooring systems. This is of particular
importance for turret moored ships at deep water
locations;
÷ Decoupled floater motion analysis should be supported
by coupled floater motion analysis when significant
coupling effects are identified;
÷ Effects from simplified treatment of LF floater motions
in terms of an additional offset should be evaluated for
deep water concepts. Statistical correlation as well as
effects from LF response on WF response (e.g. LF
variation of effective tension) should be addressed.
Such studies should at least be carried out for new deep
water concepts;
÷ Regular wave analyses should always be verified by
irregular analyses. This is in particular important for
systems that may be subjected to resonance dynamics;
÷ Many riser concepts are sensitive to wave loading in the
splash zone. The effect of disturbed kinematics due to
the presence of the floater should be carefully
evaluated. Simplified modelling in terms of adjustments
of hydrodynamic coefficients must be evaluated by
more advanced techniques considering transfer
functions for wave kinematics consistent with the
floater motions;
÷ Any structural modelling simplifications to gain
computational efficiency should be validated against a
DNVOSF201 Dynamic Risers, October 2010
Page 82 Appendix D
DET NORSKE VERITAS
more comprehensive structural model (e.g. omission of
bending stiffness, simplified modelling of components,
use of average of crosssectional properties, simplified
modelling of boundary conditions etc);
104 Analytical verification should be performed
whenever possible to verify modelling and input parameters.
Examples of simple analytical checks are given in the
following:
÷ Verification of static effective tension distribution of
top tensioned risers. The effective tension distribution
of top tensioned risers can be found by accumulation of
effective weight along the riser. This check represents a
verification of the mass (pipe, components, internal
fluid etc), buoyancy modelling (pipe, additional
buoyancy components etc) and tensioner modelling of
the system;
÷ The static configuration of single line compliant riser
configurations can be verified by use of catenary
equations disregarding the effect of bending stiffness.
The catenary configuration solution will in most
situations represent a close approximation because the
effect of bending stiffness to the overall static
configuration normally is negligible. Simple
equilibrium iteration is however required in obtaining
the static configuration (e.g. using the socalled
‘shooting’ approach). The primary purpose of this
check is to verify mass and buoyancy modelling, but it
will also give a verification of the shape of the static
configuration;
÷ Eigenmodes of top tensioned risers can be verified by
analytical calculations. Approximate solutions are given
in terms of closed form expressions for tensioned beams
and cables with uniform crosssectional properties.
105 It has been experienced that surprisingly many
modelling mistakes can be traced back to a few common
problem areas. Two important examples are discussed
below:
÷ Input of floater motion transfer functions in terms of
amplitude and phase angle (or alternatively on complex
form) as function of wave frequency and direction
related to a local floater coordinate system. Definition
of amplitude, phase angle, wave direction and floater
coordinate system differ from program to program.
Conversion between different definitions is usually
required to apply output from hydrodynamic floater
motion analysis (e.g. diffraction/radiation approach) as
input in global riser analyses. Such operations should be
performed very carefully with emphasis on thorough
verification. Floater terminal point motion (i.e. motion
of a point on the floater at some distance from origin of
vessel coordinate system) generated in global riser
analysis should in particular be verified by analytical
calculations for different wave directions and floater
directions. Animation showing floater motions, waves
and riser deflections is a very useful tool for verification
of floater motions;
÷ Buoyancy can be treated in terms of effective tension as
discussed in Appendix A or alternatively by integration
of the hydrostatic pressure acting on the outer riser
surface. Both formulations are correct and will hence
give the same riser response when applied correctly.
The latter formulation will however require a very
careful modelling of the exposed outer area for complex
riser systems with variable outer diameter (e.g. Spar
risers, systems with attached buoyancy elements etc). It
is therefore recommended that use of computer
programs based on pressure integration for
representation of hydrostatic pressure should be
validated against other codes using the effective tension
formulation.
106 Independent analyses of selected critical conditions
are in addition highly recommended as a part of the design
process of riser systems. The independent analyses should in
principle always be carried out using a different recognised
computer program. Furthermore, it is crucial to utilise
information from model tests as well as fullscale
measurements whenever possible for validation, calibration
and enhancement of computer analysis of riser systems.
107 Sensitivity studies are also recommended to
investigate the influence from uncertain system parameters
(e.g. equivalent multipipe model, hydrodynamic coefficients
in moonpool, soil data etc). The main purpose should be to
quantify model uncertainties, support rational conservative
assumptions and identify areas where a more thorough
investigation is needed to achieve an acceptable modelling
(e.g. calibration against model test).
C. Verification of numerical
procedures
101 Numerical approximations will typically involve
spatial discretisation of the structure into a finite number of
elements as well as time and/or frequency discretisation of
the dynamic loading. Investigation of convergence in the
solution by repeated analyses considering successive
refinement of the discretisation is the basic principle to
verify that the discretisation is adequate. The discretisation
is considered adequate when the change in response
between two successive discretisation is acceptable seen in
relation to the purpose of the analyses. In this situation,
there will be no practical gain by further refinement of the
discretisation.
C 200 Spatial discretisation
201 Repeated static and dynamic analyses considering
successive refinement of the element mesh can be applied to
assess the adequacy of the spatial discretisation. Special
attention should be given to the following parts of the riser
system:
÷ Areas with high curvature (e.g. hog and sag bend);
÷ Contact areas (touch down, hull supports);
÷ Terminations to fixed structures;
÷ Areas with high load intensities (e.g. splash zone);
÷ Areas with significant change in crosssectional
properties (e.g. taperjoint, bend stiffener etc);
DNVOSF201 Dynamic Risers, October 2010
Appendix D Page 83
DET NORSKE VERITAS
÷ Areas with change in element lengths. The relative
change in length between adjacent elements with
uniform crosssectional properties should not exceed
1:2. A lower relative change may be required in case of
nonuniform crosssectional properties;
202 The convergence should be assessed for all relevant
response quantities. This is because the rate of convergence
normally will be different for different response quantities
(e.g. slower convergence is normally observed for shear
forces and bending moments when compared to effective
tension).
203 The convergence study must be performed for the
actual element used in the analyses. This is because e.g.
beam elements based on conventional displacement
formulation may display a significantly different numerical
behaviour when compared to hybrid elements used in a
mixed formulation. Furthermore, static as well as dynamic
analyses should be considered in the evaluation studies.
C 300 Frequency discretisation
301 Floater motion transfer functions are represented in
terms of amplitude and phase angle as function of a number
of discrete wave frequencies and directions. The discrete
frequencies and directions must be selected carefully to
obtain an adequate description of the floater motions:
÷ The frequencies should be selected to cover the
resonance peaks in vessel motion transfer functions
(e.g. heave, roll and pitch resonance frequencies);
÷ Possible cancellation frequencies should be identified
and covered by the discrete representation. (relevant for
e.g. semisubmersibles and TLP’s);
÷ The frequency range should cover relevant frequencies
in the wave excitation. It should also be clarified how
the actual computer program handles possible excitation
outside the frequency range of the floater motion
transfer function (this is a well known source to
erroneous excitation);
÷ Discretisation of wave direction with a spacing in the
range of 1530 deg. is normally sufficient to give a
good representation of the floater motions;
302 Results from frequency domain analysis are given in
terms of auto and crossspectral densities at a number of
discrete frequencies. The frequency spacing will hence be
decisive for the variance and covariance found by
integration of the corresponding response spectra. The
adequacy of the frequency discretisation can be assessed by
repeated analysis considering successive denser frequency
spacing.
C 400 Time discretisation
401 Numerical time integration is applied in time domain
analyses to produce discrete response timeseries.
Unconditional stable, single step integration procedures such
as Newmark β and HilberHughes α methods are
frequently applied. The latter approach is normally preferred
in variable time step algorithms due to explicit control of
numerical damping to suppress possible high frequency
noise introduced by change of time step. Choice of time step
is crucial for the stability and accuracy of direct time
integration methods, some aspects are discussed in the
following:
÷ The time step required to obtain a stable numerical
solution is to a large extent governed by the highest
eigenmode present in the discrete structural model. This
is because all eigenmodes need to be integrated
accurately to obtain a stable solution (i.e. also modes
that are of no significance for the response description)
Typical time step is in the range of 0.1 0.4s for
numerically wellbehaved systems;
÷ Nonlinear analyses will in general require a shorter time
step to obtain a stable numerical solution when
compared to linearized analyses. This is in particular
the case for numerical sensitive systems, e.g. systems
with significant displacement dependant nonlinearities
such as low tension problems including snap loading,
instability problems, contact problems and significant
nonlinear material behaviour (e.g. momentcurvature
hysteresis);
÷ Variable time step integration methods may introduce
high frequency noise when applied to numerically
sensitive systems. It is therefore recommended to apply
constant time step algorithms when analysing
numerically sensitive systems. Use of variable time
stepping procedures should at least be validated against
constant time step algorithms when unphysical noise is
detected in response time series;
÷ Quality checks of response time histories should always
be considered to identify possible unphysical noise
reflecting an inaccurate numerical solution. The
overview statistics discussed in Appendix A is a very
useful tool for detection of possible unphysical response
peaks. Identified suspicious locations along the riser
should be subjected to closer examination by spectral
and statistical analyses as well as visual inspection of
the response time histories;
÷ Study of convergence considering successive
refinements of the time discretisation is a useful
exercise to determine the required time step to obtain an
adequate numerical solution.
402 Time domain analyses considering stochastic wave
loading will typically require generation of discrete time
histories for floater motions and wave kinematics according
to a specified wave spectrum. The load time histories are
represented in terms of a finite number of harmonic
components. The amplitude of each harmonic component is
normally computed from the specified spectral
representation of the load process, while the phase angle is
assumed to follow a uniform probabilistic distribution over
the interval (02π). Important aspects regarding load
discretisation is discussed in the following:
403 The generation of load time histories can be carried
out very efficiently by use of the FFT (Fast Fourier
Transform) technique using equidistant frequency
representation of the load process. The main advantage of
this approach is that almost no additional cost is related to
use of many frequencies to describe the load processes. This
DNVOSF201 Dynamic Risers, October 2010
Page 84 Appendix D
DET NORSKE VERITAS
is of particular importance to describe the relevant
frequency content of vessel motion transfer function and
wave spectrum as well as the response process in case of
resonance dynamics. The repetition period of the generated
load time history is also uniquely determined by the
frequency spacing of the harmonic components (see
Appendix A) The main drawback is however that time series
must be generated prior to the simulation at fixed locations
along the riser. Interpolation in time and space is hence
necessary during the simulation. The spatial interpolation
should in particular be considered carefully to obtain an
adequate representation of the loading close to sea surface.
Variable spacing of interpolation points (i.e. points where
load time histories are pregenerated) along the riser is
normally considered to obtain efficient analyses. Benchmark
validation by successive increase of number of interpolation
points is recommended to verify the spatial interpolation. A
time step in the range of 0.251s is typically sufficient to
facilitate adequate time interpolation of WF excitation.
404 Direct accumulation of harmonic components
representing floater motions and wave kinematics can
alternatively be performed during the simulation to
overcome the interpolation problem related to the FFT
approach. The main advantage is that wave kinematics can
be calculated at instantaneous spatial position allowing for
consistent representation of wave kinematics in case of large
riser displacements (e.g. combined LF and WF floater
motions). This approach is however far more time
consuming than the FFT approach and will only be
applicable when relatively few frequencies are considered
for representation of the load processes (typically 100200).
These frequencies must hence be selected very carefully to
give an adequate representation of the loading (e.g.
resonance peaks in the vessel motion transfer function and
peak period in wave spectrum). Furthermore, use of variable
frequency spacing is required to cover the relevant
frequency range with as few harmonic components as
possible. Several strategies have been proposed, see e.g.
Garrett et al (1995) and McNamara and Lane (1984).
Benchmark validation by successive increase of number of
frequencies is recommended.
405 An additional practical problem related to use of
variable frequency spacing is that it is more complicated to
assess the repetition period of the generated time histories.
Approximate closed form expressions are available for some
algorithms. Judgements based on the autocorrelation
function estimated from the generated realisation can
alternatively be applied to assess the repetition period
(Garrett et al 1995).
406 The quality of the generated floater motions and
wave kinematics depends on the ability of random number
generator to produce statistically independent phase angles.
The numerical behaviour of the random generator may
depend on the actual computer used in the analyses. Quality
checks of generated wave realisations are recommended in
connection with new computer installations to ensure that
the generated realisations are Gaussian. Statistical properties
of the process and individual peaks should be considered for
several realisations with rather long duration (e.g. 36
hours).
D. References
Garrett D L, Gu G Z, Watters A J (1995) “ Frequency
Content Selection for Dynamic Analysis of Marine System”
OMAE 1995.
McNamara, J F, Lane M (1984) “Practical Modelling of
Articulated Risers and Loading Columns” journal of Energy
Resources Technology, Vol. 106.
DNVOSF201 Dynamic Risers, October 2010
Appendix E Page 85
DET NORSKE VERITAS
APPENDIX E VIV ANALYSIS GUIDANCE
Contents
A. General
A 100 Objective
B. Fatigue Assessment
B 100 Simplified Assessment of Fatigue Damage
B 200 Multimodal Response Analysis Based on
Empirical Models
B 300 Methods Based on Solution of the NavierStokes
equations
C. Methods for reduction of VIV
C 200 Modification of Riser Properties
C 300 Vortex suppression devices
D. References
A. General
A 100 Objective
101 This Appendix proposes a fourstep method for
assessment of VortexInduced riser response amplitudes and
corresponding fatigue damage. These steps of increasing
complexity is defined as follows:
÷ Simplified assessment of fatigue damage;
÷ Multimodal response analysis based on empirical
hydrodynamic coefficients (and tests);
÷ Computational Fluid Dynamics solving the Navier
Stokes equations;.
÷ Laboratory test.
102 The fundamental principle is that for cases where
VortexInduced Vibrations (VIV) are likely to represent
design problem, refined assessment methods preferably
supplemented with tests are required.
Guidance note:
Often, the main design focus is to evaluate if the fatigue
capacity is sufficient. Accordingly, a simplified (i.e.
conservative) VIV analysis will suffice if the resulting fatigue
damage is within the tolerated limit. If the simplified analysis
indicates insufficient fatigue capacity, more sophisticated
should be applied. The method should be chosen according to
the specific case investigated.
 end  of  Guidance  note 
B. Fatigue Assessment
B 100 Simplified Assessment of Fatigue Damage
101 A simplified estimate of the induced fatigue damage
can be computed by neglecting the influence of the waves,
assuming undisturbed current velocities to apply. The
following procedure can subsequently be applied (B 102 B
106).
102 Identify the planes of vibration for the relevant mode
shapes in relation to the specified current directions.
Guidance note:
For rotationally symmetric riser systems, the crossflow
vibration will generally be perpendicular to the current
direction. For nonsymmetric systems, the crossflow vibration
is assumed to occur in the plane of the relevant modeshapes
 end  of  Guidance  note 
103 Identify dominant mode shapes and natural
frequencies as follows:
a) Determine the natural frequencies and mode shapes for
bending in the crossflow direction based on analytical
models or by numerical FEM analysis.
b) Define a band of local vortex shedding frequencies f
s
along the riser using:
D
U
S f
t s
·
(E.1)
Where U is the local tangential flow velocity and D is
the outer riser diameter. S
t
is the Strouhal number where
upper and lower bound values should be checked
(Typically S
t
= 0.14 to 0.25)
c) For each mode, check for which parts of the riser the
natural frequency for the mode is within the limits of
the local shedding frequency.
d) Identify the most likely mode shapes to be excited by
VIV and select the one with the highest curvature for a
unit modal amplitude. Typically, this will be the mode
with the highest frequency among the “probable
modes”
104 For a given flow velocity compute the vibration
amplitude for the anticipated mode according to Sarpkaya
(1979):
) K S 2 ( 06 . 0
32 . 0
D
A
S
2
t
⋅ ⋅ π +
·
γ
(E.2)
Where K
s
is the stability parameter, and γ is the mode
participation factor, see e.g. Blevins (1990).
105 Compute corresponding stress range:
S = A SCF⋅ E ⋅ κ⋅ (Dt) (E.3)
where E is the modulus of elasticity and SCF is a stress
concentration factor . κ is the curvature of the mode
shape φ(s) at the point (s, φ(s)) to be calculated as:
DNVOSF201 Dynamic Risers, October 2010
Page 86 Appendix E
DET NORSKE VERITAS
2 / 3
2
2
2
1 ) (
−
,
`
.

,
`
.

∂
φ ∂
+
∂
φ ∂
· κ
s s
x
(E.4)
Guidance note:
If a finite element model is applied, the stresses corresponding
to unit mode shape amplitude is first computed based on the
stiffness matrix for the relevant element. The resulting stress
range is subsequently obtained by multiplication with 2 A SCF.
 end  of  Guidance  note 
106 The fatigue damage is estimated by application of
the relevant SNcurve as:
m L n
F
S
a
T f
D ⋅
⋅
·
(E.5)
where f
n
is the frequency of the relevant mode, T
1
is the
design life of the riser, S is stress range and m and a are
constants defining the SNcurve, see Appendix B.
For screening purposes a 1year velocity with associated
velocity profile is considered conservative. Otherwise, a
weighted summation of computed damage over the long
term current distribution for velocities and direction must be
performed.
B 200 Multimodal Response Analysis Based on
Empirical Models
201 If significant VIV induced fatigue damage is likely,
more thorough calculations should be conducted. The next
level of refinement is typically methods for multimodal
response analysis based on empirical or semiempirical
values of the hydrodynamic coefficients. One way of
achieving this is by application of a generalisation of the
procedure described above. There are also two other main
approaches for calculating the response:
÷ Calculate modal response in the frequency domain. This
approach can incorporate general current profiles. A
correlation function for the loading process at two
points along the riser is introduced. Subsequently, a
double integration is performed. The parameters
entering the calculation of load and response generally
requires calibration with model field data.
÷ Calculate response in the time domain. Here one must
have a considerable database of crosssection tests
giving force coefficients, frequencies and phase angles
for various combination of incident velocity and cross
section vibration. In the end such a simulation would
hopefully stabilise or maybe repeat, and a response
spectrum can be retrieved.
Guidance note:
For vertical risers in wellknown environmental conditions
recognised semiempirical programs may be applied, see e.g.
SHEAR7 (Vandiver & Li, 1996). Larsen and Halse (1995)
conducted a comparison between programs showing
considerable discrepancies concluding that at present no
generally accepted program exist for calculation of VIV
response. The excitation is directly dependent on the response.
Jumps from one mode to another may happen. Proper force
coefficients for vibrations including more than one frequency
are still lacking.
 end  of  Guidance  note 
B 300 Methods Based on Solution of the Navier
Stokes equations
301 The analysis based on solution of the full Navier
Stokes equations implies a set of twodimensional fluidflow
analysis for sufficiently many crosssections along the riser,
also including modelling of the dynamic boundary
conditions. The direct solution of the complete flow
equation is until now restricted to low Reynolds numbers
(no turbulence in the near wake). However, for marine risers
the wake will be turbulent, requiring very small timesteps
or a good turbulence model. Even if this approach at present
stage is very time consuming and possibly not correctly
modelled for high Re, it is likely that this will be a feasible
approach in the future.
302 For compound multipipe riser geometries, the
computations generally become increasingly complex and
time consuming. Validation of the numerical results by
sensitivity studies with respect to key parameters should
accordingly be performed. Comparison with results obtained
from fullscale or model experiments is also essential for
calibration and finetuning of the numerical algorithms.
C. Methods for reduction of VIV
101 If the calculated VIVresponse is a problem, there are
two main approaches:
÷ Modify the properties of the riser, i.e., tension,
diameter, structural damping.
÷ Introduce vortex suppression devices.
C 200 Modification of Riser Properties
201 There are several different ways of reducing the
amplitude of vortex induced vibration. It is usually possible
to avoid the resonant crossflow region when the highest
reduced velocity is below 3, i.e. below the resonant region.
To be well above the resonant area is much more
complicated. There will always be a higher natural mode
with a frequency that corresponds to f
s
. However, according
to Vandiver (1993), the presence of shear flow in the region
of the higher modes greatly reduces the probability for lock
in.
202 A different approach is to increase the reduced
damping. Blevins (1990) states that a reduced damping
greater than 64 reduces the peak amplitudes to less than 1 %
of the diameter. In marine applications, the reduced
damping is usually lower than one and it is very seldom
possible to increase the damping to such an extent.
C 300 Vortex suppression devices
301 A second possibility is to add vortex suppression
devices to the cylinder. Zdravkovich (1981) classifies the
DNVOSF201 Dynamic Risers, October 2010
Appendix E Page 87
DET NORSKE VERITAS
means of suppression to three categories according to the
way it influence the vortex shedding:
÷ surface protrusions (wires, helical strakes etc.)
triggering separation;
÷ perforated shrouds, axial slats etc. (breaking the flow
into many small vortices); and
÷ near wake stabilisers, preventing the building of the
vortex street.
In Blevins (1990), eight different devices are shown, and
comments on their use and effects are given. Common for
all (except the ribboned cable) is that they increase the cost
of the riser, and that they will complicate handling during
installation. Some of the devices also reduce the drag
coefficient, especially the streamlined fairing. However, in
most cases the inline drag coefficient is increased rather
than being reduced by introducing vortex suppression
devices.
D. References
Blevins, R. D. (1990). FlowInduced Vibration (Second ed.).
New York, USA: Van Nostrand Reinhold.
Sarpkaya, T. (1979). Vortexinduced oscillations, a selective
review. J. of Applied Mechanics 46, 241258.
Larsen, C.M. and Halse, K.H.(1995):”Comparison of
models for vortex induced vibrations of slender marine
structures. In Proceedings of the Sixth International
Conference on FlowInduced Vibration, London UK, pp.
467482
Vandiver, J. K. (1993). Dimensionless parameters important
to the prediction of vortexinduced vibration of long,
flexible cylinders in ocean currents. J. of Fluids and
Structures 7, 423455.
Vandiver & Li (1996). User Guide for SHEAR7 Version
2.0. MIT, September 1996.
Zdravkovich, M. M. (1981). Review and classification of
various aerodynamic and hydrodynamic means for
suppressing vortex shedding. J. of Wind Engineering and
Industrial Aerodynamics 7, 145189.
DNVOSF201 Dynamic Risers, October 2010
Page 88 Appendix F
DET NORSKE VERITAS
APPENDIX F FRAMEWORK FOR BASIS OF DESIGN
Contents
A. General
A 100 Objective
A 200 Application
B. Design basis
B 100 General
B 200 General design requirements
B 300 Internal fluid data
B 400 Environmental data
B 500 Data for Floater and Stationkeeping System
B 600 Riser system and interfaces
B 700 Analysis methods and load cases
B 800 Miscellaneous
A. General
A 100 Objective
This Appendix defines the items normally to be included
in the design basis document.
A 200 Application
Design basis shall be prepared for all risers.
B. Design basis
B 100 General
101 A design basis document shall be created in the
initial stages of the design process to document the basis
criteria and analysis methodology to be applied in the
structural design of the riser system.
102 When the design has been finalised, a summary
document containing all relevant data from the design and
fabrication phase shall be produced, i.e. a Design,
Fabrication and Installation (DFI) résumé.
103 This section presents the essential of the
information that must be available to the designer, in order
to be able to design the riser according to this standard.
This information is normally included in a design basis
document.
104 Typical information needed to perform a riser
design includes as a minimum:
÷ general riser system design requirements;
÷ functional requirements of the riser system;
÷ operational requirements of the riser system;
÷ internal fluid data;
÷ environmental data;
÷ floater data;
÷ interface requirements and equipment/component
data;
÷ structural analysis methodology including load cases
to be considered;
÷ verification procedures;
÷ miscellaneous.
B 200 General design requirements
201 The operator should specify project specific design
requirements, e.g.:
÷ riser location;
÷ general requirments;
÷ description of the riser system including extent, main
interfaces, configuration, boundary conditions, main
dimensions and main components;
÷ choice of applicable design codes, standards and
regulations;
÷ nominal and minimum internal diameter of equipment
bores interfacing with the riser;
÷ length of each component type;
÷ number off, for each component type;
÷ required service life;
÷ testing ;
÷ fire protection ;
÷ material selection, coating, corrosion protection and
corrosion allowances.
B 300 Internal fluid data
301 The operator should specify all relevant internal
fluid parameters. As relevant, the parameters listed in
Table F1 should be specified. For uncertain data, the
parameters should be specified as realistic ranges
(min/normal/max). Expected variations in the internal fluid
parameters over the service life should be specified.
302 If temperature and pressure is correlated, extreme
combinations of temperature and pressure may be
provided in the form of a design envelope diagram.
303 If rapid decompression of internal gas may occur,
the corresponding adiabatic temperature drop inside
should be calculated by the supplier, and reflected in the
minimum design temperature.
DNVOSF201 Dynamic Risers, October 2010
Appendix F Page 89
DET NORSKE VERITAS
Table F1 Internal fluid parameters
Parameter Comment
Internal
pressure
The following internal pressures should be
specified:
÷ maximum internal pressure including
operating, design and incidental pressure
with possible pressure profile through
service life ;
÷ mill and system test pressure
requirements ;
÷ minimum internal pressure (including
vacuum condition if applicable).
Temperature The following temperature should be
specified:
÷ operating temperature or temperature
profile through service life ;
÷ design maximum temperature ;
÷ design minimum temperature ;
Fluid
composition
Including produced fluids, injected fluids,
exported fluids, and continual and occasional
chemical treatments (dosages, exposure times,
concentrations and frequency) ;
÷ all parameters which define service
conditions, including partial pressure of
H
2
S (sour) and C0
2
(sweet) ;
÷ fluid density range corresponding to
relevant pressure and temperature ;
÷ fluid/flow description including fluid
type and flow regime. ;
÷ sand or particle erosion data ;.
Service
definition
Sweet or sour in accordance with fluid
composition.
Fluid/flow
description
Fluid type and flow regime including slugs.
Annulus fluids for multipipe systems
Flow rate
parameters
Flow rates, fluid density, viscosity.
Thermal
parameters
Fluid heat capacity.
B 400 Environmental data
401 The operator should specify all relevant
environmental parameters. As relevant, the parameters
listed in Table F2 should be considered. Combined wind,
wave and current conditions should be specified for
relevant return periods (e.g. 1, 10 and 100 year return
periods).
402 For temporary (retrievable) risers, the operator
should specify the required range of environmental
conditions (weather window) and planned field locations
for which the riser should be suitable.
403 For environmental conditions at the limits of the
weather window, it should either be possible to safely
retrieve the riser, or it should sustain being hangoff
throughout a design storm specified by the operator.
Table F2 Environmental parameters
Parameters Comment
Location Geographical data for planned fields of
operation.
Water depth Design water depth (minimum and maximum),
tidal variations, storm surge and subsidence.
Seawater data Density, pH value, and minimum and
maximum temperatures.
Air temperature Minimum and maximum during storage,
transportation, installation and operation.
Soil data Description, shear strength or angle of internal
friction, friction coefficients, seabed scour and
sand waves (soil/well and/or soil/pipe structure
interaction characteristics). To be used for
analysis/design riser base foundation, soil
restraint for conductors and soil/structure
interaction evaluation for touch down region for
catenary risers.
Marine growth Maximum values and variations along length of
thickness, density and surface roughness.
Current data Current velocity as a function of water depth,
direction and return period, and including any
known effects of local current phenomena.
Wave data In terms of significant and maximum wave
heights, associated periods, wave spectra, wave
spreading functions and wave scatter diagrams
as function of direction and return period.
Wind data Wind velocity as function of direction, height
above water level and return period.
Ice Maximum ice accumulation, or drifting
icebergs or ice floes.
Earthquake data Ground motions described by means of spectra
or time series.
B 500 Data for Floater and Stationkeeping
System
501 The operator shall specify all data for the floater
and stationkeeping system of relevance for design and
analysis of the riser system.
502 The following general floater data should be
included as relevant for the actual installation:
÷ Main hull dimensions;
÷ Detailed hull geometry, draughts, mass, radii of gyration etc
required to required to perform hydrodynamic
motion/excitation analysis of the floater;
÷ Detailed moonpool geometry, if relevant;
÷ Location of riser supports and riser supporting
structures/devices (e.g. tensioner, moonpool supports etc)
÷ Specification of possible interference areas, including
other risers, mooring lines, platform columns, floater
pontoons, keel, surface equipment and deck, surface
jumper and deck, etc. and definition of allowable
interference/clashing if any.
503 Floater motion characteristics should normally be
specified in the design basis. The following information is
required for documentation of the floater motion
characteristics:
DNVOSF201 Dynamic Risers, October 2010
Page 90 Appendix F
DET NORSKE VERITAS
÷ WF floater motion transfer functions in 6 degrees of
freedom with a clear cut definition of amplitudes and phase
angles as well as wave directions;
÷ The floater motion transfer function shall be given for
relevant loading conditions (i.e. draughts);
÷ The actual water depth at the location and together with the
slender structure restoring force for the actual mooring/riser
system design shall be applied in calculation of WF floater
motion transfer functions;
÷ The floater attached coordinate system used as
reference for floater motion transfer functions shall be
documented in terms of origin (i.e. motion reference
point) and directions of coordinate axes. ;
÷ DP system performance (e.g. position tolerances and
capability curves), if relevant
÷ Mean position and second order motions for relevant
design conditions including intact as well as damaged
conditions due to e.g. mooring line breakage shall be
specified;
504 The design basis document may include relevant
data for evaluation of the global performance of the
installation. The following additional information is
required to conduct coupled and /or decoupled station
keeping analyses
÷ WF and LF transfer functions for hydrodynamic
excitation on the floater.
÷ Frequency dependent added mass and damping for the
floater.
÷ Wind and current coefficients for the floater.
÷ Detailed description of the tethers/mooring system.
For slack/semitaut/taut mooring systems this will
typically include layout pattern of the mooring lines
and detailed mooring line composition (e.g. material
data, description of possible clump weights/buoys,
suspended line lengths, location of anchors and floater
attachment points etc)
÷ DP system characteristics in case of DP assisted
mooring systems
÷ Detailed description of the riser system
÷
A clear cut definition of must be provided for transfer
functions and coefficients (e.g. reference coordinate
system, directions, amplitudes and phase angles etc) to
allow for implementation of these data in the actual
software for stationkeeping analysis.
B 600 Riser system and interfaces
601 The customer should provide the required
information on any interfaces between riser pipe and
adjacent structures, equipment and component data.
602 An overall layout of the riser system should be
provided together with a clear definition of scope of
design, i.e. specification of which parameters/components
of the riser system that are subject to design (typical
examples are wall thickness, material quality, buoyancy
modules, stress joints etc). Indications of preferred
solutions should be given to the extent possible. Examples
of information that may be included in the design basis
document are:
÷ Riser configurations;
÷ Arrangement of risers, in case of more that one riser ;
÷ Riser joints including cross section data, annulus
content, riser joint length, connectors, attachments etc;
÷ Description of buoyancy modules such as aircans,
midwater arch and distributed buoyancy modules;
÷ Description of additional external lines , umbilical etc;
÷ Description of structural components of relevance for
the actual installation (e.g. stress joints, flex joints,
mechanical connectors, tension joint, ball joints
emergency disconnect package, etc)
603 A general description of the top interface between
riser system and adjacent structure should include
information, such as:
÷ Floater support boundary conditions;
÷ geometry, stroke, pulling capacity, load/displacement
characteristics (linear/nonlinear) and failure tolerance
of tensioner systems, if any;
÷ design of temporary and permanent riser top
suspension systems (spiders, etc.);
÷ surface equipment like surface flow tree, jumpers, etc.
604 A general description of the bottom interface and
subsea equipment should be included in the design basis
document. The following information may be included as
relevant for the actual installation:
÷ wellhead datum relative to sea level;
÷ seafloor conditions including characteristic soil
properties (e.g. stiffness, friction coefficients etc);
÷ conductor stiffness and soil restraint;
÷ subsea template dimensions and stiffness ;
÷ subsea equipment like BOP, subsea tree, EDP, LMRP,
LWRP, etc.
605 The operator should provide information on the
permissible loading (e.g. pressure, tension and bending
moment) of the wellhead equipment and the top
suspension, to which the riser is connected.
606 For temporary top tensioned risers, the maximum
allowable disconnect angle of the emergency disconnect
package (EDP) should be defined by the operator for input
to the operating condition limits for the riser analysis.
607 For risers equipped with flexjoints, the maximum
permissible deflection angle should be defined for the
relevant tension and pressure ranges.
B 700 Analysis methods and load cases
701 The intended procedures to be adopted in the
design of the risers shall be documented. All applicable
limit states for all relevant temporary and operational
design conditions shall be considered. The following
should be included:
702 Design criteria for all relevant temporary phase
conditions including, as relevant:
DNVOSF201 Dynamic Risers, October 2010
Appendix F Page 91
DET NORSKE VERITAS
÷ limiting pressure, functional and environmental load
criteria and design load combinations (cases) ;
÷ essential design parameters and analytical procedures
associated with temporary phases e.g. transportation,
lifting/handling, installation, retrieval, connection and
disconnection ;
÷ relevant ALS criteria ;
÷ riser abandonment.
703 Design criteria for all relevant operational phase
conditions including, as relevant for the actual installation:
÷ limiting pressure, functional and environmental load
criteria and design load combinations (cases) ;
÷ essential design parameters and procedures associated
with operational phases e.g. top tension, vessel offset,
internal pressure and related internal fluid density;
÷ relevant ALS criteria, e.g. tensioner failure, drive/drift
off, collision, explosion, fire, dropped objects etc;
÷ relevant SLS criteria for the riser pipe and structural
components
704 A general description of analysis models to be
utilised, including description of :
÷ global analysis model(s) including modelling for wave
and current loading and floater motions;
÷ local analysis model(s);
÷ load cases to be analysed.
705 A general description of the structural evaluation
process, including:
÷ description of procedures to be utilised for
considering global and local responses;
÷ description of procedures to be utilised for combining
global and local responses;
÷ criteria for limit state checking;
÷ description of fatigue evaluation procedures
(including use of design fatigue factors, SNcurves,
basis for stress concentration factors (SCF’s), etc.) ;
description of procedures to be utilised for code checking.
B 800 Miscellaneous
801 A general description of other essential design
information, including:
÷ inservice inspection criteria general philosophy for
inspection, maintenance and repair/replacement;
÷ Procedures/scope for verification of the riser design
(e.g. testing and independent review/analyses of the
design);
÷ weak links (if relevant).
FOREWORD
DET NORSKE VERITAS (DNV) is an autonomous and independent foundation with the objectives of safeguarding life, property and the environment, at sea and onshore. DNV undertakes classification, certification, and other verification and consultancy services relating to quality of ships, offshore units and installations, and onshore industries worldwide, and carries out research in relation to these functions. DNV service documents consist of amongst other the following types of documents: — Service Specifications. Procedual requirements. — Standards. Technical requirements. — Recommended Practices. Guidance. The Standards and Recommended Practices are offered within the following areas: A) Qualification, Quality and Safety Methodology B) Materials Technology C) Structures D) Systems E) Special Facilities F) Pipelines and Risers G) Asset Operation H) Marine Operations J) Cleaner Energy
O) Subsea Systems
The electronic pdf version of this document found through http://www.dnv.com is the officially binding version © Det Norske Veritas Any comments may be sent by email to rules@dnv.com For subscription orders or information about subscription terms, please use distribution@dnv.com Computer Typesetting (Adobe Frame Maker) by Det Norske Veritas
If any person suffers loss or damage which is proved to have been caused by any negligent act or omission of Det Norske Veritas, then Det Norske Veritas shall pay compensation to such person for his proved direct loss or damage. However, the compensation shall not exceed an amount equal to ten times the fee charged for the service in question, provided that the maximum compensation shall never exceed USD 2 million. In this provision "Det Norske Veritas" shall mean the Foundation Det Norske Veritas as well as all its subsidiaries, directors, officers, employees, agents and any other acting on behalf of Det Norske Veritas.
Offshore Standard DNVOSF201, October 2010
ACKNOWLEDGEMENT This Offshore Standard has been developed in close cooperation with the industry. The basis for the standard was developed within the recently completed 4 year Joint Industry Project “Design Procedures and Acceptance Criteria for Deepwater Risers”. The JIP was performed by DNV, SINTEF and SeaFlex and supported by international oilcompanies and national authorities. In addition to the feedback from the JIP steering committee the Standard has been circulated on extensive internal and external hearing. The following organisations have made contributions to the standard. Coflexip Stena Offshore DST ELF Europipe Exxon Prod. Research Company MCS Norway Norsk Hydro NPD Phillips Petroleum Saga Petroleum SeaFlex Stolt Offshore SINTEF Stress Engineering Shell Statoil
DNV is grateful for the valuable cooperations and discussions with the individual personnel of these companies. CHANGES
• General • Main changes
As of October 2010 all DNV service documents are primarily published electronically. In order to ensure a practical transition from the “print” scheme to the “electronic” scheme, all documents having incorporated amendments and corrections more recent than the date of the latest printed issue, have been given the date October 2010. An overview of DNV service documents, their update status and historical “amendments and corrections” may be found through http://www.dnv.com/resources/rules_standards/.
Since the previous edition (January 2001), this document has been amended, most recently in October 2009. All changes have been incorporated and a new date (October 2010) has been given as explained under “General”.
DET NORSKE VERITAS
Offshore Standard DNVOSF201. October 2010 DET NORSKE VERITAS .
.............................22 Extreme Combined Load Effect Assessment .................................... 11 DESIGN PHILOSOPHY AND DESIGN PRINCIPLES.............................21 Floater Motion.... B 100 B 200 B 300 B 400 B 500 B 600 B 700 C..... 29 Resistance Factors.........34 Fatigue assessment using SN curves ..27 Load Effects........................................................... D 100 D 200 D 300 D 400 D 500 Functional Loads .......... C 100 C 200 C 300 D.......... 20 Definition .................................................. 31 General.................................................. 19 Determination of Pressure Loads................. A 100 A 200 B.......6 Verbal forms .................................................................................................... C 100 C 200 D.......12 Objective.... 20 Definition ..............29 Geometrical Parameters..................................................... 12 Safety Objective ...........20 Waves .................. 13 Operational considerations.....32 Propagating Buckling. C 100 C 200 D. 28 Design Load Effects..............................23 Design Based on Environmental Statistics ............................................................................................................................................................................................. 15 Design by WSD Method......................36 SECTION 3 A...20 Environmental Loads ........................12 SECTION 4 A....... 14 Quality Assurance and Quality System...................................................35 Inservice Fatigue Inspections ........27 Limit States.............. E 100 E 200 E 300 E 400 F........ 22 General ..19 Definition............ 27 SECTION 2 A........................................ 17 Design by Testing........................................................................................................................................................................................ 27 Objective .........35 Accidental Limit State ................................18 Objective................29 Material Strength.....................................CONTENTS SECTION 1 A................................................................................. 14 Design by LRFD Method ............................................................................................................................................... 17 LOADS ......................................................................14 Basic Considerations....................................................................22 Application .................................................................................... 1 Objectives ..........................33 Displacement Controlled Conditions .................................... 3 Rules ........................................................................................... 3 Recommended Practices .........28 Load Effect Factors ......................................... 14 Design Format............................................................................................................................................................22 Riser Analysis Procedure ...........33 Fatigue Limit State ............................ D 100 D 200 D 300 GENERAL....... 24 General........................................................................................................ A 100 A 200 A 300 A 400 A 500 B........................................................................31 System Hoop Buckling (Collapse)......................................................................20 Environmental Load Condition.. B 100 B 200 B 300 B 400 B 500 B 600 B 700 C....................... 20 Pressure Ratings...............................................................................................................................................................................................................21 ANALYSIS METHODOLOGY............................................................................................................................................. 22 Fundamentals .......22 Generalised Load Effect.. 1 Scope and Application ...................................................... C 100 C 200 General....................................................................... C 100 C 200 C 300 C 400 C 500 C 600 SECTION 5 A....................................................................... 3 Guidelines........................ 2 Normative References .....................................................................1 Introduction............................24 Global Analysis ....30 Ultimate Limit State .............................................. 2 Structure of Standard ......... 3 Certification notes and Classification notes ........ 1 Other Codes ..............................................24 Fatigue Analysis ... B 100 B 200 B 300 B 400 B 500 C..............................................................................................................................................23 Load Cases ................3 Offshore Service Specifications ..... 36 Functional requirements ............................................................................................................................ 12 Systematic review.......................................................... 12 Application........31 Bursting....................... 18 Application........................ 10 Greek Characters ............. 12 Safety Philosophy........27 Application .............36 Categories of accidental loads....12 General........................................................ 20 DET NORSKE VERITAS ........................... A 100 A 200 A 300 B............33 Alternative WSD Format......................................................................................................... 13 Fundamental requirements ..........................................................................................................................................................................24 Design Based on Response Statistics ........ 3 Other references ..................................................................................................................................................................18 General................................................ 3 Offshore Standards ........................................................................ 3 Definitions. A 100 A 200 A 300 B............................................................................................................................................................. 34 General............... 16 Reliability Based Design ............................................ 22 Objective ...............................................25 DESIGN CRITERIA FOR RISER PIPES ..................................................................................................................... 6 Abbreviations and Symbols ........................................................................................................................................................ 6 Definitions....... 14 Safety Class Methodology .............................................................. 18 Loads ...20 Determination of Functional Loads............................................................................. 19 Pressure Control System.......................................29 Resistance...... B 100 B 200 C............ B 100 B 200 B 300 B 400 General ................................................................................ 13 Design Principles....................................................................................................................................20 Current...... A 100 A 200 A 300 B.............. D 100 D 200 D 300 D 400 D 500 D 600 D 700 E.............9 Abbreviations ................. F 100 F 200 General ...................................... 9 Symbols ...............................................................34 Fatigue assessment by crack propagation calculations.32 Combined Loading Criteria .................. 18 Pressure Loads ..........................................................................1 C.......
........ 47 Installation and Operation........64 Hydrodynamic coefficients ............................................. A 100 A 200 A 300 B............... B 100 B 200 B 300 B 400 B 500 B 600 B 700 C........................... 38 Riser stroke............................... 39 Unstable Fracture and Gross Plastic Deformation ...............................51 Condition Summary................................................................... C 100 C 200 C 300 General ....................................... J....................................................................................50 General.. 42 Connector Designs ...............38 General...................................................47 Design47 Design basis ...................54 Compliant riser configurations.... B 100 B 200 General ................63 Slug flow...........61 Marine growth.. 45 DOCUMENTATION AND VERIFICATION.........49 General requirements ..... 45 Long term properties ............................................................. A 100 General ...................................................................39 Interference.. C 100 C 200 C 300 C 400 C 500 D...................60 Morison equation for double symmetric crosssections .................... 43 Documentation.................... 36 Design against accidental loads............................. B 100 B 200 B 300 B 400 C.F 300 F 400 G.............. A 100 B............ 42 Design and Qualification Considerations ......................................................................... H 100 H 200 H 300 H 400 I.................. B 100 B 200 B 300 B 400 C.......................64 SECTION 7 A................ 38 Examples. C 100 C 200 C 300 C 400 C 500 C 600 Inservice Inspection................................................. Replacement and Monitoring .....45 General........44 Objective........................................ C 100 C 200 General ................................................... 40 CONNECTORS AND RISER COMPONENTS....................... 47 Design analysis ...................... H 100 H 200 H 300 Characteristic accidental load effects ..............................63 Hydrodynamic loading in moonpool........................ 48 Verification............... A 100 B............................................................................................................................................ G...51 Extended Service life........62 Hydrostatic pressure loading ................................................................................................................................................... 43 Operating and maintenance manuals ........................58 Efficient analysis strategies considering coupling effects ..................................................................62 Steady flow..............53 Physical Properties of riser systems ....................................................................................................50 Riser monitoring...................................... 43 Local Analysis ................59 General.......................................47 Objective.............................................................................................................................................................................51 Ultimate Strength........64 General........ 47 Documentation..................... H............................................................................54 Nonlinearities...........................43 Documentation..........................63 Forced Floater Motions .....................................................51 Material Properties................................62 Accelerated uniform flow............47 SECTION 8 A....................................... 49 OPERATION....................................................................................................... MAINTENANCE AND REASSESSMENT......42 Objective....................58 Coupled system analysis .........................50 Riser Inspection .....................................................................54 Top tensioned risers ......................................51 Reassessment .........52 Dimensions and Corrosion Allowance.................62 General.................................... 49 Verification during the fabrication phase.........................58 Coupled floater motion analysis ..................57 Combined floater/slender structure analysis ......................................................................54 General...............56 Static finite element analysis .................................. 43 MATERIALS.................................................................42 B........................ 47 Manufacture and fabrication .......................... 42 Seals .55 Purpose of global analysis ................................................................................................................50 Objective.........................................................................................................................................................................42 Functional Requirements ... 37 Serviceability Limit State .........................62 Internal fluid flow........................................................................................................................ 48 DFI Résumé................................. D 100 D 200 D 300 D 400 D 500 E................. 49 Verification during the design phase.....................................55 Global riser system analysis...........52 SECTION 6 A........................................................................................................................................... 38 Special Considerations ......................................................59 Hydrodynamic loading on slender structures .................................53 Objective....................................................................... 50 DET NORSKE VERITAS ............. J 100 J 200 J 300 GLOBAL ANALYSIS ................ 44 Additional Requirements ............. 38 Ovalisation limit due to bending .................................... G 100 G 200 G 300 G 400 H.........................................................................44 APPENDIX A A..................59 Morison equation for circular crosssections.51 Guidelines for inspection intervals .............................................................................53 General .......................................... 48 Filing of documentation.......59 Decoupled floater motion analysis ..................................................................................... 40 Global Buckling.....52 Cracked Pipes and Components .............................64 Moonpol kinematics ...........................................................................................58 General.......................................................... 44 Material Selection ..........................................................50 SECTION 9 A........................................................................................................................... 44 Application...........................................................................56 Finite element eigenvalue analysis .........60 Principles for selection of hydrodynamic coefficients................................... B 100 B 200 B 300 B 400 B 500 C............................................................................................ A 100 B..............................................................................57 Dynamic finite element analysis .................................................. E 100 E 200 E 300 E 400 F....................................................................................55 General modelling/analysis considerations......................................................51 General...
...........90 Analysis methods and load cases ..... 70 Wide band Fatigue Damage Assessment......................90 Miscellaneous .. 67 Fatigue design.........................88 General design requirements ......................... 71 Semiempirical Solutions............... 85 APPENDIX E A.................................. 77 Computer implementation ...................... 65 FATIGUE ANALYSIS ............................................. 80 Standards....................................................... 77 APPENDIX F A..................................................................................88 Internal fluid data...64 Global Rayleigh damping model.. 77 Implementation in design analyses ... 67 Application...............................................................77 General.......................87 References ...................... B 100 B 200 B 300 C.......74 General....... C 200 C 300 D...... 85 Objective ............ C................ 65 References...............................................................................82 Frequency discretisation .......................................................................... Guidelines and Handbooks ...69 General.....................................................................71 General ....................................... B 100 B 200 B 300 C........................88 Application ..78 Statistical uncertainty and simulation planning............................................................. A 100 B...................... A 100 A 200 A 300 A 400 B........................................................................................................ B 100 B 200 B 300 B 400 B 500 B 600 B 700 B 800 General...........................................................................................................................................................85 Multimodal Response Analysis Based on Empirical Models ..................................... 75 Generalised load effect..............................................65 Standards............. General...........................................................................85 Fatigue Assessment............... 74 Design based on response statistics .......................................................................................................................89 Riser system and interfaces.......................................... 77 Shortterm extreme load effect estimation... A 100 A 200 B.......... 77 Implementation of the WSD design format.............. 76 ULS Analysis Procedure...................................................................................................................80 VERIFICATION OF GLOBAL ANALYSIS MODEL.....91 DET NORSKE VERITAS ..............................81 Verification of theoretical models .....................78 Extreme response estimation..... E 100 General ........... 71 Fatigue Capacity SN Curves . 75 Implementation of the LRFD design format................. 71 References................................................................................................67 Objective.................. 68 Basic fatigue damage methodology............................................................................... 81 APPENDIX B A...79 Response surface approach........................75 General..79 Longterm load effect statistics........................... 67 Methods for fatigue damage assessment.............. 88 APPENDIX C A.........................................................................................67 E 200 E 300 E 400 F........................................................................ G 100 G 200 Envelope statistics........... General.........................................................88 Design basis ..................... 85 Simplified Assessment of Fatigue Damage .....81 Introduction........................................................................................................................83 Time discretisation...................... 75 Long term acceptance criteria .................................................................... C 200 C 300 C 400 D.............................................................. 74 Design based on environmental statistics .........71 General............... 65 Technical references ....K.............................. 76 Post processing procedures ....... 88 Objective .........................80 Technical references . D 100 D 200 E............................. C 100 C 200 C 300 D........... 69 Narrow Band Gaussian Fatigue Damage ............................................................ 75 Shortterm acceptance criteria................................................. E 100 F................................... 71 Analytical Solutions for Bimodal Spectra .............................................................. B 100 B 200 B 300 C................................. General ..............83 References ...................... A 100 A 200 B........88 Environmental data ... 67 Fatigue analysis procedures ............. 70 Narrow Band NonGaussian Fatigue damage...........................................................................................86 Methods for reduction of VIV ...... A 100 B.............................................................68 General..............77 General............. 86 Modification of Riser Properties........................................ D 100 D 200 D 300 D 400 E....... K 100 K 200 L.. 84 VIV ANALYSIS GUIDANCE..................................................................................................................................................... 68 Global fatigue analysis procedures .............................................................. 69 Narrow Band Fatigue Damage Assessment........................................................................... 79 General................................................... 87 FRAMEWORK FOR BASIS OF DESIGN.... 88 General............................................... 71 Cycle counting............................74 Objective...................................80 References ............ F 100 F 200 G.............................................................. C 100 C 200 C 300 C 400 C 500 C 600 C 700 D.........................................................................................................................................................................73 ASSESSMENT OF EXTREME LOAD EFFECT FOR COMBINED LOADING74 APPENDIX D A... Guidelines and Handbooks....................... 81 Verification of numerical procedures .. 81 Objective .86 Vortex suppression devices ............................................. 82 Spatial discretisation......................... 64 Local Rayleigh damping models ...........................89 Data for Floater and Stationkeeping System............ 74 Design principles......86 Methods Based on Solution of the NavierStokes equations .............................................. L 100 L 200 Structural damping .......
Multitube crosssections (i. completion/ workover. guidance and requirements for efficient global analyses and introduce a consistent link between design checks (failure modes). allowance for the use of innovative techniques and procedures. riser application and configuration. Aspects relating to documentation. A 100 A 200 A 300 A 400 A 500 B.g. 304 There are. or transportation of hydrocarbons (import/export) in the petroleum and gas industries. However.Guidance .end . 302 The scope covers design. serve as a technical reference document in contractual matters. As an alternative. such as reliabilitybased design methods. DET NORSKE VERITAS . riser connectors and other riser components such as tension joints and stress joints. C 100 C 200 D. pipes inside pipes) are not considered explicitly. The main purpose is to cover design and analysis of top tensioned and compliant steel riser systems operated from floaters and fixed platforms. A 300 Scope and Application 301 This standard applies to all new built riser systems and may also be applied to modification. requirements and guidance on structural design and analysis of riser systems exposed to static and dynamic loading for use in the offshore petroleum and natural gas industries.g. special attention shall be given to identify possible new failure mechanisms. maintenance and reassessment of riser systems. testing. operation and upgrading of existing risers. verification and quality control are also addressed. Boundary conditions of the pipes. However. October 2010 Section 1 Page 1 SECTION 1 GENERAL Contents A. validity/adequacy of analysis methodology and new loads and load combinations. B 100 B 200 B 300 B 400 B 500 B 600 B 700 C. and reflect the stateoftheart and consensus on accepted industry practice and serve as a guideline for riser design and analysis. temperature and local contact loads should be considered in particular.e. D 100 D 200 D 300 General Introduction Objectives Scope and Application Other Codes Structure of Standard Normative References Offshore Service Specifications Offshore Standards Recommended Practices Rules Certification notes and Classification notes Guidelines Other references Definitions Verbal forms Definitions Abbreviations and Symbols Abbreviations Symbols Greek Characters A 200 Objectives 201 The main objectives of this standard are to: provide an international standard of safety for steel risers utilised for drilling. operation. materials. water depth. production and export/import of hydrocarbons and injection of fluids). 103 The basic design principles and functional requirements are in compliance with stateoftheart industry practice. The standard applies for permanent operation (e. provision of stateoftheart limit state functions in a Load and Resistance Factor Design (LRFD) format with reliabilitybased calibration of partial safety factors. production/injection.g.DNVOSF201 Dynamic Risers. application of safety class methodology linking acceptance criteria to consequence of failure. 303 This standard is applicable to structural design of all pressure containing components that comprise the riser system. this standard may be applied for design of each individual tubular of such crosssections provided a realistic (conservative) distribution of the loading on each individual tubular are assumed. General A 100 Introduction 101 This standard gives criteria. drilling and completion/workover activities).note  102 The major benefits in using this standard comprise: provision of riser solutions with consistent safety level based on flexible limit state design principles. with special emphasis on: single pipes with a ratio of outside diameter to wall thickness less than 45.of . for novel applications where experience is limited. in principle. a simple conservative Working Stress Design (WSD) format is also given. fabrication. load conditions and load effect assessment in the course of the global analyses. no limitations regarding floater type. as well as for temporary operation (e. . umbilical) if the loading on the pipe can be adequately predicted. Guidance note: This standard may also be applied to design of single steel pipes used as components in more complex composite crosssections (e. A.
g. This may typically involve validation of computational methodology by physical testing.of . The section is supported by appendix A providing additional information on global analyses. and testing. for reeling) shall comply with DNVOSF101. appendix C on assessment of extreme load effect for combined loading. October 2010 Page 2 Section 1 Guidance note: For application of this standard to new riser types/concepts (e.g. DNVOSF101 forms the primary reference for materials. 502 The main part is organised as follows: Section 1 contains the objectives and scope of the standard. Important internal pressure definitions are given.of . 405 This standard is compatible with the DNV Offshore Standard for Submarine Pipeline Systems DNVOSF101. novel hybrid systems. . definitions and abbreviations.note touchdown point) are load controlled unless otherwise argued and documented. A 500 Structure of Standard 501 This standard consist of two parts: 1. load effect assessment and load cases. testing and fabrication for riser pipes.end . loads and global analysis aspects valid for all riser materials. complex riser bundles etc) it shall be documented that the global load effects can be predicted with same precision as for conventional riser systems. . to the safety philosophy and analyses methodology set forward by this standard. for consistency.end . appendices containing practical guidance and background information on topical issues. Section 2 contains the fundamental design philosophy and design principles. 403 The framework within DNV Riser standards and RP’s is illustrated in Figure 11. These Recommend Practice (RP) documents subscribe.Guidance . unless otherwise noted.Guidance . global analysis. This is discussed in more details in Section 5. It further introduces essential concepts. Figure 11 Framework for DNV Riser Standards and RP’s 404 This standard provides the design philosophy. DET NORSKE VERITAS . − the API codes (RP2RD and RP1111) implicitly assumes displacement controlled riser configuration with secondary bending stress for ULS design checks. the requirements of this standard shall prevail. the valid revision shall be taken as the revision that was current at the date of issue of this standard. appendix D on verification of global analysis model. Examples of some typical components/important areas included in typical riser systems are illustrated in Figure 13.DNVOSF201 Dynamic Risers. A 400 Other Codes 401 In case of conflict between requirements of this standard and a reference document. The limit state design checks for this standard and DNVOSF101 is similar but due to difference in the governing failure modes and prevailing uncertainties some difference in safety factors exist. It introduces the safety class methodology and normal classification of safety classes. 2. Guidance note: The major differences/conflicts in design principles compared to current industry practice reflected by API (RP2RD and RP1111) are: − in the ASME and API codes the hydrostatic pressure test is fundamental and often drives the design of pipelines and export risers. The limit state based DNVOS aim to design for the actual modes of failure and the safety margin is ensured by a combination of material requirements. Strain limits and acceptance criteria for displacement controlled conditions of pipes (e. Section 4 contains the framework for global analysis methodology. Specific acceptance criteria for steel are given in this standard while titanium and composite materials are currently under development in associated recommended practices.note  OSF201 Design Criteria STEEL Design Philosophy Loads Analyses OSF101 Material Testing Installation 305 Examples of typical floater and riser configurations are shown schematically in Figure 12. The DNVOS a priori assumes that important riser locations (top. appendix B on fatigue analyses. This implies that the fatigue criterion in API is used as an implicit control of excessive bending rather than explicit ULS design checks where relevant as in this standard. see B 700. a main part providing minimum requirements in terms of explicit criteria where relevant and functional requirements elsewhere. functional loads and environmental loads. It provides a consistent link between design checks for combined loading. RPF201 TITANIUM Material Testing Design Criteria RPF202 COMPOSITE Material Testing Design Criteria Rules FLEXIBLES 402 Where reference is made to codes other than DNV documents. Section 3 contains a classification of loads into pressure loads. As an alternative an appropriate conservatism in design should be documented.and In addition a number of supporting documents may be required as listed in section B.
5 DNV CN 30. Section 8 contains requirements for documentation and verification of the riser system. Type Approval Conformity Certification Services. October 2010 Section 1 Page 3 section 5 on acceptance criteria for combined loading Section 5 contains acceptance criteria for the riser pipe for ULS.2 DNV CN 30. manufacture. General Principles on Reliability for Structures Petroleum and natural gas industries Design and operation of subsea production systems .note  B 600 Guidelines DNV B 100 Offshore Service Specifications DNVOSS301 Certification and Verification of Pipelines Submarine Pipeline Systems Structural Design of TLPs by the LRFD Method Structural Design of Deep Draught Floating Units Cathodic Protection Design Fatigue Strength Corroded Pipelines Mechanical Pipeline Couplings Free Spanning Pipelines Factory applied Pipeline Coatings for Corrosion Control Fracture Control for Reeling of Pipelines Titanium Risers Composite Risers B 700 Other references BS 7910 B 200 Offshore Standards DNVOSF101 DNVOSC105 DNVOSC106 API RP1111 API RP2RD B 300 Recommended Practices DNV RP B401 DNV RPC203 DNV RPF101 DNV RPF104 DNV RPF105 DNV RPF106 DNV RPF108 DNV RPF201 DNV RPF202 EUROCODE 3 ISO/FDIS 2394 IS0/CD 136287 Guidance note: The latest revision of the DNV documents may be found in the publication list at the DNV website www.end .Guidance .5 Conformity Certification Services.DNVOSF201 Dynamic Risers.note  DET NORSKE VERITAS . SLS. ALS and FLS. Approval of Manufacturers.com. Section 6 contains the fundamental functional requirements for connectors and riser components.6 B. Section 7 contains requirements for materials. .Guidance . fabrication and documentation of riser pipe and components where the principles and requirements in OSF101 is adhered. This includes a definition of resistance and load effects and safety factors for explicit limit states.of . Operation.of . Appendix F provides additional information.Part 1.2 DNV CN 1. Metallic Materials Ultrasonic Inspection of Weld Connections Fatigue Strength Analysis for Mobile Offshore Units Foundations Environmental Conditions and Environmental Loads Structural Reliability Analysis of Marine Structures Guidelines for Flexible Pipes Guide on methods for assessing the acceptability of flaws in fusion welded structures Design.dnv.1: General rules and rules for building.Part 7: Completion/workover riser systems DNV CN 7 DNV CN 30. and Maintenance of Offshore Hydrocarbon Pipelines (Limit State Design) Design of Risers for Floating Production Systems (FPSs) and TensionLeg Platforms (TLPs) Design of steel structures . DNV RP O501 Erosive Wear in Piping Systems B 400 Rules DNV Rules for Certification of Flexible Risers and Pipes DNV Rules for Planning and Execution of Marine operations DNV Rules for Classification of Fixed Offshore Installations B 500 Certification notes and Classification notes DNV CN 1. Normative References The latest revision of the following documents applies: Guidance note: Explicit reference to paragraphs in DNVOSF101 should relate to January 2000 version. Construction. If other codes are applied additional evaluations is required.4 DNV CN 30.end . . Section 9 contains basic requirements for operation and inservice operations.
DNVOSF201 Dynamic Risers. October 2010 Page 4 Section 1 Figure 12 Examples of metallic riser configurations and floaters DET NORSKE VERITAS .
October 2010 Section 1 Page 5 Figure 13 Examples of riser components DET NORSKE VERITAS .DNVOSF201 Dynamic Risers.
personnel or environment is significantly reduced. such as waves.DNVOSF201 Dynamic Risers. local: Buckling mode implying deformations of the cross section.g. 221 Flex joint: A laminated metal and elastomer assembly. guide tools. that is positioned in the riser string to reduce the local bending stresses. Definitions C 100 Verbal forms 101 “Shall”: Indicates requirements strictly to be followed in order to conform to this standard and from which no deviation is permitted. "by agreement": Unless otherwise indicated. in the event of failure or malfunction of any part of the system. in terms of resistance of relevant structural models obtained in accordance with specified principles. 208 Connector or coupling: A mechanical device use to connect adjacent components in the riser system. having a central throughpassage equal to or greater in diameter than the interfacing pipe or tubing bore. 214 Dynamic Positioning (DP. Ship. 211 Design load: The combination of load effects. 203 Buckling. or that a certain course of action is preferred but not necessarily required. unintended and undesirable event. DET NORSKE VERITAS . 102 “Should”: Indicates that among several possibilities. global: This is usually referred to as elastic Euler buckling or bar buckling. to reduce the submerged weight of the riser. loss of component or system function. Semi. 204 Buckling. 218 Failure: An event causing an undesirable condition.e. automatic station keeping): A computerised means of maintaining a floater on location by selectively driving thrusters. i. buoyancy control line and mud boost line. without mentioning or excluding others. 219 Fatigue: Cyclic loading causing degradation of the material. TLP. e. representing one or more failure modes.g. serve as a running and retrieving string for the BOP. this means agreed in writing between Manufacturer/ Contractor and Purchaser. C 200 Definitions 201 Accidental loads: Loads acting on the riser system. 216 Environmental loads: Loads due to the environment. 209 Corrosion allowance: The amount of wall thickness added to the pipe or component to allow for corrosion/erosion/wear. 206 Collapse capacity: Resistance against external overpressures. and drilling strings. hoop buckling failure (collapse). October 2010 Page 6 Section 1 C. 220 Fail safe: Term applied to equipment or a system so designed that. current. Deep Draft Floater etc. devices are automatically activated to stabilise or secure the safety of the operation. e. wave frequency motions and low frequency wind and wave motions. 213 Drilling riser: A riser utilised during drilling and workover operations and isolates any wellbore fluids from the environment. taking into account the floater mean offset. ice and earthquake. connecting two joints of riser pipe endtoend. Typical accidental event has an annual probability of occurrence less than 102. 202 Auxiliary line: A conduit (excluding choke and kill lines) attached to the outside of the riser main pipe such as hydraulic supply line. 104 "Agreement". 207 Completion/Workover riser (C/WO riser): Temporary riser used for completion or workover operations and includes any equipment between the subsea tree/tubing hanger and the workover floaters tensioning system. Drilling risers may also be used for well completion and testing. multiplied by their respective load effect factors. or deterioration of functional capability to such an extent that the safety of the unit. strapped or clamped to the exterior of riser joints. 215 Effective tension: The axial wall force (axial pipe wall stress times area) adjusted for the contributions from external and internal pressure. 222 Floater: Buoyant installation. which is floating or fixed to the sea bottom by mooring systems in temporary or permanent phases. support auxiliary lines.g. Spar. 217 Export/import riser: Export/import risers transfer the processed fluids from/to the floater/structure to/from another facility. be due to external pressure (hoop buckling) and moment (wrinkling) or a combination thereof. wind and waves. because of a sudden. This can e. 205 Buoyancy modules: Structure of light weight material.g. wind. usually foamed polymers. one is recommended as particularly suitable. 210 Design checks: Design checks are investigations of the structural safety of the riser under the influence of load effects (design load cases) with respect to specified limit 224 Floater mean offset: The offset created by steady forces from current. The major functions of drilling riser systems are to provide fluid transportation to and from the well. 103 “May”: Verbal form used to indicate a course of action permissible within the limits of the standard. states. e. Other possibilities may be applied subject to agreement. 212 Design resistance: The resistance divided by the appropriate resistance factor(s). 223 Floater offset: The total offset of the floater. which may include an another platform/floater or pipeline.
emergency disconnect. strain. The numerical definition of out of roundness and ovalisation is the same. 232 Interface loads and displacements: Loads and displacements at a particular boundary between two systems. leak or unstable fracture. 250 Pressure definitions 236 Load effect factor: Partial safety factor by which the load effect is multiplied to obtain the design load (effect). This has the form as an elliptic cross section. 249 Permanent riser: A riser. such as bending moment. effective tension. 233 Limit state: The state beyond which the riser or part of the riser no longer satisfies the requirements laid down to its performance or operation.e. Set Point Maximum Allowable Incidental Pressure. etc. 240 Material resistance factor: Partial safety factor transforming a resistance to a lower fractile resistance.DNVOSF201 Dynamic Risers. p inc 238 Load and Resistance Factor Design (LRFD): Design format based upon a Limit State and Partial Safety Factor methodology. Incidental Pressure. The partial safety factor methodology is an approach where separate factors are applied for each load effect (response) and resistance term. 228 Functional loads: Loads caused by the physical existence of the riser system and by the operation and handling of the system. excluding pressure loads. strain. which will be in continuous operation for a long time period. irrespective of environmental conditions. overpull testing. Incidental Operation: Conditions that are not part of normal operation of the system. and termed the wave frequency (WF) regime. Pressure Safety System Tolerance Mill Test Pressure. e. Such conditions may for example be surges due to bullheading. 242 Mode of operation: The riser mode of operation includes typically running. flattening. stress. or a local out of roundness. landing and connecting or such as laying. sudden closing of valves. 230 Hangoff: Riser when disconnected from seabed. deformation. 247 Out of roundness: The deviation of the perimeter from a circle. Figure 14 Pressure definitions DET NORSKE VERITAS . local buckling) or operational limitations (stroke or clearance). landing and connecting. 231 Installation: The operation related to installing the riser system. motion. i. 241 Maximum operating condition: Maximum condition in which the normal operations are carried out. etc. This should include steady flow conditions over the full range of flow rates as well as possible packing and shutin conditions where these occur as part of routine operation. The numerical definition of out of roundness and ovalisation is the same. or near surge. 234 Load: The term load refers to physical influences which cause stress. sway and yaw eigenperiods for the floater. 229 Global analysis: Analysis of the complete riser system. connected shutin.g. tiein. connected production (well access). 246 Operation. 239 Low Frequency (LF) motion: Motion response at frequencies below wave frequencies at. causing the platform to move at periods typically between 3 – 25 seconds. Examples are structural failure (rupture. deformation. Such conditions may lead to incidental pressures. pressure testing. in the riser. etc. 227 Fracture analysis: Analysis where critical initial defect sizes under design loads are identified to determine the crack growth life to failure. October 2010 Section 1 Page 7 225 Floater wave frequency motions: The motions that are a direct consequence of first order wave forces acting on the floater. wellkill. hangoff (disconnected). 245 Operation. 226 Flowline: Any pipeline connecting to the subsea tree assembly. i. Normal Operation: Conditions that are part of routine (normal) operation of the riser system.e. for a catenary riser. Set Point Pressure Regulating System Tolerance Maximum Allowable Operating Pressure. 237 Location class: A geographic area classified according to the distance from locations with regular human activities. or failure of the pressure regulating system and activation of the pressure safety system. 235 Load effect: Response or effect of a single load or combination of loads on the structure. 248 Ovalisation: The deviation of the perimeter from a circle. This can be an ovalisation. disconnecting. displacement. 243 Nominal value: Specified value. such as running of riser joints. 244 Operating envelope: Limited range of parameters in which operations will result in safe and acceptable equipment performance. an elliptic cross section. LF motions typically have periods ranging from 30 to 300 seconds.
the environment and to assets and financial interests. completion and other purposes. design is the maximum internal pressure during normal operations. stress joints. The design pressure must take account of steady flow conditions over the full range of flow conditions as well as possible packing and shutin conditions. 270 Riser joint: A riser joint consists of a pipe member mid section. 258 Pressure. unintended shutin pressure. injection. bullheading (waterhammer) or any temporary incidental condition. Maximum Allowable Incidental (MAIP): The maximum pressure at which the riser/pipeline system shall be operated during incidental (i. with riser connectors at each end. 276 Safety class: The concept adopted herein to classify the criticality of the riser system. ball joints. 256 Pressure. 273 Riser tensioner stroke: The total upward and downward vertical movements of the riser relative to the floater. minimum: The local minimum internal pressure in the riser. (normally performed as hydrostatic testing). all integrated riser components and corrosion protection system. 265 Ratcheting: Accumulated plastic deformation during cyclic loading. 252 Pressure. System test: The surface internal pressure or local internal test overpressure applied to the riser or riser section during testing after completion of the installation work to test the riser system for tightness. Tension variations are controlled by the stiffness of the unit. 261 Pressure. Injection risers transport fluids to the producing reservoir or a convenient disposal or storage formation. Incidental pressure is referred to the same reference height as the design pressure and includes the situations where the riser is subject to surge pressure. this is the system which ensures that. tension joints. Maximum Allowable Operating (MAOP): The maximum pressure at which the riser/pipeline system shall be operated during normal operation. 269 Riser disconnect: The operation of unlatching of a riser connector. The (characteristic) resistance is normally based on a defined fractile in the lower end of the distribution function for the resistance. 253 Pressure. 275 Risk analysis: Analysis including a systematic identification and categorisation of risk to people. surge: The pressure produced by sudden changes in the velocity of fluids inside the riser.. 271 Riser pipe (riser tube): The pipe.24m) lengths. 262 Pressure. incidental pressure or test pressure. to 50 ft. 259 Pressure regulating system: For export risers and in relation to pipelines. 257 Pressure. 272 Riser system: A riser system is considered to comprise the riser.e. 268 Riser component: Any part of the riser system that may be subjected to pressure by the internal fluid. irrespective of the upstream pressure. tees. 264 Production/injection riser: Production risers transport fluids produced from the reservoir. propagating: The lowest pressure required for a propagating buckle to continue to propagate. local: The internal pressure at any point in the riser for the corresponding design pressure. The maximum allowable operating pressure is defined as the design pressure less the positive tolerance of the pressure regulating system. October 2010 Page 8 Section 1 251 Pressure. 277 Safety class resistance factor: Partial safety factor multiplied on the resistance reflecting the safety class. Riser joints are typically provided in 30 ft. ensures that the allowable incidental pressure is not exceeded. DET NORSKE VERITAS . flexjoints. The production riser may be used for well workovers. bends. A conservative estimate is to assume zero. “pup joints”. independent of the pressure regulating system. reducers and valves. connectors. characteristic: The nominal value of a strength parameter to be used in determination of design resistance.DNVOSF201 Dynamic Risers. the pressure at the reference height plus the static head of the transported/test medium due to the difference between the reference height and the height of the section being considered. This is equal to the minimum pressure at the reference height plus the static head of the fluid. a set pressure is maintained at a given reference point. This includes items such as flanges. For example. may also be provided to ensure proper spaceout.14m to 15. Shorter joints. The maximum allowable incidental pressure is defined as the maximum incidental pressure less the positive tolerance of the pressure safety system. which forms the principal conduit of the riser joint. telescopic joints. i. incidental: The maximum internal pressure that is unlikely to be exceeded during the duration/life of the riser or the maximum permitted internal pressure due to incidental operation of the riser. the riser pipe is the conduit for containing the production fluid flow from the well into the surface tree. slick joints.e. 260 Pressure safety system: The system which. initiation: External overpressure required to initiate a propagating buckle from an existing local buckle or dent. 254 Pressure. 267 Resistance. (9. 255 Pressure. 263 Process shutdown: A controlled sequence of events that ensures that the well is secured against accidental release of hydrocarbons to the environment. transient) operation. 266 Resistance: Capability of a structure or part of a structure to resist load effects also noted strength or load carrying capacity. 274 Riser tensioner system: A device that applies a tension to the riser string while compensating for the relative vertical motion (stroke) between the floater and riser.
tidal variations. This factor is used to account for the increase in the stresses caused by geometric stress amplifiers. which occur in the riser component. submerged weight or effective weight). The central safety factor is the ratio between a resistance and the load effect. 287 Submerged weight: Weight minus buoyancy (commonly referred to as weight in water. 279 Service life: The length of time assumed in design that a component will be in service. October 2010 Section 1 Page 9 278 Serviceability: A condition in which a structure is considered to perform its design function satisfactorily. astronomical tide (LAT). and which can be retrieved in severe environmental conditions. 288 System Effects: System effects are relevant in cases where many riser pipe sections are subjected to similar loading conditions. 285 Stress Concentration Factor (SCF): Equal to the local peak alternating principal stress in a component (including welds) divided by the nominal alternating principal stress near the location of the component. net lift. and potential structural failure may occur in connection with the lowest structural resistance among riser pipe sections. The determination of the splash zone includes evaluations of all relevant effects including wave height. to control curvature and reduce local bending stresses. wave diffraction effects. which is essentially kept straight and tensioned in all parts. subsidence and vertical motions of the riser in the splash zone. 291 Tubing: Pipe used in wells to conduct fluid from the well's producing formation into the subsea tree or to the surface tree. essentially marine/drilling risers and completion/workover risers. Safety and Environment 286 Stress joint: A specialised riser joint designed with a tapered cross section.DNVOSF201 Dynamic Risers. 283 Specified weather window: Limits to environmental conditions specified in operation manual. by applying a top tension to it. wet weight. 292 Water Level. wind and pressure induced storm surge and settlements and subsidence if relevant. 294 Wellbore annulus: Annular space between the production tubing and the well casing. Also named apparent weight.5 % elongation of the specimen gage length. The tidal range is defined as the range between the highest astronomical tide (HAT) and lowest DET NORSKE VERITAS . 290 Tensioned riser: A riser. The mean water level (MWL) is defined as the mean level between HAT and LAT. Maximum Still Water Level Highest astronomical tide Mean water level MWL) ( Lowest astronomical tide LAT) ( (SWL) (HAT) Storm surge 280 SN fatigue curve: Stress range versus number of cycles to failure. 295 Working Stress Design (WSD): Design method where the structural safety margin is expressed by one central safety factor for each limit state. The design maximum still water level (SWL) is to include astronomical tidal influences. Fabrication and Installation Dynamic Positioning Engineering Criticality Assessment Factory Acceptance Tests Frequency Domain Fatigue Limit State Failure Mode Effect Analysis Floating Production System Heat Affected Zone Hazard and Operational Analysis Hydrogen Induced Pressure Cracking High Integrity Pressure Protection System Health. Astronomical tidal range 282 Specified Minimum Yield Stress (SMYS): The minimum yield strength (stress) at room temperature prescribed by the specification or standard under which the material is purchased. The tensile stress at 0. 281 Specified Minimum Tensile Strength (SMTS): The minimum tensile strength (stress) at room temperature prescribed by the specification or standard under which the material is purchased. D. 289 Temporary riser: A riser which is used intermittently for tasks of limited duration. Figure 15 Definition of water levels 293 Wave Frequency (WF) motion: Motion of the floater at the frequencies of incident waves. Abbreviations and Symbols D 100 Abbreviations ALS BOP CMn CRA CTOD DDF DFF DFCGF DFI DP ECA FAT FD FLS FMEA FPS HAZ HAZOP HIPC HIPPS HSE Accidental Limit State Blow Out Preventer Carbon Manganese steel Corrosion Resistant Alloys Crack Tip Opening Displacement Deep Draft Floater Design Fatigue Factor Design Fatigue Crack Growth Factor Design. 284 Splash zone: The external region of the riser that is periodically in and out of the water. settlements.
October 2010 Page 10 Section 1 IM LF LRFD MQL MWL NDT RFC SCR QA QC QL QRA SCF SLS SML SMTS SMYS SRA SWL TD TRB TLP ULS UO UOE VIV WF WSD Installation Manual Low Frequency Load and Resistance Factor Design Material Quality Level Mean water level Non Destructive Testing Rain Flow Counting Steel Catenary Riser Quality Assurance Quality Control Material Quality Level Quantitative Risk Analysis Stress Concentration Factor Serviceability Limit State Seamless Pipe Specified Minimum Tensile Strength Specified Minimum Yield Stress Structural Reliability Analysis Still Water Level Time Domain Three Roll Bending Tension Leg Platform Ultimate Limit State Pipe fabrication process for welded pipes Pipe fabrication process for welded pipes. Plastic collapse pressure Propagating pressure Vector of resistances stress range at knee of bilinear SN curve time Pipe wall thickness. see section 5 m m1 ME MF Mk N ni Ni O pb pc pd pe pel pi pinc pld pli pmin D 200 A Ai Ae As D Dfat Di Dmax Dmin E Symbols Cross section area π (D − 2 ⋅ t )2 4 π External cross sectional area D 2 4 Internal fluid area π (D − t ) ⋅ t Pipe steel cross section area Nominal outside diameter. e.t3 DET NORSKE VERITAS . Accumulated fatigue damage or Miner sum D2tnom =Nominal internal diameter Greatest measured inside or outside diameter Smallest measured inside or outside diameter Young's Modulus pp ppr Rk SSW t t1.DNVOSF201 Dynamic Risers. expanded Vortex Induced Vibrations Wave Frequency Working Stress Design m2 M MA Md max Md f0 fy fu fk g g(t) h Hs log(a ) log(a1 ) log(a2 ) D max − D min D Yield strength to be used in design Tensile strength to be used in design Material strength Ovality. Gravity acceleration Generalised load effect Height from the riser section to the reference point for design pressure Significant wave height intercept of SN curve intercept of right leg of bilinear SN curve intercept of left leg of bilinear SN curve inverse SN curve slope inverse SN curve slope for right leg of bilinear SN curve inverse SN curve slope for left leg of bilinear SN curve Bending moment Bending moment from Accidental loads Design bending moment Maximum design bending moment.g. t2. in short term sea state Bending moment from Environmental loads Bending moment from Functional loads Plastic bending moment resistance Axial force in pipe wall ("true" force) (tension is positive) Number of stress blocks Number of stress cycles to failure at constant amplitude Out of roundness. Dmax – Dmin Burst resistance pressure Collapse pressure Design pressure at reference point External pressure Elastic collapse pressure Internal pressure Incidental pressure Local internal design pressure Local incidental pressure Local minimum internal pressure taken as the most unfavourable internal pressure plus static head of the internal fluid.
in short term sea state Design effective tension (force) Plastic axial force resistance Wave peak period True wall tension Wave zeroupcrossing period D 300 αc αfab αU γA γc γE γF γm γSC λn ν η ρe ρi Greek Characters Flow stress parameter accounting for strain hardening Manufacturing process reduction factor Material quality factor Load factor for accidental loads Condition factor Load effect factor for environmental loads Load effect factor for functional loads Resistance factor to take into account uncertainties in material properties Resistance factor dependent on safety class (consequence of failure) nth order spectral moment Poisson’s ratio for pipe wall material usage factor Density of external fluid (e. October 2010 Section 1 Page 11 tcorr tfab Internal and external corrosion allowance Absolute value of the negative tolerance taken from the material standard/specification of the pipe Nominal wall thickness of pipe (uncorroded). design) temperature Maximum design effective tension.g.F Te max Ted Ted Tk Tp Tw Tz DET NORSKE VERITAS . as specified on the drawing/specification Effective tension from Accidental loads Effective tension from Environmental loads Effective tension from Functional loads Effective tension (axial force) (Tension is positive).g.E Te. e. wave period or calculation (operating.DNVOSF201 Dynamic Risers. sea water) Density of internal fluid (contents) tnom Te.A Te.
The choice of safety class should also include consideration of the expressed safety objective. − the impact on the environment shall be reduced to as low as reasonably possible (ALARP).DNVOSF201 Dynamic Risers. If no policy is available. All companies have policy regarding human aspects. installation. one could also start with a risk assessment. it can be a point of discussion as to whether this is being accomplished in the actual project. measurable requirements. requalification.note  B. or if it is difficult to define the safety objective. The risk assessment could identify all hazards and their consequences. . and then enable backextrapolation to define acceptance criteria and areas that need to be followed up more closely. but more detailed objectives and requirements in specific areas may follow them. environment and financial issues. commissioning. General A 100 Objective 101 The purpose of this section is to present the safety philosophy and corresponding limit state design format applied in this standard. These are typically on an overall level. A 200 Application 201 This section applies to all risers that are to be built in accordance with this standard. Safety Philosophy B 100 General 101 The objective of this standard is that design. operation. Statements such as those above may have implications for all or individual phases only. It is therefore recommended that the overall Safety Objective be followed up by more specific. Typical statements can be: − all work associated with the transportation. 102 The integrity of a riser system constructed to this standard is ensured through a safety philosophy integrating the different aspects illustrated in Figure 21. planned and implemented covering all phases from conceptual development until retrieval or abandonment. C 100 C 200 C 300 C 400 C 500 C 600 General Objective Application Safety Philosophy General Safety Objective Systematic review Fundamental requirements Operational considerations Design Principles Quality Assurance and Quality System Design Format Basic Considerations Safety Class Methodology Design by LRFD Method Design by WSD Method Reliability Based Design Design by Testing Safety Objective Systematic Review Fundamental Operational Requirements Considerations Design Principles Quality Assurance Figure 21 Safety hierarchy B 200 Safety Objective 201 An overall safety objective shall be established. materials. October 2010 Page 12 Section 2 SECTION 2 DESIGN PHILOSOPHY AND DESIGN PRINCIPLES Contents A. repair. the structural failure probability is reflected in the choice of safety class. operation and maintenance of the riser system shall be such as to ensure that no single failure will lead to lifethreatening situations for any person. They are typically more relevant for the work execution and specific design solutions. B 100 B 200 B 300 B 400 B 500 B 600 B 700 C. fabrication. and abandonment of riser systems are safe and conducted with due regard to public safety and protection of the environment. installation/ retrieval. A 100 A 200 B. The section also provides guidance for extension of this standard in terms of new criteria etc. or to unacceptable damage to material or the environment. Having defined the Safety Objective. In this standard. − no releases of fluid content will be accepted during operation of the riser and pipeline system. Guidance note: A.Guidance .end .of . DET NORSKE VERITAS . These policies should be used as a basis for defining the Safety Objective for a specific riser system.
the materials and products shall be used as specified in this standard or in the relevant material or product specification . for drilling and/or workover operations. Spar).of .g. bullheading etc. inspection and maintenance philosophy. for production/injection and export/import of fluids and temporary risers for drilling/workover where it is not allowable to disconnect in extreme conditions (e. This may provide an estimation of the overall risk to human health and safety. transportation and operation shall be carried out by personnel having the appropriate skill and experience.end . operated and maintained in such a way that: with acceptable probability. fabrication and operation shall be communicated in an understandable manner to avoid misunderstandings. the criticality of the planned operation and previous experience with similar systems or operations. and − consequence and risk assessment. 502 Safe operation of a riser requires that: the designer shall take into account all realistic conditions under which the riser will be operated. floater interfaces. and comply with. such that necessary remedial measures can be taken. However. Other methodologies for identification of potential hazards are Failure Mode and Effect Analysis (FMEA) and Hazard and Operability studies (HAZOP). risers shall be designed by appropriately qualified and experienced personnel .g. e.note  the riser shall be adequately maintained including inspection and preservation when applicable . the following requirements apply: the design shall be in compliance with this standard. it will sustain all foreseeable load effects and other influences likely to occur during the service life and have adequate durability in relation to maintenance costs.Guidance . relevant information between personnel involved in the design. at least at an overall level to identify critical scenarios that might jeopardise the safety and reliability of a riser system. manufactured. 402 In order to maintain the required safety level. the riser shall be operated in accordance with the design basis and the installation and operating manuals. and with appropriate degree of reliability. including design and fabrication. fabrication. B 500 Operational considerations 501 Operational requirements are system capabilities needed to meet the functional requirements. shall be assessed. Permanent risers are normally designed to stay connected and operate when subjected to the extreme environment. A temporary riser may be designed to be disconnected. for the environment and for assets and financial interests. October 2010 Section 2 Page 13 B 300 Systematic review 301 A systematic review or analysis shall be carried out at all phases in order to identify and evaluate the consequences of single failures and series of failures in the riser system. Operational considerations include matters which designers should address in order to obtain a design that is safe and efficient to install. . limits for safe operations. design reviews shall be carried out where all contributing and affected disciplines (professional sectors) are included to identify and solve any problems. It should be noted that legislation in some countries requires risk analysis to be performed. operating limits may be introduced for some temporary conditions.g. Operational requirements include operational philosophy. having due regard to its service life and its cost. B 400 Fundamental requirements 401 A riser shall be designed. adequate supervision and quality control shall be provided during manufacture and fabrication. relative vertical motions between riser and floater (stroke). Temporary riser system operational parameters normally are closely monitored at all times to ensure that the riser is being operated within prescribed limits. floater offset. fabricated. and − temporary risers. e. it will remain fit for the use for which it is intended. retrieved or hungoff when the operating limit for the riser is about to be exceeded. DDF. manufacture. Section 9. 302 The Operator shall determine the extent of risk assessments and the risk assessment methods. risers run and retrieved many times during their service life. top tension. The operational parameters may include parameters such as internal pressure and density. The extent of the review or analysis shall reflect the criticality of the riser system. wave height. The consequences include consequences of such events for people. shut down. The extent of the verification and the verification method in the various phases.g. inservice operations. handling. − assessment of probabilities of failure events. environment and assets and comprises: − hazard identification. flex joint/ball joint angle and stress joint stresses. TLP. Reference is made to recognised standards for personnel qualifications . riser installation and retrieval. verification shall be performed to check compliance with provisions contained herein in addition to national and international regulations.g. Guidance note: Risers generally fit into two main operational types: − permanent risers. e. manufacture. risers installed and left for (many) years until subsequent retrieval. on site and during operation . see Section 8. floater motions and environmental limits. the operations personnel shall be aware of. DET NORSKE VERITAS . see e. operate and maintain. Guidance note: A methodology for such a systematic review is quantitative risk analysis (QRA). − accident developments.DNVOSF201 Dynamic Risers.
fatigue cracks. the location of the part of the riser that is being designed. angles. testing. and be robust with respect to use. . Design Format C 100 Basic Considerations 101 The design objective is to keep the failure probability (i. stresses. all subsea valves should be closed before the riser system is removed. The LRFD method allows for a more flexible and optimal design with uniform safety level and is considered superior to the WSD method. operation and maintenance activities to assist compliance with the requirements of this standard. 602 The riser system including riser pipe and interfaces. structural resistance and resistance against material degradation.note – Guidance note: ISO/CD 136287 give guidance on the selection and use of quality systems.of . If riser recovery is required following an emergency disconnect event. the design should facilitate monitoring of its behaviour in terms of tension. such as riser installation including connection.of . see C 300 Working Stress Design (WSD) method. which cannot be designed according to this principle.end . maintenance. The riser system shall (on a component level if relevant) be classified into one or more safety classes based on the failure consequences. probability of exceeding a limit state) below a certain value. fluid category. or the weather suddenly and unpredictably deteriorates beyond the riser's operating threshold.Guidance . the riser joints and components shall be made such that fabrication can be accomplished in accordance with relevant recognised techniques and practice. be designed “fail safe”. erosion and wear. DET NORSKE VERITAS . The WSD format is included as a more easytouse conservative alternative. design of structural details and use of materials shall be done with the objective to minimise the effect corrosion. i.DNVOSF201 Dynamic Risers.e. details and components. corrosion etc. which are normally limited due to weather conditions. structural safety of the riser is ensured by use of a safety class methodology. There are two levels of riser disconnection: normal or planned disconnection and rapid or emergency disconnection. Consideration is to be given in the design to possible early detection of failure or redundancy for essential components. end . provide adequate access for inspection. Rapid or emergency disconnection of the riser system may be necessary if floater or well system emergencies occur. the floater stationkeeping/tensioning system fails. testing (fullscale or model) conducted in accordance with valid experimental methods may be used to determine or verify riser system load effects. 702 A quality system shall be applied to the design. retrieval including disconnection and pressure testing. see C 600 Guidance note: The LRFD method separates the influence of uncertainties and variability originating from different causes by means of partial safety factors. vibrations. 102 The following design methods may be applied: Load and Resistance Factor Design (LRFD) method. riser mechanical components shall.note – C. special case design problems. whether the riser is in operating or temporary state.of . see C 400 Reliability analysis. retrieval. see C 200. manufacture and fabrication phases and quality assurance during all relevant phases. competence of personnel performing the work and verification activities during the design.Guidance . manufacturing. All relevant failure modes for the riser shall be identified and it shall be verified that no corresponding limit state is exceeded. October 2010 Page 14 Section 2 Both temporary risers and permanent risers normally have certain operations. All equipment should be designed to be failsafe to prevent the escape of fluids from the riser/well bore/pipeline to the environment during disconnection. abrasion. C 200 Safety Class Methodology 201 This standard gives the possibility to design risers with different safety requirements. As an alternative or supplement. for conditions where limited experience exists and for (re) calibration of safety/usage factors. fabrication.note  B 600 Design Principles 601 In this standard. as far as practicable.Guidance . B 700 Quality Assurance and Quality System 701 The design format within this standard requires that the possibility of gross errors (human errors) shall be prevented by requirements to the organisation of the work. wear. permit simple and reliable installation.e. see C 500 Design by testing. shall be designed according to the following basic principles: the riser system shall satisfy functional and operational requirements as given in the design basis. replacement and repair. The WSD method adopted herein addresses the same limit states as the LRFD but accounts for the influence of uncertainty in only a single usage factor. The safety class of a riser depends on: the hazard potential of the fluid in the riser. depending on the safety class to which the riser belongs. the riser system shall be designed such that an unintended event does not escalate into an accident of significantly greater extent than the original event. end . Reliability analysis is mainly considered as applicable to unique.
This is implicit by the concept of safety classes defined in Table 23. 4) Risers that are pressurised in temporary condition may require to be treated as inservice risers. Flammable and toxic substances. Contents not specifically identified shall be classified in the category containing substances most similar in hazard potential to those quoted. significant environmental pollution or very high economic or political consequences. High 205 The safety class is a function of the riser status (phase) and location class. ethane. retrieval and hangoff.DNVOSF201 Dynamic Risers. significant environmental pollution or very high economic or political consequences. If no such analyses are performed. 204 Riser design shall be based on potential failure consequences. transportation. bottom connection) and system pressure test performed with incompressible medium is classified as safety class low. For normal riser use. 2) Temporary conditions include handling. singlephase gas which is mainly methane. which could have an adverse effect on the environment if released. The extent of location class 2 should be based on appropriate risk analyses. installation. failure modes). 203 The riser system shall be classified into a location class 1 and 2 as defined in Table 22. E Location class 1 2 NA NA Low Normal Normal High NOTES 1) Testing like overpull to test connection (e. which are gases at E ambient temperature and atmospheric pressure conditions and which. If the category is not evident. The operator shall specify the safety class to which the riser shall be designed. propane. argon and air. are conveyed as gases or liquids. carbon dioxide. 5) Riser status (phase) Testing1) Temporary with no pipeline/well access2) Inservice with pipeline/well access Riser content Fluid category A. Table 24 Normal classification of safety classes 3). and chlorine. toxic liquids and other liquids. butane.g. e. C 300 Design by LRFD Method 301 The fundamental principle of Load and Resistance Factored Design (LRFD) method (also denoted partial safety factor method) is to verify that factorised design load effects do not exceed factored design resistance for any of the considered limit states (i. Table 22 Classification of locations Location 1 2 Description Area where no frequent human activity is anticipated The part of the riser in the near platform (manned) area or in areas with frequent human activity. For conditions where failure implies risk of human injury. Guidance note: This separation of loads is done in order to cope with sources of uncertainties in a rational way. Typical examples would be hydrogen. pressure load effect (static) . the most hazardous category shall be assumed. B Flammable and/or toxic substances which are liquids at ambient temperature and atmospheric pressure conditions. 3) Riser with nonflammable content but under pressure may require to be classified as safety class Normal. For operating conditions where failure implies high risk of human injury. ammonia. methane (not otherwise covered under category D). ethylene.C Location class 1 2 Low Low Low Low Low Normal Fluid category B Location class 1 2 Low Low Low Low Normal Normal Fluid category D. uncertainties in the 302 In the LRFD approach it is distinguished between: DET NORSKE VERITAS . natural gas liquids.. petroleum products. landing. environmental load effects (mainly dynamic) and accidental load effects. the safety classes in Table 24 apply. October 2010 Section 2 Page 15 202 Fluids in the riser system shall be categorised according to their hazard potential as given in Table 21. 4). 5) If deemed necessary. a minimum horizontal distance of 500 m may be adopted. functional load effects (static) .e. connecting. Typical examples would be nitrogen. Other classifications may exist depending on the conditions and criticality of the riser. a riser can always be designed to the requirements of a more strict safety class. D Nontoxic. disconnection. Typical examples would be oil. Table 23 Classification of safety classes Safety class Low Normal Definition Where failure implies low risk of human injury and minor environmental and economic consequences. C Nonflammable substances which are gases at ambient temperature and atmospheric pressure conditions. liquefied petroleum gas. Table 21 Classification of fluids Category Description A Typical nonflammable waterbased fluids.g.
e. γ F ⋅ SF . The usage factor is also named Allowable Stress factor or Design Factor in some WSD codes and standards.end .γF ⋅SF.Guidance .note  303 The general LRFD safety format can be expressed as: g (SP . .1) 305 The load effects and resistance in this standard are usually given as percentile values (i. October 2010 Page 16 Section 2 environmental load effects are typically larger compared to those in pressure or functional load effects implying a higher safety factor . For design criteria where the load effects and resistance can be separated the LRFD format can be written in the more familiar format: Rk Sd (SP . Rk . They shall be based on reliable data. η is the usage factor and g( •) is the generalised load effect as discussed for the LRFD safety format. The resulting design formulas provide design criteria as a totality of model uncertainty. The following comments apply: the load effect factors and resistance factors depend on the limit state category identical load effect factors will apply to limit states and safety classes. . failure consequence) Resistance factor to account for material and resistance uncertainties Resistance factor to account for special conditions Time (2. return period values for load effects) of the respective probability distributions. Further. Guidance note: The characteristic resistances in this standard do not necessarily reflect either mean values or certain percentile values.γSC.Guidance .Guidance . γm. g(•)<1 implies a safe design and g(•)> 1 implies failure. bias loads etc.e. care shall be taken when recalibrating these formulas to ensure this totality. Hence. (e. 403 The general WSD design format can be expressed as: g( S. The usage factor. γE ⋅ SE . due to combined DET NORSKE VERITAS .2) 404 where S is the total load effect. the set of resistance factors are adapted to the particular failure mode being considered and safety class.Guidance .g.end . γc is applied where appropriate in order to account for conditions with specific load effects or resistances.of .of .note  an additional safety factor.end . The timedependent generalised load effect g( •) defined above covers the general case for combined loading.of . . t) ≤ 1 (2. γ A ⋅ SA .note  C 400 Design by WSD Method 401 The Working (allowable) Stress Design (WSD) method is a design format where the structural safety margin is expressed by one central safety factor or usage factor for each limit state. SP SF SE SA γF γE γA Rk γSC γm γc t = = = = = = = = = = = = Pressure loads Load effect from functional loads ( vector or scalar) Load effect from environmental load ( vector or scalar) Load effect from accidental loads ( vector or scalar) Load effect factor for functional loads( vector or scalar) Load effect factor for environmental loads Load effect factor for accidental loads Generalised resistance ( vector or scalar) Resistance factor to take into account the safety class (i. may be interpreted as an inverted weighted product of partial safety factors. γc .note  304 The acceptance criteria presented in this standard are calibrated using a reliabilitybased methodology for the different safety classes. It is emphasised that S is the total load effect (scalar or vector). Rk is the resistance.note  The generalised load effect g( •) is discussed in more detail in section 4.DNVOSF201 Dynamic Risers. γE ⋅ SE. . R k .Guidance . Resistance factors typically account for variability in strength and basic variables including the effect of dimensional tolerances and model uncertainties due to incomplete resistance model. Guidance note: The usage factor accounts for the integrated uncertainty and possible bias in load effects and resistance. η.end .t) ≤1 g(•) is the generalised load effect. Guidance note: g(•) is a function of time for systems exposed to time varying excitations. using recognised statistical techniques.end .γA ⋅SA.) ≤ γSC ⋅ γm ⋅ γ c 402 The WSD method adopted herein applies explicit design checks similar to the LRFD method but accounts for the influence of uncertainty in only a single usage factor.of .of . η. in case of prevailing system effects where many pipe sections are exposed to the same loading) Guidance note: Load effect factors typically account for natural variability in loads and model uncertainties due to incomplete knowledge or models leading to possible inaccurate calculation of load effects.
Statistical uncertainties with respect to a limited number of test results are to be included in the determination of model load effects or resistance. for touch down regions. 3) Per riser imply for the riser in each location class 4) The failure probabilities provided for SLS are not mandatory. as indicated by this standard.DNVOSF201 Dynamic Risers.3) Annual per riser Annual per riser Annual per riser Annual per riser Safety classes Low Normal 101 101102 103 104 High 102103 105 NOTES 1) The failure probability from a structural reliability analysis is a nominal value and cannot be interpreted as an expected frequency of failure.and accidental loads as relevant for the actual limit state and load case. Certain vital riser components and materials including seals may. the method complies with DNV Classification Note no. October 2010 Section 2 Page 17 action from pressure. due to their specialised and unproven function. data acquisition and processing simplifications and uncertainties. wear characteristics. mechanical characteristics. environmental. SLS are used to select operational limitations and can be defined according to the operator’s preference. Note that exceedence of a SLS conditions require a subsequent ALS design check. Guidance note: Load effect model tests are normally performed to determine the floater responses as wave induced motions and drift motions.6 or ISO 2394. 602 When implementing experimental test results into design. i. vortexinduced vibrations. load effect model tests should be considered to verify methods for predicting systems load effect (response) for concepts with little or no field experience and cases with high uncertainty in analysis models.note  Sd (S) ≤ ηR k . functional. 30. DET NORSKE VERITAS . Table 25 Acceptable failure probabilities 1) vs. the target safety level shall be based on the failure type and class as given in Table 25. 503 As far as possible.Guidance .Guidance . require extensive engineering and prototype testing to determine and confirmation of anticipated design performance including fatigue characteristics. The values are nominal values reflecting structural failure due to normal variability in load and resistance but excluding gross error. In general. a probabilistic design approach based on a recognised structural reliability analysis may be applied provided that: it is used for calibration of explicit limit states outside the scope of this standard. structural resistance and resistance against material degradation.end .end . all relevant deviations between the model test and reality shall be considered including: scaling effects. 2) The probability basis is failures per year for permanent conditions and for the actual period of operation for temporary conditions. fracture characteristics. If this is not feasible.of . model/testing simplifications and uncertainties. target reliability levels shall be calibrated against identical or similar riser designs that are known to have adequate safety based on this standard. 5) The FLS probability basis is failures per year.note  C 500 Reliability Based Design 501 As an alternative to the design formats specified in this standard.of .e. shielding effects. interference and soilstructure interaction i. Design by testing or observation of performance shall be supported by analytical design methods. often last year of service life or last year before inspection.e. the approach is demonstrated to provide adequate safety for familiar cases. and extension into new areas of application shall be supported by technical verification. safety class Limit state SLS4) ULS FLS5) ALS Probability bases 2. 502 Suitably competent and qualified personnel shall perform the structural reliability analysis. . corrosion characteristics. These tests may include tests for evaluation of hydrodynamic coefficients. Guidance note: It should be observed that the generalised load effect for the WSD formulation could be derived as a special case of the generalised load effect for the LRFD formulation. For design criteria where the load effect and resistance can be separated the WSD format can be expressed in the more familiar format: C 600 Design by Testing 601 Testing (fullscale or model) conducted in accordance with valid experimental methods may be used to determine or verify riser system load effects.. uncertainties with regard to longterm effects and failure modes.
(section C). e. i. coatings 6). waves and current. B 100 B 200 B 300 B 400 C. 6) Includes also absorbed water. contents and attachments Weight of internal fluid Applied tension for toptension risers Installation induced residual loads or prestressing Preload of connectors Applied displacements and guidance loads. 4) Earthquake load effects shall be considered in the riser design for regions considered being seismically active. General A 100 Objective 101 This section defines the loads to be considered in the design of riser systems.F 400) Table 31 gives some examples on how the various loads are categorised. DET NORSKE VERITAS .e. environmental (E) loads. A 100 A 200 A 300 B.of . A 300 Loads 301 Loads and deformations shall be categorised into four groups as follows: pressure (P) loads (section B). Table 31 Examples of categorisation of loads 1) Eloads Waves Internal waves and other effects due to differences in water density.note  A 200 Application 201 This section describes the loads to be applied in the adopted LRFD criteria. .: Mean offset including steady wave drift. slug flow. The loads are classified into different load categories. 2) For temporary risers. tubing. functional (F) loads. October 2010 Page 18 Section 3 SECTION 3 LOADS Contents A.Guidance . Guidance note: The aim of the load classification is to relate the load effect to the different uncertainties and occurrences. marine growth can often be neglected due to the limited duration of planned operations. 7) Possible dynamic load effects from Ploads and Floads shall be treated as Eloads. (section D) accidental (A) loads. Current Earthquake4) Ice3) Floater motions induced by wind.DNVOSF201 Dynamic Risers. as relevant Water Levels Floads Weight and buoyancy 6) of riser. both size and frequency. 3) Ice effects shall be taken into account in areas where ice may develop or drift. wind and current forces Wave frequency motions Low frequency motions Ploads7) External hydrostatic pressure Internal fluid pressure: hydrostatic. (section 5. 5) Slugs and pressure surges may introduce global load effects for compliant configurations. D 100 D 200 D 300 D 400 D 500 General Objective Application Loads Pressure Loads Definition Determination of Pressure Loads Pressure Control System Pressure Ratings Functional Loads Definition Determination of Functional Loads Environmental Loads Definition Environmental Load Condition Waves Current Floater Motion A. buoyancy modules. including active positioning of support floater Thermal loads Soil pressure on buried risers Differential settlements Loads from drilling operations Construction loads and loads caused by tools NOTES 1) Accidental loads.end . marine growth2). C 100 C 200 D. for a specific riser and floater may be defined by a risk analysis. static and dynamic5) contributions.g. anodes.
1*p d Pressure that is unlikely to be exceeded during life/period of operation of riser Drilling riser with both surface and subsea BOP stacks Production or injection riser used as extension of production casing Outer casing of dual casing production or injection riser with surface tree Tubing (single pipe) riser or flowline from subsea satellite well Import riser from subsea manifold Export/import riser from/to pipeline Surface shutin pressure with subsea valves open Surface shutin pressure with subsea valves open unless pressure can be reliably limited to a lower value by e. are loads that are strictly due to the combined effect of hydrostatic internal and external pressures see Table 31. Internal pressure definitions at riser surface (top)2) Design pressure.g. This should be taken into account when calculating the maximum allowable shutin pressure for the specific application . Table 32 Riser Type Drilling riser above subsea BOP stack Drilling riser with surface stack 102 The following internal pressure definitions apply at the surface (top) of the riser. B 200 Determination of Pressure Loads 201 It is the responsibility of the owner to determine design surface. however. maximum under –balance pressure) Zero (or if drilling underbalanced. Incidental pressure. Normally to be taken as 1. which determine an envelope of the (p. p d . where ρ i is the density of the internal fluid. Maximum surface shutin pressure with subsea valves open unless pressure can be reliably limited to a lower value Maximum surge pressure defined with low lifetime probability of occurrence. T. ρ) .Guidance . P. maximum underbalance pressure) Specified maximum annulus pressure1) or maximum sustained pressure allowed by regulation or company policy No requirement or specified pressure.and temperature based on the guidelines given above and Table 32.regime of the credible extreme values. pinc Maximum diverter line back pressure Design as an extension of the last casing string that will be drilled through. pd Zero Zero (or if drilling underbalanced. or the space between the inner casing and the tubing/work string in the case of a dualcasing riser. is the maximum surface pressure during normal operations. if used. Guidance note: Gas mixed with oil in the riser could reduce the hydrostatic internal pressure acting downstream of the closed valve. Assume subsea BOP will be closed before pressure rises higher. Pressure caused by nearsurface or nearbottom leak of inner tubing/casing maximum operating pressure. single pipe risers or outer riser of multitube risers).and incidental surface internal pressures together with internal content density. is the surface pressure that is unlikely to be exceeded during the life of the riser. pld = pd + ρi ⋅ g ⋅ h pli = pinc + ρ i ⋅ g ⋅ h (3.of . Annual average seawater density and mean sea levels shall be used to establish the external hydrostatic pressure.end . Pressure caused by nearsurface leak of shutin tubing (maximum). see Table 32: Design pressure. and g is the acceleration of gravity. The owner shall also specify surface operating pressure and minimum surface stresses with corresponding temperature and density. This applies to both outer riser and inner riser. It may be necessary to specify pressuretemperaturedensity values (p. ρ). a pressure reduction system (HIPPS) Maximum export/import pressure during normal operations Highest pressure that will be seen for an extended time Other riser type NOTES 1) Annulus refers to the space between the external riser pipe and the tubing/workstring/drillstring in the case of a singlecasing production/workover/drilling riser.1) Guidance note: The external pressure should not be taken as higher than the water pressure at the considered location corresponding to low DET NORSKE VERITAS . The content and pressure of the outer annulus for a dualcasing riser can normally be assumed constant and as specified. 2) Internal pressure may also be specified at subsea wellhead. Pressure Loads B 100 Definition 101 Pressure loads.DNVOSF201 Dynamic Risers. Maximum surge pressure or maximum well kill pressure. p inc. Incidental pressure.g.note  202 The local internal design pressure pld and local incidental pressure p li are determined based on the definitions given in B 100 as follows 203 The hydrostatic seawater pressure governs the external pressure on pipes directly exposed to seawater (e. T. h is the height difference between the actual location and the internal pressure reference point. they are considered separately in this standard. October 2010 Section 3 Page 19 B. Such loads are often included in the general class of functional loads. Surface pressure that will handle most well control situations.
The principal environmental parameters are waves.of . see Table 31. the equipment could be used at pressures above the marked pressure rating. flanges and other equipment shall have pressure rating not less than the surface pressure or local overpressure of the riser.Guidance . current and temperature.Guidance .end . E loading condition shall be considered. taking into account biological and other environmental phenomena relevant for the location.5 may be used as basis for establishing the environmental load conditions. 302 Wave conditions may be described either by a deterministic design wave or by stochastic methods applying wave spectra. Example is intended vessel offset. Pressure rated components like valves. are defined as loads that occur as a consequence of the physical existence of the system and by operating and handling of the system. Pressurecontrolling components (such as valve bore sealing mechanism and tubing plugs) may be isolated from the external ambient pressure under certain operating conditions. Other parameters of interest.of .of . currents and floater motions. F. valves in subsea gas service cannot be used in applications where the shutin pressure would exceed the maximum rated working pressure stamped on the equipment. Guidance note: Most spectra is described in terms of a few statistical wave parameters such as significant wave height. such as the maximum wave height Hmax and the associated wave period T Hmax can be derived from these. pH. Pressurecontrolling components on subsea oil wells may benefit from “external” downstream pressure due to hydrostatic head of the oil column in the riser. buoyancy. the most unfavourable with respect to the combined P. T p. . B 400 Pressure Ratings 401 The local differential pressure may form the basis for selection of pressure rated components. Such biological and environmental factors include water salinity.note  204 The hydrostatic annulus pressure governs the external pressure on the inner riser and tubing in multitube risers. the extreme value shall be used.DNVOSF201 Dynamic Risers. spectral peak period.Guidance .Guidance . In such cases. similar to internal pressure).note  C 200 Determination of Functional Loads 201 The following apply when the characteristic values of the Fload shall be determined: In the case of welldefined functional loads.e. D 200 Environmental Load Condition 201 Environmental phenomena that are relevant for the particular location and operations in question shall be taken into account. . The hydrostatic annulus pressure should be defined in terms of the density of the annulus content together with a reference pressure at a given location (i. Environmental Loads D 100 Definition 101 Eloads are loads imposed directly or indirectly by the ocean environment. In the case of variable functional loads. and can approach the riser from one or more directions simultaneously. spectral shape and directionality. without environmental or accidental load.note  B 300 Pressure Control System 301 A pressure control system may be used to prevent the internal pressure at any point in the riser system rising to an excessive level.of . In most cases.note  D. Examples are accurate data of the riser weight. oxygen content. Hs. October 2010 Page 20 Section 3 tide when external pressure increase the resistance and high tide when external pressure decreases the resistance. contents and applied tension. In the case of functional load caused by deformation. The pressure control system comprises the pressure regulating system. C. can vary in length and height. Such waves are irregular in shape. see DNVOSF101.end .. Guidance note: Riser components at any point along the riser should be designed for or selected to withstand the maximum differential pressure between internal and external pressure to which the components will be exposed to during operating conditions. D 300 Waves 301 Wind driven surface waves are a major source of dynamic environmental forces on the risers. Guidance note: The effect of marine growth on riser shall be considered.end . the expected value of the load shall be used. F. DET NORSKE VERITAS . see Table 31. pressure safety system and associated instrumentation and alarm systems. Examples of functional loads are listed in Table 31. . Sensitivity analyses should be performed to quantify criticality.end . Example is change in weight due to corrosion and effects due to marine growth (weight and effects on hydrodynamic loading). The principles and methods as described in DNV CN 30. The estimation of hydrodynamic load on risers subjected to accumulated marine growth shall account for the increase in effective diameter and surface roughness. Functional Loads C 100 Definition 101 Functional loads. .
For part of the riser below the splashzone linear wave theory is usually adequate in connection with irregular seastates.g.first order wave induced motions . Guidance note: This has relevance e. October 2010 Section 3 Page 21 303 The selection of appropriate wave theories depends on the actual application and link to assumptions used for adjacent structures e. D 500 Floater Motion 501 Floater offset and motions constitute a source of both static and dynamic loading on the riser.g. linear wave theory combined with wheeler stretching should be considered in addition to disturbed kinematics if relevant. Note however that disturbed kinematics e. pulldown/set down . for TLP’s). The main data regarding floater motions needed for riser designs are: static offset . wind induced current.g. for monohull vessels (FPSO's and Drill Ships) where large roll motions may introduce high bending moments due to beam swell sea in combination with wind driven head sea. floater motion transfer function Guidance note: Normally.note  D 400 Current 401 The design current velocity. reference is made to appendix F.of .Guidance . For further details. 304 Combination of wind driven waves and swell from different directions must be taken into account in design.end . low frequency motions . profile and direction shall be selected using the best statistics available.note  DET NORSKE VERITAS . density induced current. . for semisubmersibles and TLP’s may effect the kinematics close to the floater.end . wind and current loads . . The resulting current velocities shall include contributions from tidal current. storm surge current. wave frequency motions .mean offset due to wave.of . eddies that spin off from a circulating current and other possible current phenomena.due to the combined effect of mooring lines/tether constraints and floater offset (e. global ocean current.motions due to wind gust and second order wave forces .DNVOSF201 Dynamic Risers.g.Guidance .
C 100 C 200 General Objective Application Riser Analysis Procedure Extreme Combined Load Effect Assessment Fundamentals Generalised Load Effect Load Cases Design Based on Environmental Statistics Design Based on Response Statistics Global Analysis General Fatigue Analysis Riser system •Service life •Diameter(s) •Internal fluid data •Environmental data •Functional requirements •Operational requirements •Vessel data •Interfaces •Material selection Design basis Updated Riser Design Preliminary riser design Pressure design checks Combined Loading See Chapter 5 ALS SLS ULS FLS Fatigue C200 Loading Conditions Pressure P Functional F Environmental E Accidental A A. define generalised load effect for combined design criteria defined in Section 5. No Yes OK? Final riser design Interface loads Top tensions Figure 41 Design Approach B. establish loading conditions defined in B 300. F. perform preliminary riser design and static pressure. by FMEA. For accidental load and load effects see also Section 5. ULS and ALS limit states shall reflect the most probable extreme combined load effect over a specified design time period. Long Term Assessment of Extreme Load Effects F(gmax)=11/N 202 Section B considers extreme load effect assessment for SLS. establish extreme generalised load effect estimate based on environmental statistics B 400.. design checks (bursting. Focus is on assessment of global structural load effects in connection with design criteria specified in Section 5. hoop buckling and propagating buckling) specified in Section 5. conduct riser analysis using appropriate analysis models and methods defined in C. consider all relevant loads defined in Section 3. e. or on response statistics. Extreme Combined Load Effect Assessment B 100 Fundamentals 101 The characteristic load condition for SLS. A 100 A 200 A 300 B. B 500. check that no relevant limit state is exceeded. ULS and ALS while FLS is discussed in C 200. functional and environmental loads are provided below. A Dyear return period value corresponds to an annual exceedence probability of 1/D.g. Design Criteria Define Generalised Load Effect Safety Class Environmental Statistics Load cases Riser analysis Static Short Term Assessment of Extreme Load effects gmax(short term) Limit State Checks Dynamic Response Statistics Load cases A 200 Application 201 Combined load effects from pressure. g max <1 Adjust: −Design −Vessel −Operational requirements A 300 Riser Analysis Procedure 301 An overview of the (ULS)design approach is shown in Figure 41. The design approach may be summarised as: identify all relevant design situations and limit states. General A 100 Objective 101 The purpose of this section is to provide requirements for global analysis. the most probable extreme generalised load effect during D years is commonly also denoted the Dyear return period value.DNVOSF201 Dynamic Risers. HAZOP and design reviews. October 2010 Page 22 Section 4 SECTION 4 ANALYSIS METHODOLOGY Contents A. B 100 B 200 B 300 B 400 B 500 C. Guidance note: For permanent conditions. DET NORSKE VERITAS .
Guidance note: This implies that it is assumed that the design pressure (or minimum pressure) is likely to occur during an extreme environmental condition. The return periods shall be defined such that the probability of exceedence in the temporary state is no greater than that of the longterm operational state. which determines the generalised load effects to be used for limit states controls. (4.end . represent the range of operating conditions and functional applications.of . Guidance note: If more information is not available the following return period values may be applied: − a 100 year return period if duration in excess of 6 months. i.g. the 100 year return period value) of the generalised load effect. October 2010 Section 4 Page 23 . See also section 3.. Guidance note: The generalised load effect indicates the level of utilisation.of . 303 Environmental load effects generally depend on the applied Floads since Floads may influence the dynamical properties of the system (e. Ted denote design values for bending moment and effective tension. This approach may yield acceptable results when the design is driven by one dominating dynamic component.end . Rk is a vector of crosssectional capacities and Λ is a vector of safety factors (i. i. This approach will automatically account for the correlation between effective tension and bending moment DET NORSKE VERITAS . the acceptance criteria can be expressed by the following generic equation: B 300 Load Cases 301 The load cases form the basis for riser analysis.end . 103 For temporary operational conditions the load effect return period value depends on the seasonal timing and duration of the temporary period. Conservative shortterm estimates can be obtained by separate estimation of design values for effective tension and resulting bending moment disregarding correlation effects.end . depending on the duration of the operations and the consequences of exceeding the selected conditions.of .end . see Appendix C for details.1) 202 Where g(t) is the generalised load effect and M d . .Guidance note components and is hence capable of optimal design (i.2) 304 For operating extreme conditions for combined load effects the pressure should be taken as the design pressure or a minimum value whichever is the more conservative. ∆p. R k . An adequate set of load cases (loading conditions) should be examined in order to: reflect extreme combined load effects. In case of design based on environmental statistics the standard framework for response processing of results from time domain analyses can be directly applied for code checks. ∆p. − For temporary conditions with duration less than 3 days or operations which can be terminated within a 3 days window an extreme load condition may be specified and startup /shut down of the operation is then based on reliable weather forecasts. Furthermore. max g M max .e. see Section 5.Guidance note 302 Different conditions may be selected for various stages in the operation. g(t)< 1 imply a safe design and g(t)>1 imply failure. safety class and condition factors). and study sensitivities to the variation of critical parameters at different locations along the riser. Ted . The generalised load effect applies to design based on response statistics in establishing the long term probability distribution as well as shortterm extreme load assessment for design based on environmental statistics.) Sensitivity studies shall therefore be performed to identify the most unfavourable Fload with respect to combined load effects at critical locations 203 The code checks for combined loading is hence equivalent to extreme value prediction (e. . represent both permanent and temporary conditions .Guidance note  ( ) (4. An advantage of this formulation is that it is applicable to time domain. Such a generic formulation covers LRFD as well as WSD acceptance criteria for combined loading. represent all relevant limit states.DNVOSF201 Dynamic Risers.of . This will typically include application of response envelopes in case of regular wave analysis and statistical extreme value prediction in case of irregular wave analysis. respectively.3) where the indices “max” indicate extreme values.of . applied top tension and mass per unit length will influence the dynamic properties of the system. Λ ≤ 1 d 102 For permanent operational conditions a 100year return period (102 annual exceedence probability) apply. Λ) ≤ 1. − a 10 year return period for the actual seasonal environmental condition if duration is in excess of 3 days but less than 6 months. material.Guidance note  Guidance note: The importance of this formulation is that the combined time dependent action of bending moment and effective tension is transformed into a scalar process expressed by the generalised load effect.e. Ted ( t).Guidance note  B 200 Generalised Load Effect 201 For combined loading. . ∆p denotes the local differential pressure. allows for maximum utilisation).as well as frequency domain analyses. . g( t ) = g( M d ( t).e. R k .g.e. g max ≤ 1 (4.
g. systems with significant nonlinear response characteristics. A feasible approach for establishing long term response statistics is proposed in Appendix C. verification and/or calibration of results should be performed in the following cases: new concepts. Knowledge of governing nonlinearities for the DET NORSKE VERITAS . This is of special importance for regular wave analyses. splash zone loads in case of disturbed kinematics). mass. 106 Static analyses should be carried out using a full nonlinear approach. 104 The riser shall be discretised with sufficient number of elements to represent environmental loading and structural response and to resolve load effects in all critical areas. peaks in floater motion transfer function. peak period etc) and a given duration (e. Global Analysis C 100 General 101 Global riser analysis shall be conducted for the specified design cases. Several alternatives are available in subsequent dynamic analysis restarted from the static equilibrium configuration.g. numerical simulations or physical testing or a combination of these methods. environmental loading and soil mechanics to determine reliable load effects on the riser system.g. The main challenge is that the return period for the characteristic load effect is unknown due to the nonlinear dynamic behaviour of most riser systems. significant wave height.) The main purpose is to quantify model uncertainties. In particular.end . which may be subjected to severe bias for dynamically sensitive systems. .of . component modelling.g. Different combinations of wind. 103 The global riser model shall include the complete riser system considering accurate modelling of stiffness. A general guidance on global load effect analysis of risers is given in Appendix A. Consistent assessment of the Dyear generalised load effect will in general require a probabilistic description of the load effect due to the longterm environmental load on the riser system. The main challenge is to establish the longterm load effect distribution due to the nonlinear dynamic behaviour experienced for most riser systems. The load effect analysis may be based on analytical calculations. model discretisation.g. Guidance note: It has traditionally been common practice to adopt the most unfavourable load effect found by exposing the riser system to multiple stationary environmental conditions as the extreme load effect.end . Furthermore.g. disturbed wave kinematics. soil data. The principles for model validation as outlined in Appendix D should be adopted. This will in general lead to an inconsistent safety level for different design concepts and failure modes. strength of materials. 105 Sensitivity studies shall be performed to investigate the influence from uncertain system parameters (e. October 2010 Page 24 Section 4 B 400 Design Based on Environmental Statistics 401 Design criteria based on environmental statistics may be applied to establish characteristic load effects. support rational conservative assumptions and identify areas where a more thorough investigation is needed to achieve an acceptable modelling (e. . waves and current consistent with the environmental conditions at the actual site is normally applied. 403 Wave period variation shall be considered for regular and irregular wave analyses to identify most unfavourable loading condition. Each design condition is described in terms of a limited number of environmental parameters (e.of . waves and current yielding the same return period (e. appropriate drag and inertia coefficients for the selected method shall be applied. 102 The global analyses shall be based on accepted principles of static and dynamics analysis. hydrodynamic coefficients. period dependencies in load intensity (e. 36 hours).Guidance note  Guidance note: Design based on response statistics is the more correct approach and should be considered when deemed important. calibration of computer model against physical testing) B 500 Design Based on Response Statistics 501 Design based on response statistics is generally the recommended procedure for consistent assessment of characteristic load effects. Treatment of nonlinearities is the distinguishing feature among available dynamic analysis techniques. The period variation shall be performed with due consideration of the following: statistical variation of wave period. see B 100 to check the relevant limit states for the riser system and establish component load effects and riser interface data. damping and hydrodynamic load effects along the riser in addition to top and bottom boundary conditions. eigenvalues of the riser system. The methodology in B 500 may be applied for verification and/or calibration purposes. and dynamically sensitive systems. 402 If the design is based on environmental statistics.DNVOSF201 Dynamic Risers. Acceptable results can however be expected for quasistatic systems with moderate nonlinearities.Guidance note  C. Time and/or frequency discretisation shall be verified to ensure that the desired accuracy is obtained. Guidance to computational strategies for shortterm assessment of extreme load effects is given in Appendix C. structural damping etc. the most severe directional combination of wind. A sufficient number of loading conditions in terms of stationary environmental conditions must be analysed in order to capture the extreme generalised load effects for all critical locations on the riser. corrosion allowance. 100 years) for the combined environmental condition are typically applied.
110 It shall be documented that the duration of irregular time domain analyses is sufficient to obtain extreme load effect estimates with sufficient statistical confidence. frequency domain analyses may be applied provided that the adequacy of such analyses is documented by verification against time domain analysis. Special FLS analyses for systems or parts of systems with highly nonlinear response characteristics (e. validation against irregular sea. Collision/interaction with other slender structures NA Linearised at static equilibrium position (stochastic linearisation in case of irregular excitation) Linearised at static equilibrium position NA Structure Geometric stiffness Nonlinear material Seafloor contact. Extreme load effect analyses should preferably be carried out by use of time domain analyses. October 2010 Section 4 Page 25 actual system as well as treatment of nonlinearities in established analysis techniques is crucial for the accuracy and hence the choice of adequate analysis strategy. Reference is made to Appendix A for a more detailed discussion. see Appendix E.g. This is DET NORSKE VERITAS . If frequency domain analysis is used. Global analysis. 202 The fatigue response due to the first two contributors may be calculated with the same methods as for extreme response calculation. The methodology as outlined in Appendix C may be applied. second order floater motions . For further details see Appendix D. regular wave analysis in time domain (design wave). irregular wave analysis in the frequency domain 109 The irregular wave analysis refers to modelling of water particle kinematics and floater motions. Finite element (FE) methods overview Nonlinearities Special loads Slug flow. 107 An overview of commonly used dynamic FE analysis methods is given in Table 41. 108 One or combinations of the following methods should be applied: irregular wave analysis in the time domain (design storm). time domain analysis shall be performed. the validation as specified in Table 43 should be considered for representative (critical) load cases.g. top tensioned risers ) Screening analyses. touchdown area of compliant configurations) Verification/validation of simplified methods (e. LTD. Variable hull contact Large 3D rotations Linearised at static equilibrium position Linearised Time domain (LTD) Frequency domain (FD) Table 42 Method NTD Typical analyses techniques versus applications Typical applications Extreme response analysis of systems with significant nonlinearities. in particular compliant configurations exposed to 3D excitation.DNVOSF201 Dynamic Risers. However. 111 Any use of simplified modelling and/or analysis techniques should be verified by more advanced modelling and/or analyses. thermal and pressure induced stress cycles vortex induced vibrations. main techniques for dynamic analysis is indicated in Table 42. FD) Extreme analysis of systems with small/moderate structural nonlinearities and significantly nonlinear hydrodynamic loading (e. collisions All modes of operations including connected. In particular. Table 43 Validation analysis methods overview Method for validation Nonlinear time domain analysis Time domain analysis Irregular wave analysis LTD Applied method Linearised time domain analysis Frequency domain analysis Regular wave analysis FD C 200 Fatigue Analysis 201 Fatigue analysis of the riser system shall consider all relevant cyclic load effects including: first order wave effects (direct wave loads and associated floater motions) . FLS analyses of systems with small/moderate nonlinearities of particular concern in case of combined WF and LF loading. Typical application of the Table 41 Method Environmental Loads Nonlinear Time domain (NTD) Morison loading Integration to actual surface elevation. running and hangoff must be considered if relevant.g.
g.response FD FD FD TD TD TD TD for combined WF+LF excitation Where : FD TD WF = Global frequency domain analysis = Global time domain analysis = Wave frequency LF NB RFC = Low frequency = Narrow band approximation = Rain flow cycle counting DET NORSKE VERITAS . Table 44 Fatigue analysis methods overview Fatigue damage assessment WFdamage LFdamage NB NB NB RFC RFC RFC RFC for combined WF+LF response Combined WF+LF damage Summation / bimodal Summation Summation Method of Analysis WF. using half the corrosion allowance in the cross section values for inservice assessment 204 Recommended procedures for shortterm fatigue damage calculation for commonly used global analysis strategies are given in Table 44. e. For further details see Appendix B. Sensitivity analysis is needed to map criticality and give input to DFI.DNVOSF201 Dynamic Risers.response LF. October 2010 Page 26 Section 4 203 Fatigue analyses normally apply nominal values.
DNVOSF201 Dynamic Risers.e. C 100 C 200 C 300 D. E 100 E 200 E 300 E 400 F. components. manufacture. Accidental Limit State (ALS) is a ULS due to accidental loads (i. 202 Requirements for materials. D 100 D 200 D 300 D 400 D 500 D 600 D 700 E. the riser pipes and connectors shall be designed for (not limited to) the potential modes of failures as listed in Table 51 for all relevant conditions expected during the various phases of its life. A 200 Application 201 This standard provides design checks with emphasis on ULS. SLS and ALS load controlled conditions. For operating condition this limit state corresponds to the maximum resistance to applied loads with 102 annual exceedence probability. Design of connectors and riser components are covered in Section 6. A 100 A 200 A 300 B. FLS. October 2010 Section 5 Page 27 SECTION 5 DESIGN CRITERIA FOR RISER PIPES Contents A. B 100 B 200 C. but not necessary be able to operate. F 100 F 200 F 300 F 400 General Objective Application Limit States Load Effects Design Load Effects Load Effect Factors Resistance Resistance Factors Geometrical Parameters Material Strength Ultimate Limit State General Bursting System Hoop Buckling (Collapse) Propagating Buckling Combined Loading Criteria Alternative WSD Format Displacement Controlled Conditions Fatigue Limit State General Fatigue assessment using SN curves Fatigue assessment by crack propagation calculations Inservice Fatigue Inspections Accidental Limit State Functional requirements Categories of accidental loads Characteristic accidental load effects Design against accidental loads limit states for pipes in riser systems. 203 Mill pressure test and system pressure test shall be performed in compliance with DNVOSF101. infrequent loads) Fatigue Limit State (FLS) is an ultimate limit state from accumulated excessive fatigue crack growth or damage under cyclic loading. General A 100 Objective 101 The section provides the general framework for design of riser systems including provisions for checking of DET NORSKE VERITAS . 302 As a minimum requirement. This limit state corresponds to criteria limiting or governing the normal operation (functional use) of the riser. Ultimate Limit State (ULS) requires that the riser must remain intact and avoid rupture. A. fabrication and documentation of riser pipe. equipment and structural items in the riser system are given in Section 7. Design principles for displacement controlled conditions are discussed in D 700. A 300 Limit States 301 The limit states are grouped into the following four categories: Serviceability Limit State (SLS) requires that the riser must be able to remain in service and operate properly.
note  102 Design bending moment for functional and environmental induced load effects: M d = γ F ⋅ M F + γ E ⋅ ME + γ A ⋅ M A where: MF ME MA = Bending moment from functional loads (5. Load Effects B 100 Design Load Effects 101 Design load effects are obtained by multiplying the load effect of each category by their corresponding load effect factor. Excessive Miner fatigue damage or fatigue crack growth mainly due to environmental cyclic loading. Normally.floater deck.DNVOSF201 Dynamic Risers. surface treefloater deck. e. coiled tubing.3) = Bending moment from environmental loads = Bending moment from accidental loads 103 Design effective tension for functional and environmental induced load effects: Ted = γ F ⋅ TeF + γ E ⋅ TeE + γ A ⋅ TeA True wall tension (i. Mechanical function of a connector during makeup/breakout. local buckling and hoop buckling Unstable fracture and gross plastic deformation Liquid tightness Global buckling ALS FLS Same as ULS and SLS Fatigue failure B.g. Large relative top displacements between riser and floater that are beyond the specified operational limits for top tensioned risers. see Appendix A: (tensile force is positive): Te = TW − p i A i + p e A e Where Tw = (5.Guidance . Unstable crack growth or rest ligament rupture or cross section rupture of a cracked component. Gross plastic deformation and hoop buckling of the pipe cross section and/or local buckling of the pipe wall due to the combined effect of external overpressure.g. risermooring line. slick joint and tensioner.2) pe Ai = = = where DET NORSKE VERITAS . directly or indirectly. Limiting size of fatigue cracks may be wall thickness (leakage) or critical crack size (unstable fracture/gross plastic deformation). ULS Mechanical function Bursting Hoop buckling (collapse) Propagating buckling Gross plastic deformation and local buckling Gross plastic deformation. Note that systems can be designed for exceeding displacement limits if the structural integrity is maintained. inclination of flex joint or ball joint.end .of . Te is given by. or by normal loads after accidental events (damage conditions). Large angular deflections that are beyond the specified operational limits. Specific examples are given below for bending moment and effective tension. axial stress resultant found by integrating axial stress over the crosssection) Internal (local) pressure External (local) pressure Internal crosssectional area pi (5. Gross plastic deformation (crushing) and/or buckling (collapse) of the pipe cross section caused by external overpressure only. F+E loads and A loads is not considered simultaneously in global analyses . Leakage in the riser system including pipe and components. TeF TeE TeA = Effective tension from functional loads = Effective tension from environmental loads = Effective tension from accidental loads Guidance note: Accidental loads are included in the above design load effects for completeness. Failure caused by accidental loads directly. e. axial force and internal overpressure. Overall column buckling (Euler buckling) due to axial compression (negative effective tension). October 2010 Page 28 Section 5 Table 51 Typical limit states for the riser system Limit State Category SLS Limit State Clearance Excessive angular response Excessive top displacement Failure definition/ Comments No contact between e. Propagating hoop buckling initiated by hoop buckling. Gross plastic deformation (rupture/crushing) of the pipe crosssection in combination with any local buckling of pipe wall (wrinkling) due to bending moment. stroke of telescope joint. surface equipment and drill floor. subsea treeseabed. Membrane rupture of the pipe wall due to internal overpressure only.e.1) 104 The effective tension. effective tension and bending moment.g. surface jumper. riserhull. riserriser.
Several combinations may have to be checked when load effects from several load categories enter one design check.3.g. see F.11) 1.4) NOTES 1) If the functional load effect reduces the combined load effects.Guidance .7) 204 Guidance note: t 1 is the minimum wall thickness and is relevant for design checks where failure is likely to occur in connection with a low capacity.g. Table 52 Limit state ULS FLS SLS & ALS C 200 Geometrical Parameters 201 The nominal outside diameter D applies in resistance calculations for all failure modes. October 2010 Section 5 Page 29 Ae = External crosssectional area the material resistance factor in Table 54 can be reduced to 1. γ E shall be taken as 1/1.note  102 Unless otherwise stated.0 t1 = t nom − t fab Operational condition (5. end . material resistance factor γ m to account for material and resistance uncertainties a condition factor γc to account for special conditions specified explicitly at the different limit states where relevant. see G.The following approximation may be applied for a stationary corrosive environment: t 3 = t nom − 0.e. The load effect factors shown in Table 52 shall be used wherever the design load effect is referred to for all limit states and safety class. Table 53 Low 1.of . 202 For burst and collapse pressure design checks (i.15 Guidance note: 205 Variation in pipe wall thickness over the design life of the riser system shall be considered in longterm fatigue damage calculations (i.05. installation etc) the pipe wall thickness shall be taken as: DET NORSKE VERITAS .8) For fatigue damage calculations prior to permanent operation (e. towout.Guidence .32) 1.5) C.note B 200 Load Effect Factors 201 The design load effect is used in the design checks. (see Section 2. γ F shall be taken as 1/1. t 2 is used for design checks governed by the external loading and failure is likely to occur in connection with an extreme load effect at a location with average thickness. see e. .26 Table 54 Material resistance factor γ m SLS & FLS 1. For ALS.14 High 1.DNVOSF201 Dynamic Risers.0 Eload effect γ E 2) 1.6) Otherwise t 2 = t nom − t corr (5.C): safety class factor γSC linked to the actual safety class and accounts for the failure consequence. Table 511. t1 = t nom − t fab − t corr where: tnom tfab tcorr = Nominal (specified) pipe wall thickness = Fabrication (manufacture) negative tolerance = Corrosion/wear/erosion allowance (5. Resistance C 100 Resistance Factors 101 The following resistance factors apply. the set of resistance factors can be defined by the owner. In cases. the set of safety factors depends on the frequency of occurrence and is to be defined from case to case. where a conservative estimate is applied. the resistance factors applicable to all limit states are specified in Table 53 and Table 54. 2) If the environmental load effect reduces the combined load effects.e.5 ⋅ t corr For SLS.1. operational condition).end . inplace. where the inherent uncertainty related to the accidental load is negligible and.04 Safety class resistance factor γSC Normal 1.0 Aload effect γA NA NA 1.0 ULS & ALS 1.0 1.0 1. An average representative pipe wall thickness may be applied in nominal fatigue stress calculations . 203 Resistances for all other limit states related to extreme loading shall be calculated based on wall thickness as follows: Installation/retrieval and system pressure test t 2 = t nom (5. D 200 and D 300) the resistance shall be calculated based on wall thickness as follows: Mill pressure test and system pressure test condition Load effect factors Fload effect γ F 1) 1. (5.of .
temp SMTS fu.10) Stress DeRating (MPa) f f k = min f y .temp tensile strength.00 Normal 0. during blown down in gas risers. applies for pipes manufactured by the UOE.end . compressive circumferential yield strength for hoop buckling. Table 56 Material strength factor α u Supplementary requirement U 1.end . This includes: yield strength.15 Compressive circumferential material strength 120 100 80 60 40 20 0 0 20 50 100 Temperature C 150 200 CMn f k = f y ⋅ α fab Longitudinal material strength (5. local buckling and propagating buckling limit states.note  Supplementary requirement U has a testing regime which shall ensure that SMYS is at least 2 standard deviations below the mean yield strength and that SMTS is at least 3 standard deviations below the mean tensile strength. fu. Note that αc is a function of the pressure among others. Unless otherwise documented.g. DET NORSKE VERITAS . The supplementary requirement ensures increased confidence in material strength. Beneficial effect on this reduction factor due to heat treatment is allowed if documented. see 302. the fabrication factor αfab in Table 57. see 302. October 2010 Page 30 Section 5 t 3 = t nom (5. i.note  Characteristic yield and tensile strength f y = SMYS − f y .DNVOSF201 Dynamic Risers.temp αU is the Specified Minimum Yield Stress at room temperature based on the engineering stressstrain curve. 180 160 140 25Cr 22Cr C 300 Material Strength 301 The characteristic material strength to be used in the resistance calculations fk is given by: Tensile circumferential material strength (5. α fab is a fabrication factor given by 305 and αc is a strain hardening factor given by 306. low temperature effects. UO or three roll bending (TRB) or similar cold deforming processes.Guidance .of .96 Guidance note: For reeling the effect of plastic straining after the pipe mill shall be evaluated and included in the material property.e.of . Guidance note: The increased utilisation may be applied for connectors made of forging and bolts provided an equivalent testing scheme is adopted . Possible influence on the material properties from the temperature shall be considered at temperatures above room temperature. which is reflected in a higher material strength factor αU .end . SMTS) is normally specified at room temperature. .temp Young's modulus.temp ⋅ α U Where SMYS ( ) Tensile strength f u = SMTS − f u.of . is the temperature derating factor for the tensile strength. thermal expansion coefficient.e. temp ⋅ α U ( ) fy.Guidance . e. is the material strength factor. 305 A fabrication factor α fab applies to the design 303 Derated material properties at design temperatures shall be established as input to the design and verified under manufacture.11) f k = fy ⋅ α C (5.note  302 The material strength (SMYS. u 1. should be considered when establishing mechanical and physical material properties.Guidance . . Further. see 304 304 The material selection may include selection of supplementary requirement U according to DNV OSF101. fy. is the temperature derating factor for the yield stress. is the Specified Minimum Tensile Strength at room temperature based on the engineering stressstrain curve. i. 22Cr Duplex or 25Cr Duplex stainless steel in Figure 51 below may be used. Table 55 Yield stress Figure 51 Derating values for yield strength Likewise. given in Table 56.9) Guidance note: If no other information on derating temperature effects of the yield strength exists the recommendations for CMn steel.12) Where fy and fu denote the characteristic yield and tensile strength given in Table 55.
For compliance.8 D 200 Bursting 201 Pipe members subjected to net internal overpressure shall be designed to satisfy the following condition at all cross sections: αc 1.DNVOSF201 Dynamic Risers. p b ( t) = 2 2⋅t f ⋅ min f y . − hoop buckling collapse criterion is formulated in terms of the minimum (t 1) rather than nominal (t 2) thickness.925 Tensile strength or seamless pipe 1. 1. Normally the incidental surface pressure. section 5.4 0. Pressure and functional loads normally govern wall thickness sizing for pipelines while extreme environmental loads and fatigue govern typical dynamic riser design. i. In addition a few minor differences exist.85 0. γp=1. The following comments apply to this standard in relation to DNVOSF101: − load combination a) in DNVOSF101. Design principles for displacement controlled conditions are discussed in D 700.4 + qh )(60 − D / t 2 ) / 45 0 (p ld − pe ) 2 q h = p b (t 2 ) 3 0 p ld is the local design pressure defined in Section 3.15) t is a “dummy variable” to be substituted by t1 or t 2 where relevant.: 102 This section provides design checks with emphasis on load controlled conditions.end .4 + q h ) β = (0. − the additional safety class resistance factors for pressure containment for compliance with ISO is not required for dynamic risers. October 2010 Section 5 Page 31 linear elastic and ductile materials. Further. see Section 3. u 3 D −t 1. p e is the external pressure and p b is the burst resistance given in D 200.note  306 αc is a parameter accounting for strain hardening and wall thinning given by: α c = (1 − β) + β ⋅ fu fy for D / t 2 < 15 for 15 < D / t 2 < 60 for D / t 2 > 60 for pld > pe else (5. p d .0 herein.20 1. Ultimate Limit State D 100 General 101 The riser pipe shall be designed against relevant modes of failure listed in Table 51.15 ⋅ (5.2 0.of . 202 The local incidental pressure. Table 57 Fabrication factor α fab Compressive strength for welded pipe UOE/ UO/TRB 0.00 where: p li pe = Local incidental pressure. p inc is taken 10% higher than the design pressure.16) DET NORSKE VERITAS . In general the same limit states apply for pipeline systems and dynamic riser systems but the governing failure modes differ due to different functional requirements between pipelines and risers.18. 1⋅ pd where: (5.e.04 (pli − pe ) ≤ 0 10 20 30 40 50 60 70 pb ( t1 ) γ m ⋅ γ SC (5. − anisotropy is not considered explicitly but the effect is implicit in the combined loading criteria for internal overpressure.20. p li is the maximum expected internal pressure with a low annual exceedence probability.13) ( 0 . see DNV0SF101. αc is not to be taken larger than 1. .18 The burst resistance pb is given by: D.Guidance .24 qh 1. − the propagating buckling criteria is similar but may be relaxed if the buckle is allowed to travel a short distance.D 300 is not required for dynamic risers.08 1.16 0 0. αc is for illustration purpose given in Figure 52 in case of (fu /fy ) = 1.00 accumulated plastic deformation is considered unlikely and “shakedown” can automatically be assumed Guidance note: A high degree of compatibility with the DNVOSF101 Submarine Pipeline Systems has been attempted where relevant.12 1.6 0. 103 If the design is based on: load controlled (LC) conditions design loads based on global riser analysis p li = p ld + 0. see Section 3 = External pressure D/t Figure 52 α c versus D/t ratio and pressure ratio qh for (fu/fy ) = 1.14) 1.
19) The plastic collapse pressure is given by: DET NORSKE VERITAS . 15 ⋅ +1 (p li − p e ) 3 γ m γ SC .note  t nom = t1 + t corr + t fab when the negative fabrication thickness tolerance is absolute. tfab.005 (0.18) Solution of the equation above can be found in DNVOSF101. The elastic collapse pressure (instability) of a pipe is given by: t 2 ⋅ E ⋅ D p el ( t) = 1− ν 2 3 (5. Hence. a corresponding decreased utilisation shall be applied. October 2010 Page 32 Section 5 p ld = Local internal design pressure. Note that the burst criterion is formulated in terms of the local incidental pressure rather than a local design pressure. For installation p min equals zero. Guidance note: The burst criterion is expressed in terms of the resistance for capped pipe ends.DNVOSF201 Dynamic Risers. p pr.Guidance . the initial ovality is given by: f0 = D max − D min D (5. The ovalisation due to external pressure or moment in the asinstalled position shall not be included. Ovalisation caused during the construction and installation phase is to be included in the ovality.of . is given by: 402 If the pipe design is sufficient to meet the above propagation criterion. The nominal thickness is given by: p p ( t) = 2 t ⋅ f y ⋅ α fab D (5. u 4 1 . .e. and t nom = (t1 + t corr ) /(1 − %t fab ) when the negative fabrication thickness tolerance is given as a percentile of the nominal thickness.20) 303 The initial departure from circularity of pipe and pipe ends.Guidance . If buckle arrestors are in pipe sections subjected to fatigue. the bursting limit state designs explicitly against the extreme pressure loading condition over the lifetime in compliance with standard ULS design checks. The allowable utilisation is however in compliance with recent industry practice for wellknown riser types.end . % t fab. 302 The resistance for external pressure (hoop buckling). means to prevent or arrest them should be considered in the design.17) p pr t = 35 ⋅ f y ⋅ α fab ⋅ 2 D 25 (5. Maximum ovality from fabrication is given in section G 200. p c(t). If conditions are such that propagating buckles are possible.B 200 203 The burst criterion is valid if the mill pressure test requirement in DNVOSF101 has been met.9.0 if no buckle propagation (once initiated) is allowed.. The resistance against buckling propagation. For installation with waterfilled pipe.23) Where p min is a minimum internal pressure.5%). If the buckle is allowed to travel a short distance (where the neighboring pipe section acts as buckle arrestors) γ c may be reduced to 0. Guidance note: pmin is the local minimum internal pressure taken as the most unfavourable internal pressure plus static head of the internal fluid.note  (p e − p min ) ≤ p pr γ c γ m γ SC (5. If not.end . any fatigue degradation should be evaluated due to stress concentration factors. i. see 3. The minimum required wall thickness for a straight pipe without allowances and tolerances is given by: D 400 Propagating Buckling 401 To ensure that a possible local buckle remains local and does not lead to successive hoop buckling (collapse) of neighbouring pipe sections a propagating buckling (collapse) check is required: t1 = D f min f y . Once initiated. hoop buckling may still be initiated at a lower pressure by accidental means.21) 304 The initial ovality shall not be taken less than 0. The consequences of such a failure should be evaluated. such a collapse may form a propagating buckle that will travel along the pipe until the external pressure drops below the propagation pressure or until a change in property arrests the buckle.of . Guidance note: For a pipe designed to meet the hoop buckling (external collapse) criteria outlined above.22) D 300 System Hoop Buckling (Collapse) 301 Pipe members subjected to external overpressure shall be designed to satisfy the following condition: where γc = 1. is given by: (p e − p min ) ≤ pc ( t1 ) γ SC ⋅ γ m (5. (p c ( t ) − p el ( t )) ⋅ (p 2 ( t ) − p 2 (t ) )= p c (t ) ⋅ p el (t ) ⋅ p p ( t ) ⋅ f 0 ⋅ D c p t (5. Examples of such means would be impact or excessive bending due to tensioner failure. the system hoop buckling (collapse) criterion is also met. pmin equals pe .
Guidance note: The failure modes controlled by this semiempirical limit state is yielding and combined local buckling and hoop buckling due to combined bending.end .25) Tk is the plastic axial force resistance given by: 603 Pipe members subjected to bending moment. . gross plastic deformation and wrinkling due to combined loading.29) 502 Pipe members subjected to bending moment.75 M k is the (plastic) bending moment resistance given by: M k = f y ⋅ α c ⋅ (D − t 2 )2 ⋅ t 2 (5. ≤1 2 where: Md Ted p ld pe = Design bending moment. it may not be necessary to design the riser for propagating buckling. Guidance note: The failure modes controlled by this limit state comprise yielding.79 High 0. It is equivalent to the plastic limit bending moment capacity (including the effect of strain hardening and wall thinning) for (T ed/T k) <<1. 703 A displacementcontrolled condition is one in which the structural response is primarily governed by imposed geometric displacements.note  Te + T k 2 pld − p e + p b (t 2 ) ≤ η2 2 (5.note System effects should be considered for installation methods involving many pipe sections being exposed to a similar loading condition.05 multiplied with γSC γm apply.0. or Displacement Controlled conditions (DC secondary) or combined load types. It reduces to the traditional wall thickness Von Mises criterion. the basic usage factor shown in Table 58 apply: Table 58 Low 0.15). 604 Pipe members subjected to bending moment. (5. . see 3. tension and external overpressure. i.end . (5. see B 100 = Local internal design pressure. Instead.B 200 = Local external pressure 602 For the WSD format the design load effects equals the corresponding characteristic load effect. effective tension and net external overpressure shall be designed to satisfy the following equation: D 700 Displacement Controlled Conditions 701 Loads and load effects may be classified as follows: (5. see e. for pressure and effective tension load effects only.e.of .Guidance .g.Guidance . If detailed information is not available a condition factor γC=1. the load effect factors and resistance factors equals unity: γF =γE =γA =γSC =γ m =1.18).Guidance . October 2010 Section 5 Page 33 Connectors and riser joints may be considered equivalent to buckle arrestors. effective tension and net external overpressure shall be designed to satisfy the following equation: M M k Te + T k 2 p −p min + e p c (t 2 ) 2 ≤ η4 2 (5.83 Usage factor η for combined loading Normal 0.24) Working Stress Design (WSD) format may be used for the combined loading check for pipes with D/t ratio less than 30.e. see B 100 = Design effective tension. effective tension and net internal overpressure shall be designed to satisfy the following equation: 2 M {γ SC ⋅ γ m } d ⋅ 1− p ld − p e + Ted p (t ) T Mk b 2 k 2 pld − pe + pb (t 2 ) D 600 Alternative WSD Format 601 As a more easytouse alternative the following (5. (5. DET NORSKE VERITAS . The present WSD is based on explicit limit states for combined loading and provides results on the conservative side compared to the corresponding LRFD limit states.DNVOSF201 Dynamic Risers. 702 A loadcontrolled condition is one in which the structural response is primarily governed by the imposed loads. i. {γSC ⋅ γm}2  Md  + Ted + {γSC ⋅ γm}2 pe − pmin ≤1 Mk Tk pc (t 2 ) 2 2 2 Where the hoop buckling capacity p c(t) is given by Eq.26) effective tension and net internal overpressure shall be designed to satisfy the following equation: p − pe M ⋅ 1 − ld p (t ) Mk b 2 2 Tk = f y ⋅ α c ⋅ π ⋅ (D − t 2 ) ⋅ t 2 p b (t2 ) is the burst resistance given by Eq. end . API RP 2RD.of .of . The design criterion may be viewed as a (plastic) Von Mises criterion in terms of cross sectional forces and plastic cross sectional resistance.27) Load Controlled conditions (LC or primary).note – D 500 Combined Loading Criteria 501 Pipe members subjected to bending moment.28) where all parameters are defined in D 500.
g. calculations (see E 300).Guidance .of . the fatigue damage must hence be calculated at a number of regularly spaced points to identify the most critical location. accumulated plastic deformation must be considered. the total (primary and secondary) load effect must be checked against the strain limits and acceptance criteria for displacement controlled conditions in DNVOSF101. determination of stress concentration factor (SCF) not included in the SN curve. These sites normally include welds and details that causes stress concentrations.note  107 The stress to be considered for fatigue damage accumulation in a riser is the cyclic (i.30) 109 This combined stress varies around the circumference of the riser pipe. selection of appropriate SN curve . methods based on fatigue crack propagation. October 2010 Page 34 Section 5 704 Displacement controlled conditions should be subdivided into: conditions with static (functional and pressure) loads. governed by the geometric stiffness due to the effective tension). σ for pipes is normally a linear combination of the axial and bending stress given by: σ= Te 32⋅ M ⋅ (D − t 3 ) + π ⋅ (D − t 3 ) ⋅ t3 π ⋅ D4− ( D− 3)4 2·t ( ) (5. 108 The governing cyclic nominal stress component. 706 In dynamic DC loading conditions (lowcycle) fatigue often becomes the limiting condition for extreme loading conditions. 707 If the bending moment can be assumed secondary a condition factor γc=0.e.e. which have magnitude and corresponding number of cycles large enough to cause fatigue damage effects.. Hence. Fatigue Limit State E 100 General 101 The riser system shall have adequate safety against fatigue within the service life of the system. . the methods based on SN curves are used during design for fatigue life assessment.0 (5. 705 In static DC loading conditions the following fundamental design principles apply: the primary load effect (i. 708 Displacement controlled conditions must be documented.. a direct fatigue testing of the actual components shall be performed with due regard of the chemical composition of the internal and external environment. 202 The fatigue criterion. see e. 103 All critical sites for anticipated crack initiation for each unique component along the riser shall be evaluated. which shall be satisfied. running and hangoff shall be considered. conditions with dynamic (environmental) loads taken into account. installation.e. A more rational and fundamental design principle is to require that inelastic displacements caused by cyclic loads is not allowed.85 may be multiplied on the bending moment in D 500 and D 600. Reference is made to section 4 and Appendix B for more details with respect to fatigue design and analysis. the following shall be considered: assessment of shortterm distribution of nominal stress range . 105 Normally. E.end . towing. Guidance note: Examples where bending stress may be considered secondary: − a riser bent into conformity with a continuous curved structure such as a reel. Fatigue crack propagation calculations may be used to estimate fatigue crack growth life and to establish NDT inspection criteria to be applied during both fabrication and inservice. Pipe sections and components subjected to inelastic deformations shall be designed with due consideration of accumulated plastic deformation (ratcheting) such as incremental hoop buckling (accumulated ovality) and plastic (low cycle and ultra low cycle) fatigue. LC part of the load effect ) shall fulfil the load controlled criteria in this standard ignoring the secondary load effects (i. shall be where D fat = Accumulated fatigue damage (Palmgren DET NORSKE VERITAS . − in areas where the geometric equilibrium shape of the riser is not influenced by the bending stiffness (i. incorporate thickness correction factor. 104 The fatigue assessment methods may be categorised into: methods based on SN curves (see E 200). 106 If representative fatigue resistance data are not available.e. The latter must be documented by analysis with and without bending stiffness for both static and dynamic loading conditions. For cases where the waves are incident from several different directions.DNVOSF201 Dynamic Risers. DC part of load effect). timedependent) principal stress.31) 102 All cyclic loading imposed during the entire service life. E 200 Fatigue assessment using SN curves 201 When using the calculation methods based on SN curves. DNVRPC203 determination of accumulated fatigue damage D fat over all short term conditions.. may be written: Dfat ⋅ DFF ≤ 1. Temporary phases like transportation. the total strain must be confined to the elastic region.
Crack propagation calculations typically contain the following main steps: determination of longterm distribution of nominal stress range . Crack initiation time is normally neglected for welds . see Table 59. Time to first inservice inspection may be based on crack growth versus time results with the criteria given in Table 59 in combination with fabrication/installation records. 402 Necessary data shall be logged during the life cycle for documenting and analysis of fatigue status for temporary risers. For single sided girth welds. estimation of the initial crack size and geometry and/or any possible time to crack initiation. see Table 59 Design fatigue factors DFF Safety class Normal 6. is the expected value of defects left after fabrication and NDT.end .g. access for inspection during fabrication. The inservice inspection plans after first inspection shall be based on the inspection results obtained and the plans updated accordingly.g. lack of penetration defects is hard to detect by NDT. a i /2ci for surface inspection is considered acceptable means of inspection. This log shall be reviewed regularly to assess the need for fatigue crack inspections. Machining off the root pass is considered to significantly improve the fatigue quality. pressure. Note that these codes defines the initial crack size to be based on the 90 % probability of inspection level for the applied NDT method and not the mean level as applied in this standard. The surface crack depth to total defect length (ai/2ci) should be assumed low (less than 1:5) if no other documentation is available. E 400 Inservice Fatigue Inspections 401 The SN curve approach may be used for screening purposes to identify the most likely regions where fatigue cracks may appear during service. etc. field data (water depth. Crack depths in the range of 1 to 2 mm may be hard to find. floater data including top tension and the length of time and seastate for each mode of operation. E 300 Fatigue assessment by crack propagation calculations 301 A damage tolerant design approach applies. This implies that the riser components shall be designed and inspected so that the maximum expected initial defect size would not grow to a critical size during service life or time to first inspection. Guidance note: For surface cracks starting from the transitions between weld/base material. selection of the appropriate crack growth law with appropriate crack growth parameters. 302 The fatigue crack growth life shall be designed and inspected to satisfy the following condition: N tot N cg ⋅ DFF ≤ 1 . Some codes have reduced life requirements for fatigue crack growth vs. DFF = 403 Inplace NDT or removal of the riser for dry 303 The assumed initial defect size. integration of the fatigue crack propagation relation with respect to the longterm stress range distribution to determine the fatigue crack growth life.DNVOSF201 Dynamic Risers. .note  203 The design SN curve shall be based on the meanminustwostandard deviations curves for the relevant experimental data. manufacture method. Using a reliable welding procedure is important for such cases. determination of cyclic stress in the prospective crack growth plane.Guidance . 304 The maximum acceptable initial crack size may be used to evaluate detection limits of NDT methods for the actual component. riser configuration. The log shall typically include running sequence of joints. especially for the root pass. For nonwelded components the mean stress shall be determined.32) where: Ntot Ncg = = total number of applied stress cycles during service or to inservice inspection Number of stress cycles necessary to increase the defect from the initial to the critical defect size Design fatigue factor. surface finish. a crack depth of 0.1 mm (e. the thickness and geometry of the structure. SN. determination of final or critical crack size (through the thickness. October 2010 Section 5 Page 35 Miner rule) DFF = Design fatigue factor. defects and 2ai /2ci for embedded defects. e. For defects found. due to undercuts and microcracks at bottom of undercuts) may be assumed if other documentation about crack depth is not available. see DNVRPC203.of . full or partial penetration weld and the number of passes used to complete the weld. fatigue crack calculations to establish residual life shall be based on the sizing accuracy of the applied method and the expected value shall be used for fatigue assessment.0 High 10. The DET NORSKE VERITAS . Best estimate initial crack size (mean value) shall be applied.0 expected initial defect size (mean value) shall be established based on an evaluation of the detection capability of the inspection method. Crack growth parameters (characteristic resistance) shall be determined as mean plus 2 standard deviations.0 (5. a factor of 5 for fatigue crack growth versus 10 for SN. see Appendix B. welding method.0 Table 59 Low 3.). unstable fracture/gross plastic deformation) . density. Light grinding of hot spot areas should be considered to remove undercuts and increase reliability of the inspection.
DNVOSF201 Dynamic Risers. Accidental environmental events should be assessed assuming 1) a return period value with reasonable likelihood of not being exceeded during the design life (e. Account shall be taken of other loads that might reasonably be present at time of the accidental event. Accidental loads typically results from unplanned occurrences. accidental loads shall be determined with due account of the factors of influence. F 200 Categories of accidental loads 201 Accidental loads may be categorised into (not limited to): fires and explosions impact/collisions. Loads occurring at the time of an accidental event do not normally need to be assumed concurrent with an extreme environmental load condition. such as : F 300 Characteristic accidental load effects 301 Accidental loads and load effects are determined by the frequency of occurrence and their magnitude. safety systems and control procedures.g. tension system or draw works motion compensator loss of buoyancy. e. However. postaccidental resistance against environmental loads (if the resistance is reduced by structural damage caused by the accidental loads). equipment. incorrect operation or technical failure.Guidance .g. air cans for spar units loss of mooring line. such as : dragging anchor failure of support system. the following design checks apply: resistance against direct accidental load. earthquake) with recurrence interval from several hundred to a few thousand year. . tendon or guidewire dynamic positioning (DP) failure (driveoff or drift off) exceedence of incidental internal overpressure: loss of pressure safety system failure of well tubing or packers. DET NORSKE VERITAS . etc.note  102 Relevant failure criteria and accidental loads in terms of frequency of occurrence and magnitude shall be determined based on risk analyses and relevant accumulated experiences. Accidental loads shall be understood as loads to which the riser may be subjected in case of abnormal conditions.g. the damaged structure resulting from an accidental load event shall be able to resist relevant pressure and functional loads in an extreme environmental load condition. Characteristic accidental load effects and load combinations for different operating modes are given in Table 510. Further. 200 years) and 2) a rare intense event (e. ultimate resistance and consequence assessment due to exceedence of a SLS introduced to define operational limitations. heave compensating system malfunction (loss or stuck). October 2010 Page 36 Section 5 F. (Typically discrete events with an annual frequency of occurrence less than 102). pressure surge well kill – bullheading environmental events earthquake tsunamis iceberg Guidance note: Environmental load conditions with a 10 000 year return period as a normal “tail” behaviour in the long term probability distribution function is implicit in the ULS design criteria and need not be considered as an accidental (or abnormal) load condition for risers.of . the arrangement of the installation. operational procedures.g. Accidental Limit State F 100 Functional requirements 101 The Accidental Limit State (ALS) is a limit state due to accidental loads or events.end . Such factors may be personnel qualifications. such as: infrequent riser interference (see H 100) impact from dropped objects and anchors impact from floater/floating objects hook/snag loads. Normally. e.
9 γc = 0. i. the simplified design check proposes a total safety factor in the range 1.0 γc = 0. a high characteristic value for the resistance of the link should be used.T (5.2. 404 A simplified design check with respect to accidental load may be performed as shown in Table 511 below multiplied on appropriate load effect factors selected according to Table 52 and resistance factors according to Table 53 and Table 54. shall be recognised.8 may be disregarded 402 Design with respect to accidental load must ensure that the overall failure probability complies with the target values in Table 25. Not Operating Operating Expected value associated with the Aloads. Sound engineering judgement and pragmatic evaluations are hence required. 1 Limit State Category Floads Eloads Aloads Value dependent on measures taken and magnitude and probability of occurrence.9 Accidental loads or events γc = 0. Expected specified or expected extreme value associated with the Aloads. Example of the latter is tensioner failure where the tensioner shall provide sufficient integrity to survive certain environmental scenarios without further progressive collapse. times the structural failure probability conditioned on this event.note – ∑ Pf Di ⋅ PDi ≤ P f . and identify the need for deliberately introducing weak links in the system. PDi . does not occur in case of accidents (in particular floater driveoff or driftoff events or failure of draw works heave compensation system).DNVOSF201 Dynamic Risers. This probability can be expressed as the sum of the probability of occurrence of the i’th damaging event. For combined loading.e. Expected value Characteristic design pressure or incidental as suitable. This range is consistent with standard industry practice interpreted as corresponding to safety class Normal for accidental loads with a probability of occurrence equal to 104.T is the target failure probability according to Table 25.0 γc = 1. controlled riser failure above the subsea valve. or indirectly. end . DET NORSKE VERITAS . Eloads may relate to a season rather than a year. 403 The inherent uncertainty of the frequency and magnitude of the accidental loads.0 γc = 1. Table 511 Prob. The adequacy of simplified design check must be assessed based on the summation above in order to verify that the overall failure probability complies with the target values in Table 25.0 for accidental event with a probability of occurrence equal to 104 and survival of the riser is merely related to a conservative definition of characteristic resistance.33) where P f.11. When maximum load is calculated in a potentially weak link. F 400 Design against accidental loads 401 The design against accidental loads may be done by direct calculation of the effects imposed by the loads on the structure. The number of discretisation levels must be large enough to ensure that the resulting probability is evaluated with sufficient accuracy. October 2010 Section 5 Page 37 Table 510 Characteristic accidental load effects and combinations for different operational modes Mode of Operation Not Operating ALS Intact Structure Load effect category Ploads Expected value Characteristic design pressure or incidental as suitable. PfDi. In this standard accidental loads and events are introduced in a more general context with a link between probability of occurrence and actual failure consequence. The requirement is accordingly expressed as: Guidance note: Standard industry practice assumes safety factors equal to 1. Operating Temporary. of occurrence >102 102 103 103 104 104 105 105 106 <106 Simplified Design Check for Accidental loads Safety Class Safety Class Safety Class Low Normal High Accidental loads may be regarded similar to environmental loads and may be evaluated similar to ULS design check To be evaluated on a case by case basis γc = 1. ALS Damaged Structure Combined load effect defined with an annual probability of exceedence = 101 Not applicable NOTE 1) Eloads may be determined on weather forecast if time to repair is short and protective measures can be taken.Guidance .of . as well as the approximate nature of the methods for determination of accidental load effects. The ALS analysis may provide extreme loads for the design of wellhead and rig equipment. If the repair period is confirmed to a season. the probability of exceedance may be relaxed i.e. Such weak links may be required to ensure that unacceptable escalation.. by design of the structure as tolerable to accidents.
G 200 Ovalisation limit due to bending 201 Risers shall not be subjected to excessive ovalisation and this shall be documented. 104 Serviceability limit states for the global riser behaviour are associated with limitations with regard to deflections. G 300 Riser stroke 301 For a top tensioned riser. artificial supports and support settlements. set down/pull down effects and floater draught.e. In order to prevent premature local buckling. . effects from subsidence and settlements shall be evaluated. the Owner will specify requirements however.03 Do G 400 Examples 401 Examples of SLS for drilling and workover riser with subsea BOP are outlined in the following Table 512.0 %: D − D min f 0 = max ≤ 0 . 302 Riser systems shall be designed to have sufficient stroke such that damages to riser.note  freespan shoulders. Some examples are given in the subsequent sections. Guidance note: FMEA. a tensioner pulls upward on the top part of the riser in order to limit bending and maintain constant tension. Such events will typically be controlled by maintenance/inspection routines and by implementation of early warning or failsafe type systems in the design.and downstroke calculations must include effects from environmental response. tide. 202 The requirement may be relaxed if: a corresponding reduction in moment resistance has been included. HAZOP and design review meetings are useful systematic procedures that can lead to identification of SLS and for reviewing the consequences of setting operating limitations and of exceeding those limitations. such as pigging and tool access requirements.of . 403 Examples for export and import riser serviceability limits should be set for riser installation and pigging. 405 Other serviceability limits may be determined to limit the degradation of riser coatings and attachments or for allowances due to wear and erosion. i. Additionally. The tensioner must continue to pull as the riser and the floater move vertically relative to each other. Such point loads may arise at DET NORSKE VERITAS . Serviceability Limit State G 100 General 101 Serviceability limit states are most often associated with determination of acceptable limitations to normal operation.Guidance . the flattening due to bending together with the outofroundness tolerance from fabrication of the pipe shall be limited to 3. In many cases. 404 Examples for a production riser with a surface tree the riser is part of the well control system and may not be disconnected and hungoff. 204 Special consideration shall be made of ovalisation after loading causing plastic strains. Some SLS examples are given in Table 515. Some examples of how this may influence SLS are summarised in Table 513. In addition. 103 Exceeding a SLS shall not lead to failure and an ALS shall be defined in association with exceedance of SLS. 303 The up. the frequency and consequences of events after exceeding an SLS shall be evaluated. unreeling of pipes and riser interference/impact. For permanent risers. geometrical restrictions are met. length of slick joint. etc.34) 402 Examples when drilling with a surface BOP (e. tension changes and length variations needs to be taken into account. 305 The most unfavourable fluid density shall be considered. pressure (end cap effects). TLP. swell. The static stroke is due to current loading and set down effect due to floater mean offset.DNVOSF201 Dynamic Risers. components and equipment are avoided. e. dynamic stroke. 203 Ovalisation shall be checked for point loads at any point along the riser system. reeling. SPAR) the riser is part of the well control system and may not be disconnected and hungoff. (5.g. the designer must also carry out evaluations with respect to riser serviceability and identify relevant SLS criteria for the riser system.g. storm surge. The travel of the tensioner is called its 'stroke'.end . see Table 514. clearance between surface equipment and drill floor. draw works. The wave loading introduces relative motions between the floater and the riser. 102 It is important that all operating limitations and/or design assumptions are clearly highlighted and implemented in the operating procedures. Riser stroke influences the design requirements for tensioner. and additional cyclic stresses caused by the ovalisation have been considered. October 2010 Page 38 Section 5 G. 304 Environmental response includes static and dynamic stroke. makeup (riser production tolerances). displacements and rotation of the global riser or ovalisation of the riser pipe. tension. temperature. The floater mean offset includes effects from static wind and mean wave drift.
DNVOSF201 Dynamic Risers, October 2010 Section 5 Page 39 Table 512 Examples of SLS for drilling and workover with subsea BOP
Function Drilling with fluid returns Guide tools or assemblies into the well Overpull SLS criteria Limit fatigue on drill string and wear on wellhead/riser Excessive angle may result in getting stuck or not being able to land the string properly Avoid overloading the wellhead, BOP and connectors Approaching the resistance of the wellhead/BOP and connectors Approaching the resistance of the tensioner system Avoid damage Comment Usually monitor flexjoint angle and follow weather forecast and adjust mooring to minimise joint angle Due to tight tolerances
Flexjoint or ball joint Operating limits for specific operations
Disconnect and Hang off
Riser stroke Umbilical, choke, kill and other attachments
Hang off BOP and well control
Overpull may be used to check that a connector is made up properly or in an attempt to release a stuck string For a normal hangoff scenario sufficient time shall be allowed for pulling the downhole string Weather is resulting in excessive platform motion and offset. Risk and consequences of damage may govern criteria for interference
Table 513
Examples of SLS for drilling and workover with surface BOP
Function Drilling with fluid returns Guide tools or assemblies into the well Overpull SLS criteria Limit fatigue on drill string and wear on wellhead/riser Excessive angle may result in getting stuck or not being able to land the string properly Avoid overloading the wellhead, BOP and connectors A weather limitation would be set to avoid riser interference Comment Usually monitor flexjoint angle or stressjoint curvature. It is not normally feasible to adjust moorings Due to tight tolerances
Flexjoint or Stressjoint Operating limits for specific operations
Riser installation
Running and retrieving the riser
Overpull may be used to check that a connector is made up properly or in an attempt to release a stuck string Usually run on guidewires in close proximity to other risers
Table 514
Examples of SLS for export and import risers
Function Running and retrieving the riser Inspection or cleaning SLS criteria A weather limitation would be set to avoid riser interference Pig launching and associated temporary loading Comment Usually run on guidewires in close proximity to other risers
Riser installation Pigging
Table 515
Examples of SLS for production risers with surface tree
Function Running and retrieving the riser Reason for SLS A weather limitation would be set to avoid riser interference The tensioner may be designed for bottomout The tensioner may be designed for bottomout Comment Usually run on guidewires in close proximity to other risers Energy absorption criteria shall be specified Energy absorption criteria shall be specified
Component Riser installation
Riser stroke
Limit the frequency of bottomout Limit the design requirements for the jumper from the surface tree to the topside piping
H. Special Considerations
H 100 Interference 101 The riser system design shall include evaluation or
analysis of potential interference with other risers, mooring lines, tendons, hull, the seabed, and with any other
obstruction. Interference shall be considered during all phases of the riser design life.
102 A feasible design approach may be categorised into:
No Collisions allowed Collisions allowed
DET NORSKE VERITAS
DNVOSF201 Dynamic Risers, October 2010 Page 40 Section 5
103 A first step is hence to determine whether collisions
are likely to occur or not. If collision occur, it must be documented that the structural integrity is not endangered, i.e. the pipe capacity is sufficient for both SLS and ULS (incl. ALS & FLS) conditions. This requires an assessment of collision frequency, location, force impulse or relative riser velocity prior to the impact. Separate local calculations/analyses will in general be required for assessment of pipe stresses during impact.
Guidance note: Owing to the complexity of interference analyses, due balance between simplified and advanced analyses is recommended to obtain efficient analyses: − screening analyses using a simplified approach to identify critical conditions or configurations; − detailed analyses of identified critical conditions or components using stateoftheart interference analyses. Screening analyses may imply use of − simplified environmental loads, e.g. current only, simple profile without directionality; − simplified Wake Induced Oscillation (WIO) and Vortex Induced Vibration (VIV) models for current only or undisturbed flow models for waves; − simplified onset of collision criteria. Detailed analyses for criticality assessment of collisions may include: − − − − − hydrodynamic interaction models; global collision models; dedicated CFD calculations; explicit collision load effect models; explicit limitstate design criteria..  end  of  Guidance  note 
Guidance note: Fatigue failure in the SN curve approach, see E 200, is normally based on through wall cracks. Where through wall cracks are applied as failure criteria, it should be ensured that through wall cracks should not cause unstable fracture. Normally, brittle fracture in riser systems is avoided by selection of material with sufficient ductility and Charpy V notch impact energy and by performing NDT during fabrication to ensure that only acceptable defects are present in the riser system after fabrication. Unstable fracture may occur under unfavourable combinations of geometry, fracture toughness, crack like welding defects and stress levels. The risk of unstable fracture increases in general when the state of “plain strain” is approached at the crack tip. This occurs in general with large material thickness, low temperature, high loading rates, high strength material and deep cracks subjected to bending. Fracture toughness data as KIC, JIC or CTODC values are necessary to perform defect assessment. The failure assessment diagram is a twocriteria failure model that considers unstable fracture, gross plastic deformations (plastic limit load), and the interaction between these mechanisms.  end  of  Guidance  note 
204 For ALS evaluations, normally no partial load effect
factors are required for load effects, flaw size and toughness, i.e. all partial safety factor shall be taken as unity. Normally, the riser pipe is designed based on the principle that plastic hinges may develop without giving rise to unstable fracture. In such case, the nominal stress for unstable fracture shall not be less than the design (yield) stress of the member.
H 300 Global Buckling 301 Global buckling (Euler buckling) implies buckling of
the pipe as a beamcolumn in compression. The procedure is as for "ordinary" compression members in air using the concept of effective tension.
104 Model testing for verification of structural capacities,
hydrodynamic interaction models and global analysis methodology is recommended.
H 200 Unstable Fracture and Gross Plastic Deformation 201 Pipe members, including components and girth welds
shall have adequate safety due to unstable fracture for a representative part or throughwall crack during the service life of the riser.
302 A negative effective tension may cause a riser to
buckle as a beamcolumn in compression. Distinction shall be made between loadcontrolled and displacementcontrolled buckling. Excessive loadcontrolled buckling involves total failure and is not accepted while a displacement controlled buckling may be acceptable if the postbuckling condition is acceptable.
202 Defect assessment of crack like defects should
normally be performed in accordance with BS 7910 Level 2A failure assessment diagram Partial safety factors for flaw size, fracture toughness and yield strength should be as given in BS 7910, Appendix K, Table K2 while load effect factors shall be in accordance with B 200.
Guidance note: The partial factors in Table K2 in BS 7910 annual target probabilities of 103, 7*105, and 105 correspond to those for safety class Low, Normal and High given in this standard.  end  of  Guidance  note 
303 The global buckling resistance for loadcontrolled
condition may be calculated according to recognised stability criteria in structural design codes, e.g. ISO 138192.
304 Displacementcontrolled buckling may be acceptable,
provided it does not result in other failure modes. This implies that global buckling may be acceptable provided that: local buckling criteria are fulfilled in the global post buckling configuration; displacement/curvatures/angles of the riser are acceptable and cyclic effects are acceptable.
203 Defects assessment at fatigue sensitive locations shall
be additional to fatigue crack evaluations, see E.
DET NORSKE VERITAS
DNVOSF201 Dynamic Risers, October 2010 Section 5 Page 41
305 Special care shall be given when a small decrease in
top tension of a toptensioned metallic riser could cause excessive bending moment. In that case, the designer shall establish a minimum effective tension that gives a margin above the tension that is predicted to cause excessive bending moments.
Guidance note: It is essential that an appropriate tensionedbeam model is used for the analysis of global buckling. The consequence of a toosmall positive effective tension is excessive curvature and
bending moment near the location of minimum effective tension. Note that members above the tension joint for top tensioned risers may be subjected to compressive forces for some riser types  end  of  Guidance  note 
DET NORSKE VERITAS
October 2010 Page 42 Section 6 SECTION 6 CONNECTORS AND RISER COMPONENTS Contents A. Relevant sections include 5. include (not limited to): local buckling. .8. breakout loads. thermal load effects (trapped fluid/water. deflections and finish damage. leak tightness and fatigue resistance for all relevant load cases. Connector Designs B 100 Functional Requirements 101 Riser connectors shall allow for multiple makeup and breakout in a reliable manner. 106 For further functional requirements see ISO/CD A. 205 Deformations.8.8 B 200 Design and Qualification Considerations 201 Connectors shall be designed to sustain the design loads and deformations arising from makeup/breakout. provisions shall be made on the riser joint/connector to allow for attachment of an anode (bracelet). 136287 Section 5. pigging and maintenance operation when applicable. C 100 C 200 General Objective Connector Designs Functional Requirements Design and Qualification Considerations Seals Local Analysis Documentation Documentation Operating and maintenance manuals 103 For permanent risers. dog types. and bolted flanges designed for facetoface contact . leakage and fire resistance. cyclic loading. All relevant limit states must be considered. and shall be made to a low stress part of the connector. dissimilar metals) and thermal transients. Electrical connection to the riser shall be made by welding or other qualified method. thermal gradients and internal and external pressure loads without exceeding the connector design resistance. Reference is made to ISO/CD 136287 for further details on design. Guidance note: Riser connectors basically provide a means of connecting and disconnecting riser joints or equipment. 104 The riser pipe and components shall provide an internally flush bore to ease running components into a well. General A 100 Objective 101 This section gives requirements in relation to design. 105 The design shall ensure that any trapped water/fluid does not interfere with the installation or operation of the connector. B 100 B 200 B 300 B 400 C.of . inspection and maintenance. 6. 206 The FLS capacity shall be verified to ensure that the connector will not fail due to cyclic loading. hub type.end .5 and 6. 102 The aim of the design is to ensure that the connectors and riser components have adequate structural resistance. galling tendency between sliding elements B. thread disengagement. bending moments and effective tensions.3. if applicable. 204 Issues which may require considerations in ULS and ALS. The requirements apply also to other riser components and at transitions to the pipe wall thickness. external loads applied to the pipe body. DET NORSKE VERITAS . A 100 B. analysis and qualification of metal connectors and components used in riser design. 202 The connectors should be designed to be at least as strong as the pipe or weld with respect to strength. may require consideration in SLS. see ISO/CD 136287 section 6. unstable fracture and excessive yielding.note  203 As a minimum. the following loading parameters/ conditions shall be considered and documented by the manufacturer when designing connectors and components: makeup loads. The most commonly used types of riser connector design comprises: threaded types.Guidance . The connector may permit for interchangeability between connector halves to allow riser joints to be run in any sequence.DNVOSF201 Dynamic Risers. analysis and requirements. internal and external pressure including test pressure. Other methods may be used. Resistance against accidental loads such as fire and impact shall also be considered when applicable. which adversely affects the use. 102 The external profile of all riser components shall not restrict the passage of equipment like guideframes and specialised tooling required for riser installation/retrieval. leak tightness.5. fatigue.
end . the two seals should be of a different design without common failure modes. a double seal may be provided.Guidance .15. driveoff.note  Seals for riser connectors should be static. driftoff or tensioner system failure. Documentation C 100 Documentation 101 The documentation for the connector shall as a minimum comply with the requirements of ISO/CD 136287 section 6. Nonintegral seals use separate seal elements that can be removed and replaced.of . 308 Seals of high reliability should be used to confine flammable fluids. 208 All riser connectors shall be qualified for the application based on finite element analysis in combination with performance qualification testing.e. To achieve redundancy. which combined with external pressure results in frequent pressure reversals on sealing mechanism. or the connector shall be constructed of. leakage and fatigue (SCF's). . C 200 Operating and maintenance manuals 201 The documentation for the connector shall as a minimum comply with the requirements of ISO/CD 136287 section 6. including landing blocks. In cases where "weak links" are introduced to protect components against accidental loads.of . or coated with. Seals must be selected with consideration to the required service life. blowdown. Integral seals are built into the connector and are nonreplaceable. etc. A single seal therefore may have modest reliability. 302 Seal design for connectors and riser components shall include consideration of external pressure. plastic hinge may preferably develop in the pipe before failure of the connector occurs in order to increase the ductility in the riser system. commissioning. operation. slick joints.. should demonstrate fit for purpose of their function. installation and reassembly. B 300 Seals 301 Connectors shall provide a seal between the mating segments that is compatible with any fluids that will pass through the riser. complex riser joint cross sections (multiple pipes). For permanent risers where metaltometal seals are not utilised.16. Guidance note: All seals are sensitive to damage during handling. All operating conditions (i. 307 Connectors exposed to cyclic loading shall utilise nonloadcarrying seals in order to maintain high reliable against leakage with time. the service exposure in terms of chemical aggressiveness and temperature as well as pressure and relative displacements that need to be accommodated. either the connector including components shall be designed in such a way that an acceptable corrosion control can be implemented at the joint. To enhance the reliability. C. representative connectors of the same type may be designed by analytical methods (design equations) in combination with finite element analysis whenever necessary. taper joints. For static strength.e. Seal designs are either integral or nonintegral. Guidance on FE analysis of connectors and riser components is given in ISO/CD 136287 section 6. flex/ball joints. tension joints. 304 The seal and the connector including any bolts and preload shall be considered together as a system to determine the sealing performance. Alternatively. The effect of sealing performance by the connector includes effects such as torque of pin/box connectors and bolt resistance and preload. a connector with known breaking resistance may be applied. temperature. seals and sealing surface shall be corrosion resistant in the actual environment. sealing should take place between surfaces which have little or no movement relative to each another. However. in order to reduce the uncertainty in the preload of the connector and ensure that the preload is within the design limits. i. the minimum requirements are given above. 402 The most unfavourable combination of specified tolerances shall be used in connection with FE analysis for strength.) shall be considered. 303 Seal rings wetted with internal fluid shall include the same internal corrosion allowance as the connecting pipe and be of compatible material.8. 305 Metaltometal seals are preferred as the primary seals on riser connectors. such as friction. which should be carried out on connectors to be used on risers. a corrosionresistant material. Using analytical or numerical calibration of the qualified connector. Loads and boundary conditions for use in local analysis shall be obtained from the global analysis procedure. startup.Guidance . lubrication. fluids under high pressure and corrosive fluids. The seal must maintain its integrity under all external and internal loading conditions. i. redundant seals (primary plus backup) should be provided. 306 209 Connector makeup shall be performed according to a qualified procedure considering factors. etc. DET NORSKE VERITAS .note  B 400 Local Analysis 401 Local FE analysis should be performed for connectors and structural components. October 2010 Section 6 Page 43 207 For connectors intended to be used in corrosive environment.e. Guidance note: It is considered reasonable that the analysis or tests. . Seal design shall also consider operating conditions what may result in frequent changes in the external loads and internal pressures.8.end . This does not necessarily mean they have to be as strong and reliable as the connecting pipe or weld.8. testing.DNVOSF201 Dynamic Risers.
Submarine Pipeline Systems. inspection/ replacement possibilities and possible failure modes during the intended use. Guidance note: The external corrosion allowance in the splash zone for CMn steel is usually taken as 68 mm. welds. General A 100 Objective 101 This section specifies the requirements for materials. 305 All sliding surfaces shall be designed with sufficient additional thickness against wear and tear. Special attention should be given to the following where applicable: clamped supports. and corrosion resistant alloys (CRA) including ferritic austenitic (duplex) steel. fabrication and NDT methods and procedures in general with the exceptions given in Table 71 and in part B of this section. manufacture and fabrication. The materials for use in the riser system shall have the dimensions and 102 The riser components shall be forged/extruded rather than cast whenever a favourable grain flow pattern. fabrication and documentation of riser pipe. nondestructive testing. other stainless steels and nickel based alloys. shall be determined during design. necessary to comply with the assumptions made in the design. corrosion and wear resistance. ductility. 303 All materials liable to corrode shall be protected against corrosion. . dimensional and weight verification. If a higher utilisation is used. components. sliding supports. which can be exposed to fluids with H2 S during the lifetime of operation of the riser. A 100 A 200 A 300 B. Special attention should be given to local complex geometry. B 100 B 200 General Objective Application Material Selection Additional Requirements General Long term properties mechanical properties.e. 304 Requirements for corrosion allowance shall comply with DNVOSF101. shall be applied for all metallic materials included in this section. 203 The additional considerations in Table 71 could be met by additional evaluations and/or specifications to the applied codes or by applying the material. manufacture. loads. coatings. such as strength. based on the consequence with respect to failure and experience. with regard to the characteristic properties of materials which shall be obtained after heat treatment. 204 The design utilisation in this standard depends on the material quality and level of control. slick joints. martensitic stainless steels (“13% Cr”). final shaping and assembly. DET NORSKE VERITAS . toughness. B. areas that are difficult to inspect/repair. 302 Materials for riser systems shall be selected with due consideration of the internal fluid. consequences of corrosion damage. welding and NDT requirement in DNVOSF101 and in part B of this section. Additional Requirements B 100 General 101 Risers shall be made in seamless or longitudinally welded pipes.DNVOSF201 Dynamic Risers. temporary/permanent operations. austenitic stainless steels. expansion. 306 The possibility for “sour” service conditions shall be evaluated for all riser components. temperatures (maximum and minimum). October 2010 Page 44 Section 7 SECTION 7 MATERIALS Contents A. see Section 5. service life. mechanical and corrosion testing. manufacturing. and possibilities for electrolytic corrosion.Guidance . equipment and structural items in the riser system. 307 The quality of the materials used shall be tested/documented. ball joints and telescopic joints. Requirements to testing and control. Special consideration shall be given to the splash zone.note  A 200 Application 201 The requirements in this section are applicable for metallic risers of the following materials: carbon Manganese steel. supplementary requirement U.of . the principles and requirements in DNVOSF101. A 300 Material Selection 301 The materials selected shall be suitable for the intended use during the entire service life.end . dynamic seals. The selection of materials shall ensure compatibility of all components in the riser system. All elastomers and other nonmetallic materials shall have documented compatibility with all fluids to which they could be exposed including pressure and temperature cycles. The requirements are relevant both for pressure containing and for load carrying parts. A. clad/lined steel. external environment. 202 This standard applies to risers fabricated from linepipe material meeting internationally recognised codes for materials. i.
D1100 and sec 12. 209 Wear resistance shall be considered. Effects of corrosion shall be accounted for with a minimum of 1mm allowance unless it can be documented that a corrosion allowance can be eliminated. sec 5 D1000. internal environment as well as the maintenance and inspection procedures that shall be applied.note  203 It is strongly recommended to specify tight dimensional requirements at pipe ends for SCR’s in order to reduce the stress concentration factors associated with the girth welds. − Supplementary requirement P in DNVOSF101 shall apply for riser pipes with accumulated strain εp ≥ 2%.end . particularly for drilling risers or other wear exposed components. reduction in wall thickness due to internal corrosion shall be evaluated. 205 Weldments and other components with high fatigue loads shall be identified. DET NORSKE VERITAS . sec 5. . 206 For temporary risers manufactured from CMn steel.Guidance . Requirement for ductility in the through thickness direction shall be considered. εp . Fatigue properties 208 Special considerations shall be given to riser pipes to be used for fluids containing hydrogen sulphide and defined as “sour service” according to NACE Standard MR0175. machining and fabrication shall also be considered.DNVOSF201 Dynamic Risers. Wear 202 Adequate fatigue life of base metal and weldments shall be verified by fatigue analyses that are based on fatigue testing (SN fatigue or fatigue crack growth testing) or existing fatigue data. further testing shall be conducted. Extended NDT can take place in the form of spot checks performed by other qualified operator. Adequate wear resistance shall be verified by analyses and / or testing. Equivalent criteria have to be developed for other materials based on the fracture properties. For heavy wall components with SMYS above 420 MPa a higher ductility level may be required. resulting from installation and operation shall be treated in accordance with the principles of DNVOSF101. This can be obtained by the use of supplementary requirement D in with DNVOSF101. and be in accordance with NDT Level I in DNVOSF101 or similar.note  103 Accumulated plastic strain. 204 NDT of longitudinal welds shall include 100% control for transverse imperfections. − where sufficient and relevant test data are not available. Corrosion 106 Limitations on SMYS on parts exposed to cathodic protection shall be according with DNVRPB401. Guidance note: Treatment of accumulated strains in accordance with DNVOSF101: − Requirements and guidelines to performance of ECA at accumulated strains εp ≥ 0. the accumulated plastic strain limits of 0. − selection of SN curves shall match the weld detail and quality. and extended NDT of these shall be considered. production. storage and age control requirements shall be defined for nonmetallic pressure containing parts. .3% and 2.end . completion / workover) or if inspection only is possible by means of remote control equipment (as for permanent risers used for export. import. 104 However.3% are given in DNVOSF101. B 200 Long term properties General 207 The external surface for temporary risers shall be protected by a suitable coating system in addition to routine coating repair and preservation of damaged coating.of . 107 Generic base polymer(s) ASTM D1418. Manufacturing process.of . 105 Reduction of area Z of cast and forged CMn fine grain and low alloy steel shall be ≥ 35%. physical property requirements.0% applies only to the DNVOSF101 linepipe specification. welding and NDT applied. An evaluation shall take into consideration the material properties. injection). October 2010 Section 7 Page 45 a maximum degree of homogeneity and the absence of internal flaws are of importance. Guidance note: − when test results in terms of existing fatigue data are used as basis for fatigue analyses. the tests shall have been conducted on materials with expected fatigue properties equal to the chosen material and in a representative internal/external environment (including corrosion protection if relevant). This can be obtained by the use of supplementary requirement S in DNVOSF101. 201 The long term material properties with regard to fatigue and corrosion shall be documented. Special consideration shall be given as to whether regular inspection intervals or replacements can be applied (as for temporary risers used for drilling.Guidance .
DNVOSF101 Sec. October 2010 Page 46 Section 7 “Recognised” linepipe code Ovality 4 Mill test 5 Fracture properties. See B 202 and 203 of this section. strain hardening DET NORSKE VERITAS Fracture arrest Local buckling Strain limits 2 (0. Welding and NDT 6 Suppl. The effect of “other” stressstrain curves for high strength steel shall be evaluated if relevant Testing. U Statistics 7 Suppl. See B 103 of this section. S Sour service Suppl F Fracture arrest NDT level I 8 NDT High strength steel (yield stress > 555)9 Included 1 2 3 4 5 6 7 8 9 10 X X X X X X (X) X X X X X X X (X) (X) X The “additional considerations” shall constitute input to an evaluation regarding the highlighted topic. D Load effect calculations Collapse Resistance Content Sour service Increased utilisation Fatigue 3 . D Dimensional requirements 3 Suppl.DNVOSF201 Dynamic Risers.6 E1100 (hoop stress to be at least 96% of SMYS) Criteria to be based on a fracture mechanic assessment To document that SMYS is at least 2 standard deviations below the mean yield stress and that SMTS is at least 3 standard deviations below the mean ultimate strength “ See B 204 and 205 of this section DNVOSF101 is limited to yield stress less than 555. The moment capacity formulation is valid for (D max – Dmin)/D less than 3%.5 D800 Mill test requirement in accordance with DNVOSF101 Sec. ref. P Ductility10 Suppl. Such an evaluation shall end up with resulting specifications or guidance as required.3% & 2%) Pressure Containment Table 71 Additional Considerations DNVRecognised codes OSF101 Additional considerations1 Sec 6.
but not limiting to the following items: a summary including design check key results and illustrations in figures. including but not limited to: floater layout drawings with risers. procedure/methods. Quality Plans. Documents that are considered proprietary and confidential shall be available for review. riser fabrication drawings. The design basis document normally include information supplied by the owner. components. all applicable load cases. corrosion. tolerances. calculation input data. fabrication. introduction including the objective of the document and a brief description of the riser system. 302 Drawings shall be provided for the fabrication and construction of the riser system. procedures for riser system and component analysis including analysis models and applied computer programmes. as far as practicable. A.DNVOSF201 Dynamic Risers. relevant component and interface design loads. and should include all relevant information for all relevant phases of the lifetime of the riser system. 102 Documentation shall be available to the purchaser or the purchaser's agents. graphs for the geometric model. General A 100 Objective 101 This section gives requirements for documentation and verification of riser systems during design. nonvoluminous. explanation of notations and abbreviations. NDT procedures. Submittals and/or approval procedures shall be agreed between the purchaser and the supplier. structural and other fabricated items: material and manufacturing specifications. and manufacturer's/fabricator's quality system manual. including material details. October 2010 Section 8 Page 47 SECTION 8 DOCUMENTATION AND VERIFICATION Contents A. assumptions for calculations and details of the computer programs reference number of the standard/guideline/textbook including the reference number for the formulae. wastage and other allowances where applicable. see B 200. wall thickness selection including minimum thickness. full traceability of the calculations performed. B 100 B 200 B 300 B 400 B 500 B 600 B 700 C. design basis if not included in a separate document. Documentation B 100 Design 101 Design documentation shall. C 100 C 200 C 300 General Objective Documentation Design Design basis Design analysis Manufacture and fabrication Installation and Operation DFI Résumé Filing of documentation Verification General requirements Verification during the design phase Verification during the fabrication phase 202 A summary of those items normally to be included in the design basis document is included in Appendix F. including. including sources and assumptions. installation and operational phases. Manufacturing Procedure Specification (MPS). limit states and safety classes for all relevant temporary and operating design conditions. A 100 B. B. The design documentation shall be presented in such a form that it is readily applicable for design review and third party verification. utilisation ratios along the riser) and evaluation of the results in the light of the limit states and assumptions made in the analysis wrt. key results presented in a clear and concise manner (i. inspection and maintenance of the riser system in service. B 300 Design analysis 301 The design analysis documentation shall be selfcontained and selfexplanatory setting forth in full detail. be concise. manufacturing/fabrication procedures. including boundary conditions.e. assumptions made with respect to treatment. welding procedure specifications/qualification records if relevant. B 200 Design basis 201 A design basis document shall be established in the initial stages of the design process. and drawings of the corrosion protection system. B 400 Manufacture and fabrication 401 The following information shall be prepared prior to start of or during manufacture of pipes. DET NORSKE VERITAS . equipment.
and all nonconformances identified during manufacture and fabrication. pressure testing. fabrication and installation phase operation of the riser system and maintenance phases of the riser system shall. complete statistics of chemical composition. running/retrieving. operate and maintain the riser and its component systems. 703 The engineering documentation shall be filed by the Owner or by the engineering contractor for a minimum of 10 years. integrity evaluation. test records (visual. mechanical properties and dimensions for the quantity delivered. test requirements and acceptance criteria. installation and commissioning. detail design. etc. 503 The Riser Installation and Operational Manual should contain as a minimum the following information: stepbystep procedure for handling. fabrication and installation including. NDT. and contractor Quality System manual. running. October 2010 Page 48 Section 8 402 All relevant documentation shall be submitted to owner. responsibility. 702 The engineering and asbuilt files shall. periodical and special inspections. provide recommendations. weights and part numbers of various components. 603 The main objectives of the DFI résumé are to ensure that only necessary information is kept available. comprise the documentation from design. include final inservice inspection reports from startup.g. DET NORSKE VERITAS . installation and testing specifications and drawings. including but not limited to: fabrication procedures. Any changes to the riser system after startup will be a part of operation history and shall be reflected in a condition résumé. necessary asbuilt drawings. pipes. effective and rational operation. defines how to safely install.g. seal rings. installation Manuals. deviations. personnel qualification records. Design basis and key data for the riser system shall by filed for the lifetime of the system. fabrication. The DFI résumé is therefore not supposed to be updated based on events/changes made in the operation phase. requirements and sufficient information for the operation. coating and corrosion protection data sheets. design basis data. which should be prepared jointly by the designer and the owner. contingency procedures. preservation and storage of the riser system. riser clamps. to facilitate the safe. as a minimum. dimensional. operational procedures for e. The manual(s). operation. maintenance activities and modification or requalification throughout the entire lifetime of the installation. followon engineering. 502 The following information shall be prepared prior to start of installation: Failure Mode Effect Analysis (FMECA) and HAZOP studies. anodes. It shall contain all relevant data and documentation used for: the design. material certificates for e. inspection and maintenance procedures for each component. and repairs performed preparation of plans for periodic inspections of the riser system 602 Documentation referred to in the DFI Résumé shall be kept for the lifetime of the riser system and shall be easily retrievable at any time. B 700 Filing of documentation 701 Maintenance of complete files of all relevant documentation during the life of the riser system is the responsibility of the owner.). requirements. 605 The DFI résumé is a historical document. as a minimum. 704 Files to be kept from the operational and B 600 DFI Résumé 601 A DFI Résumé shall be prepared for riser systems including equipment and components. verification activities. emergency disconnect. transportation. provide a system description for the riser system . heat treatment. B 500 Installation and Operation 501 Installation and Operational requirements shall be documented in a Riser Installation and Operation Manual(s). and maintenance and modifications of the riser system and input to the preparation of plans for periodic inspection 604 The purpose of the DFI résumé is to: provide a reference key to the detail technical documentation. and final reports of maintenance and repair. provide a summary of all design. condition monitoring records.DNVOSF201 Dynamic Risers.. piping components. operating limits for each mode of operation. This includes documentation from design to startup and also documentation from possible major repair or reconstruction of the riser system. bolts. incl. tests on samples. inservice inspection. manufacturers drawings of the riser system components outlining critical dimensions. and critical design areas with references to underlying detailed documentation. fabrication procedure qualification reports including welding procedure qualification records. hangoff. recommended spare parts list. handling. FAT.etc. operating.
appropriate personnel qualifications and organisation of the design.DNVOSF201 Dynamic Risers. that corrosion. C 200 Verification during the design phase 201 Verification of design should include checking of the following items: that specifications are in compliance with the applicable rules and regulations etc. components and materials. The consequences of any failure or defects that may occur during construction of the riser system and its anticipated use shall receive particular attention in this assessment. C 300 Verification during the fabrication phase 301 Verification during fabrication should include the following items to check that: the specifications are in accordance with public regulations/provisions and safety requirements . that accidental loads are in compliance with the results from the risk analyses . 105 Independent analyses shall to the extent practicable possible be performed using different software as applied in design. and that the programmes are adequately tested and documented. constructions or in case of new/modified software. materials. 104 There shall be organisational independence between those who carry out the design work. taking into consideration the design. that deviations during fabrication and installation are assessed and if necessary corrected. 102 The extent of the verification and the verification method in the various phases shall be assessed. dimensions including assembly tolerances. NDT detection limits. that the design of important structural details are adequate. shall be verified. 106 Verification work and findings shall be documented. October 2010 Section 8 Page 49 C. deviation procedures are adequate during fabrication the transportation and storage of materials and fabricated assemblies are adequate. used in dealing with new problems. personnel qualifications are in accordance with the requirements.g. including material selection and corrosion protection. for environmental data. This is of particular importance when programmes are DET NORSKE VERITAS . that drawings are in accordance with calculations and specifications . The calculations should be sufficiently accurate and extensive to demonstrate clearly that the dimensions are adequate. satisfactory work instructions and procedures are prepared. e. calculations of loads and load effects. the methods and equipment of suppliers and at the fabrication site are satisfactory with regard to control of dimensions and quality of riser pipe. that measuring requirements are complied with. the usefulness of computer software. 103 The verification shall confirm whether the riser system satisfies the requirements for the specific location and method of installation and operation. Verification C 100 General requirements 101 Compliance with provisions contained in relevant national and international regulations or decisions made pursuant to such regulations. independent calculations should be performed of the riser system including riser components of significance to the overall safety. surface protection and work performance are in accordance with the basic assumptions made during design. and those who verify it. wear.and erosion protection measures are adequate. and the analysis methods used.
B 100 B 200 B 300 B 400 B 500 C. seal ring damage). Defects should be documented with respect to type. 203 Parts that are damaged repaired or particularly exposed and where failure will incur serious consequences shall be subject to particular attention in the planning of inservice inspection and maintenance. .Guidance . Criticality of components and ease of inspection should be considered early to ensure that provisions are made for adequate inspection.g. girth welds. ageing and corrosion. The safety factor should account for uncertainties in timetofailure predictions. leaks (loosening of mechanical connectors.DNVOSF201 Dynamic Risers. marine growth. Moreover. 205 The maximum interval between inspections should be based on the component's predicted time to failure divided by a safety factor. C 100 C 200 C 300 C 400 C 500 C 600 General Objective Inservice Inspection. see DNVOSF101. internal and external corrosion. B 200 Riser Inspection 201 The inspection philosophy should be an integral part of the design. lubrication. General A 100 Objective 101 The objective of this section is to provide requirements for operation and inservice inspections. A 100 B. e. internal and external wear. Replacement and Monitoring General Riser Inspection Riser monitoring Guidelines for inspection intervals Condition Summary Reassessment General Ultimate Strength Extended Service life Material Properties Dimensions and Corrosion Allowance Cracked Pipes and Components cathodic protection. The designer should also consider the time required for repairs or replacement when determining maximum inspection intervals.E 300. damage. loosened or heavily distorted coating. anode attachment welds). 206 If the maximum inspection interval is longer than the intended service life. then the component should be inspected and refurbished if necessary or replaced.g. However. risks of failure and ease of inspection.g. e.note  B. size and location. excessive marine growth. maintained and inspected to maintain an acceptable safety level throughout the service life of the riser. fatigue cracking (e.end . wear. MAINTENANCE AND REASSESSMENT Contents A. section 10. scratches. inspection is not expected to be necessary and need not be included in the operation and maintenance documents. The influence of defects on structural or pressure integrity should be assessed. periodically disconnected components and paint should generally be inspected or replaced on a scheduled basis. touch down point. dents. the equipment should be designed to facilitate these maintenance operations. pipe distortion. October 2010 Page 50 Section 9 SECTION 9 OPERATION. Inspection intervals should be developed for each mode of failure such as fatigue. A. anticorrosion/abrasion coatings. Replacement and Monitoring B 100 General 101 Risers shall be operated. DET NORSKE VERITAS . Manufacturer supplied data should include recommended maintenance operations and intervals. They should also be inspected after potentially damaging incidents and to confirm that any repairs have been properly performed. This section also provides general guidance on structural integrity assessment of risers to demonstrate fitness for purpose in case deviations from design appear during operation. 204 Risers to be inspected for fatigue cracks should be inspected in accordance with the principles given in section 5. 102 Risers should be visually examined for factors such as external damage. connectors. if during operation the intended service life is extended beyond the original maximum inspection interval of a component. general pipe configuration and sliding of buoyancy modules and/or ballast. external corrosion. soil conditions at seabed. Inservice Inspection. Inspections relating to areas such as the following may be necessary for risers and riser components: overloaded/permanently deformed riser string components. 202 The designer should ensure that necessary inspection methods or replacement procedures are available and are scheduled and described in adequate detail as part of the operating and maintenance documentation for the facility. abrasion. Guidance note: Equipment consumables such as seals.of .
Such an evaluation should be based on: reliability of inspection method(s) used. The system can also be used to record and estimate riser fatigue damage. departure form the original basis of design. This document should be updated annually. change of use that violates the original design or previous integrity assessment basis.. damage or detrimentation of the riser system shall be included in an updated condition summary. Guidance note: A riser monitoring system is not mandatory. e. modifications. Wall thickness measurements by internal means.end . justification for changing guideline inspection intervals. Reassessment C 100 General 101 An existing riser shall undergo an integrity assessment to demonstrate fitness for purpose if one or more of the following conditions exists: extension of service life beyond the originally calculated design life.g. top/bottom flexjoint etc of e. elapsed time from last inspection performed and/or inspection/repair history. This monitoring should include the recording of riser response and tension (if relevant) as well as the composition.g. even where cracks are not found it should be considered to perform a light grinding at the hot spot areas B 500 Condition Summary 501 The DFI Résumé is a historic document. change in top tension for TTR’s. present condition and service history. damage or deterioration to a riser component. age. based on the factors listed in this section. 202 The riser pipe or riser component must have sufficient ductility to develop the failure mechanism in question and large inelastic displacements or fractures due to repeated yielding must not occur. Verification activities should be considered also for the operational phase.DNVOSF201 Dynamic Risers. e.g. should be documented and retained by the owner. C 200 Ultimate Strength 201 The ultimate strength of damaged members should be evaluated by using a rational.g.g. October 2010 Section 9 Page 51 B 300 Riser monitoring 301 The riser's internal and external operating condition should be monitored to reveal whether design conditions have been exceeded. riser type and location. The riser monitoring system can also be applied in connection with active floater positioning for reduction of stresses.g.of . change in internal fluid. drilling risers. 402 The intervals given in the Table 91 should not be exceeded unless experience or engineering analysis justifies longer intervals. redundancy. DET NORSKE VERITAS . change of use. pigs and by external means at selected reference points should be considered. which is completed at the start of the operational phase. Local buckling or other nonlinear instabilities must be considered in the calculation. e.note  502 Modifications. for monitoring riser dynamics and for design verification. Component Above water components Below water components All components Cathodic protection Areas of known or suspected damage Components retrieved to surface C 300 Extended Service life 301 Extended service life may be based on results from performed inspections throughout the prior service life. DNV RPF101 may be applied. deep water or new design with few long term operating examples. change in floater. 103 In case of change of use. The condition summary shall provide user groups with an overall picture of the actual condition and functioning of the riser system during the operational phase. justifiable engineering approach. Guidance note: In some situations. pressure and temperature of the riser contents. but it is useful for setting and maintaining precise tension. by change in environmental data or relocation. specific intervals based on criteria discussed elsewhere in this section. B 400 Guidelines for inspection intervals 401 The following factors should be taken into account when determining inspection intervals: safety class. 102 Assessment of existing risers should be based on the most recent information of the riser..Guidance . e. C. In such cases. Load data should be revised according to latest metocean data and the current layout of the riser. repair. . Table 91 Guideline for inspection intervals Inspection type Visual Visual NDT Visual or Potential Survey As appropriate As recommended by manufacturer Interval 1 year 35 years As needed 35 years After exposure to design event After disconnect damage or detrimentation of the riser system. e. results of previous inspections. changes in design operating or loading conditions or prior damage and repairs. measures shall be implemented to maintain an acceptable safety level.
Guidance note: DNV RPF101 Corroded Pipelines gives guidance on assessment of pipes corrosion/erosion defects including local wall thinning due to fabrication tolerances and grind repairs. to document that they are removed. . The determination of characteristic values shall be in accordance with the evaluation procedure given in ENV 1993 11.DNVOSF201 Dynamic Risers. The yield and tensile strength may be taken as the minimum guaranteed yield and tensile strength given in material certificates. Section thickness for use in the strength assessment may be calculated from the measured section thickness combined with the expected corrosion in the remaining lifetime.end .end . C 600 Cracked Pipes and Components 601 Pipes and components containing cracks should be repaired/replace as soon as possible. reduced for corrosion allowance.of . The remaining life of such a repair should be assessed in each case. Detected cracks may be ground and inspected again.of . based on the observed rate of corrosion. Material properties may be revised from design values to ‘as built’ values based on material certificates. material tests may be used to establish the characteristic ‘as built’ yield strength. October 2010 Page 52 Section 9 of the riser systems to remove undercuts and increase the reliability of the inspection. DET NORSKE VERITAS . Annex Y.Guidance . 402 Alternatively.Guidance note  C 400 Material Properties 401 . 502 For unprotected or cathodically protected steel. the section thickness and the expected corrosion may be updated based on the measured values.note  C 500 Dimensions and Corrosion Allowance 501 Strength assessment shall be based on 'as built' dimensions. Due consideration must be given to the inherent variability in the data. .
A. A 100 Objective General B. Structural damping L. K. general overview of analysis techniques with emphasis on treatment of nonlinearities.6. The focus will be on the following essential issues: general overview of global system behaviour and important nonlinearities. F.6. General A 100 B 100 B 200 B 300 B 400 C 100 C 200 C 300 C 400 C 500 D 100 D 200 D 300 D 400 D 500 E 100 E 200 E 300 E 400 Objective General Top tensioned risers Compliant riser configurations Nonlinearities Purpose of global analysis General modelling/analysis considerations Static finite element analysis Finite element eigenvalue analysis Dynamic finite element analysis General Coupled system analysis Efficient analysis strategies considering coupling effects Coupled floater motion analysis Decoupled floater motion analysis General Morison equation for circular crosssections Morison equation for double symmetric crosssections Principles for selection of hydrodynamic coefficients GLOBAL ANALYSIS A. coupled floater/slender structure analysis).5 and A. API RP 2RD should be consulted for a more detailed technical description as well as modelling guidance of special components such as tensioner. balljoint etc (mainly sections A. Comments related to recommended procedures for specific riser systems are addressed whenever appropriate. October 2010 Appendix A Page 53 APPENDIX A Contents A. Hydrostatic pressure loading H. and stateoftheart review of recent developments regarding analysis techniques of particular interest for deep water riser systems (e. guidance to global analysis to provide consistency with the guideline requirements.1 and A.5. Marine growth G. Reference is also made to API RP 2RD. structural damping etc).6.6. internal fluid flow.3. A general generic presentation is used as the basic methods of analysis can be applied to a wide range of riser systems.2.6. Forced Floater Motions J. Combined floater/slender structure analysis E. flexjoint.5). A.1 for modelling of multiple tubular crosssections in global analyses. hydrodynamic loading. overview of important load models (e.2. C.DNVOSF201 Dynamic Risers.6. effective tension.g. general techniques and accepted modelling practice (guidelines.4. handbooks and technical papers) I. Functional description and extended use of references is applied to describe well established procedures. Guidelines and Handbooks Technical references 104 In particular.4.4. Global riser system analysis D. References DET NORSKE VERITAS . sections A. A. Internal fluid flow H H H H 100 200 300 400 General Steady flow Accelerated uniform flow Slug flow 103 The document should not be regarded as a selfcontained document on analysis but rather as an introduction to basic principles and more detailed guidance on selected important topics. Hydrodynamic loading in moonpool J 100 J 200 J 300 K 100 K 200 L 100 L 200 General Moonpol kinematics Hydrodynamic coefficients Global Rayleigh damping model Local Rayleigh damping models Standards. stressjoint. Physical Properties of riser systems 101 The objective of this Appendix is to give guidance on global riser system analysis referred to present technical level of tailormade FE computer codes for static and dynamic analysis of slender structures.g.2. Hydrodynamic loading on slender structures 102 The overall intention with the present document is to support practical implementation of the LRFD and WSD design formats and provide background information for selection of adequate method of analysis.
deep draft floaters (DDF) and semisubmersibles). 203 An alternative solution is used for Spar platforms where the top tension is obtained from buoyancy modules attached along the upper part of the riser inside the moonpool. TLP. Bending moments in risers operated from a shell Spar are mainly due to the resulting horizontal hull motion as well as hydrodynamic loading from the entrapped water in the moonpool. A taper joint may also be included at the keel of Spar and other deep draught floaters. It is therefore convenient to distinguish between top tensioned . Titanium and composite pipes are suggested for deep water applications in order to keep the top tension requirement at an acceptable level. waves and floater motions.and compliant risers to reflect the principle applied for absorption of floater motions. export/injection risers. Spar platforms. A significant higher top tension must however be applied to account for imperfect tensioner arrangements and allow for redundancy in case of partial loss of top tension.DNVOSF201 Dynamic Risers. 206 The crosssectional composition depends on the functional applications. import and low pressure drilling risers are normally single tubular risers. functional requirements and design load conditions. 204 The static and dynamic behaviour of top tensioned risers is largely governed by the applied top tension. 104 There is also a significant potential for hybrid riser configurations. workover/completion risers. Compliant risers are mainly applied as production. types of operation.g. Increased top tension can also be applied to reduce the probability of collision in riser arrays and limit the mean angles in bottom of the risers. The required overpull is system dependent with a typical range of 30%60%. A pronounced peak in the bending moment distribution is normally seen close to the wave zone. B 200 Top tensioned risers 201 Vertical risers supported by a top tension in combination with boundary conditions that allows for relative riser/floater motions in vertical direction are denoted top tensioned risers. combining the properties of tensioned and compliant risers in an efficient way. export and injection risers. Applied top tension and stroke capacity are the essential design parameters governing the mechanical behaviour as well as the application range. Bending moments are mainly induced by horizontal floater motions and transverse loading due to current and wave action. Espinasse et al (1989) 205 Steel pipes have traditionally been applied for conventional water depths. Risers are therefore commonly grouped into the following categories.g. Several supports are introduced along the riser system to constrain the riser motion in transverse hull direction. reflecting the area of application: drilling risers. The effective weight of the riser system defines the lower limitation for the applied top tension to avoid compressive effective tension in the riser at static position. These categories differ with respect to typical dimensions. The relative riser/floater motion in vertical direction is commonly denoted stroke. the effective tension distribution along the riser is mainly governed by functional loading due to the applied top tension and the effective weight of the riser. A main concern in selection of the global riser configuration is how floater motions should be absorbed by the riser system. B 300 Compliant riser configurations 301 Compliant riser configurations are designed to absorb floater motions by change of geometry. B 100 Physical Properties of riser systems General 202 101 The purpose of this section is to give a brief overview of characteristic physical properties and governing nonlinearities of riser systems. 207 A taper joint. crosssectional composition. Top tensioned risers are applicable for all functional purposes as mentioned above and will hence represent an attractive alternative for floaters with rather small heave motion (e. A flexjoint or balljoint may be applied to reduce bending stresses at termination to the floater. 102 The main functional requirements to marine risers is to provide for transfer of fluids and gas between seafloor and a floater. Export. Steel risers with buoyancy modules attached can alternatively be applied for deep water. October 2010 Page 54 Appendix A B. Multitube crosssections may typically be found in high pressure drilling. as well as allow for transportation of various well operation tools. Hence. Pronounced peaks in the bending moment distribution are normally found at the support locations. Furthermore. without use of heave compensation systems. The applied top tension is commonly specified in terms of increase relative to the effective weight of the riser system. examples are given by e. flexjoint or balljoint is applied to reduce bending stresses at termination to the seafloor. Top tensioned risers operated from TLP’s and semisubmersibles are equipped with a separate (hydraulic) heave compensation system to account for the floater motions and at the same time maintain a constant target value for the applied top tension. production risers. The required DET NORSKE VERITAS .and workover risers as well as production risers. 103 Risers will typically be operated from a floater. There are no constraints (except from friction forces) in longitudinal direction allowing the hull to move relative to the riser system. Such information is crucial when selecting analysis strategy to describe the static and dynamic behaviour when the system is exposed to environmental loading due to current. Characteristic properties of these main riser categories are discussed separately below. the riser is constrained to follow the horizontal floater motion at one or several locations. The intended (idealised) behaviour is that the applied top tension should maintain a constant target value regardless of the floater motion. also denoted as overpull.
always be present. Engseth et. Nonlinearities will also be decisive for the statistical response characteristics for systems exposed to irregular loading. Nonlinearities are introduced by the quadratic drag term in the Morison equation expressed by the relative structurefluid velocity and by integration of hydrodynamic loading to actual surface elevation. Lazy Wave. angular orientation). at least to some extent.g. The floater motion characteristics will in many situations be decisive for the dynamic tension and moment variation along the riser (e. 1988). Nonlinearities due to item 1) and 2) will. It should be noted that the frequency content in all response quantities can be WF or combined WF and LF depending on the analysis strategy applied in the global response analysis. 1) Geometric stiffness (i. the yield stress is typically higher than for steel and the specific weight is much lower (about half the steel weight). This is due to a low modulus of elasticity (half that of steel) implying a higher degree of flexibility. it is however also possible to arrange metallic pipes in compliant riser configurations. Critical locations on compliant risers are typically the wave zone. O’Brien et. Free Hanging Risers in steel have been installed in the Gulf of Mexico (Phifer et al 1994). global riser position (coordinates. Single pipe crosssections are typically applied for compliant riser configurations 402 The relative importance of these nonlinearities is strongly system and excitation dependent. October 2010 Appendix A Page 55 system flexibility is for conventional water depths normally obtained by arranging nonbonded flexible pipes in one of the ‘classical’ compliant riser configurations: Steep S. Tension variation is hence a nonlinear effect for risers. touch down area at seafloor and at the terminations to rigid structures. elongation. hog and sag bends.e. 303 Compliant riser systems will in general experience significantly larger static and dynamic excursions when compared to top tensioned risers. hog and sag bends). It should be noted that external hydrostatic pressure is not considered to be a nonlinear effect as hydrostatic pressures normally will be handled by the effective tension / effective weight concept (Sparks 1984) in computer programs tailor made for slender structure analysis (e. 2) Hydrodynamic loading. contribution from effective tension to transverse stiffness).g. A Lazy Wave configuration with increased horizontal extension termed Long Wave is proposed for application of metallic risers for a deep water Floating Production Ship and Offloading facility (FPSO) for North sea conditions (Karunakaran et al 1996). Steep Wave.e. 4) Material nonlinearities. al 1988. Important nonlinearities that always should be carefully considered can be summarised as: 102 Subsequent detailed cross sectional analysis to determine local stresses and strains can be performed using resulting cross sectional forces from the global analysis as boundary conditions and considering possible DET NORSKE VERITAS . Relevant global response quantities can be grouped into the following main categories: resulting crosssectional forces (effective tension. 305 Termination to rigid structures is an essential design issue for compliant riser configurations. global riser deflections (curvature. C 100 Global riser system analysis Purpose of global analysis 304 Titanium may offer several benefits relative to steel for some of these configurations. ball joint or flex joint. waves and 1st order floater motion) into nonGaussian system responses. and support forces at termination to rigid structures ( resulting force and moments). Item 3) is relevant for twoaxial bending due to inplane as well as out of plane excitation. 101 The purpose of global riser system analyses is to describe the overall static and dynamic structural behaviour by exposing the system to a stationary environmental loading condition. distance to other structures. These response quantities are given directly as output from global riser analyses.DNVOSF201 Dynamic Risers. torsional moment). bending moments. etc). 3) Large rotations in 3D space. B 400 Nonlinearities 401 A basic understanding of important nonlinearities of riser systems is of vital importance for system modelling as well as for selection of adequate global analysis approach. In such applications it may also be desirable to apply prebend pipe sections to reduce the dynamic curvature at critical locations along the riser (i. Lazy S. The primary design requirement is to limit bending curvature and pipe stresses. Semisubmersibles. TLPs. al. Ships). translations. Pliant Wave or Free Hanging (catenary). while 4) and 5) are more system specific nonlinear effects. Furthermore. and 5) Contact problems in terms of seafloor contact (varying location of touch down point and friction forces) and hull/slender structure contact.e. Possible solutions are carefully designed bend stiffener. C. 302 In deepwater. and Lazy Wave configurations in steel and titanium have been proposed as deepwater riser systems for TLP and Semisubmersibles. position of touch down point on seafloor. An essential issue is how nonlinear properties of the riser system and hydrodynamic loading mechanisms transform the wave frequency Gaussian excitation (i. bending moment versus curvature and torsion moment versus twist angle) is applied in such analyses. Environmental load effects will consequently also be of greater concern for compliant configurations. the secondary design requirement is to minimise forces on the supporting structure. A global cross sectional description in terms of resulting force/displacement relations (axial force versus axial elongation.
and frequency domain dynamic analyses. buoyancy modules.g. volume forces (weight and buoyancy): specified forces (e. This is in particular the case for irregular analyses where rather long simulations in general are required to estimate extreme responses with sufficient statistical confidence. the required accuracy of the analysis. With limited computer resources available this should always be kept in mind when deciding the analysis strategy.) The standard approach often applied in practical analyses is based on engineering judgement and experience possibly supported by some simple parametric studies. tension control etc. and nonlinear time domain dynamic analysis.g. 301 The purpose of the static analysis to establish the static equilibrium configuration due to static loading (weight. The most important features required for adequate modelling and analysis of deep water systems in general can be summarised as: 3D formulation allowing for unlimited translations and rotations. adequate structural damping formulation. top tension. nonlinear static analysis. regular. current modelling. it is convenient to distinguish between the following basic loading components: 1. Following standard FE terminology. The benefit from long simplified simulations versus reduced simulation length using a more advanced tool must be considered carefully when available computer resources are limited. numerous simplified analyses will normally produce more information regarding overall static. October 2010 Page 56 Appendix A external/internal pressure loading. Deviations from the unknown ‘true’ response will depend on method of DET NORSKE VERITAS . cross sectional properties.g.and dynamic analyses. the acceptable loss in accuracy by introduction of a simplified approach must be seen in relation to other uncertainties involved. A judgement regarding tradeoff between computational efficiency and model uncertainty will therefore always be involved when strategies for cost effective analysis are decided. (e. C 300 Static finite element analysis 203 The basic strategy to obtain efficient analyses is to use a simplified response model and/or use of simplified timesaving FE analysis methodologies such as 2D formulations and linearized time. It can be shown that the statistical uncertainty roughly will be reduced proportional to the square root of the simulation length. prescribed displacements (displacement of terminal points from stressfree. 103 Procedures for evaluation of LRFD capacity checks for combined loading are addressed separately in Appendix C C 200 General modelling/analysis considerations 201 A Finite Element (FE) approach is normally considered for global riser system analysis. 2. Sødahl and Larsen 1989) and simulation of critical events identified by a simplified approach (e.and irregular loading due to waves and floater motions. Issues that often must be considered in the decision process are briefly described in the following: the acceptable accuracy is dependent on the purpose of the analysis. 3. uncertainties related to modelling of environmental loading. analysis as well as the response model.e.g. hinges.DNVOSF201 Dynamic Risers. Simple methods allowing for a broad analysis scope is attractive in the early analysis stage while more specialised advanced analyses of identified critical conditions are of more interest in final stages.to specified positions).and geometric stiffness and allowing for nonlinear material properties. taper joint and bend stiffener analysis).and dynamic system behaviour when compared to a reduced number of advanced analyses. detailed design. Detailed component analysis can also be performed by application of resulting forces and deformations obtained from global analyses as boundary conditions in local quasistatic analyses (e. 204 Model uncertainties will always be present in numerical simulations of marine structures. 202 The computational efforts of nonlinear time domain dynamic analysis considering a detailed global riser response model can be substantial. terminations to seafloor and floater).g. and displacement dependant forces (current loading). special features allowing for efficient modelling of components such as swivels. floater motions. applied top tension). 4. Passano 1995) can also be applied to gain computational efficiency. the simulation length used in stochastic analyses is crucial to obtain sufficient confidence of extreme response estimates. This ‘target accuracy’ is dependent on the purpose for which the results will be used. see C 500 for a description. flexjoint. flexjoints etc. Different conclusions may be drawn depending on scope of work (feasibility. buoyancy.and bar elements including material. preliminary design. Other strategies such as use of special designed quasistatic analysis for bend stiffeners (e. It is therefore desirable to apply simplified analysis strategies as a supplement to the general advanced approach in order to achieve more efficient computer analyses. The latter approach can be highly beneficial when the wanted results are relatively few extreme responses of a complex system exposed to irregular excitation. current) for given locations of riser terminations to rigid structures (e. final verification). Static analysis is always the first step in global riser analysis and defines the starting point for subsequent eigenvalue. clump weights. Simplifications introduced to achieve analyses that are more efficient will in most cases lead to an increased model uncertainty. Static riser analyses are normally performed using a nonlinear FE approach. seafloor/riser and hull/riser contact formulations. conventional small strain slender beam. hydrodynamic loading according to the Morison equation expressed by the relative water/structure velocity and acceleration. i.g.
e. Nonlinear hydrodynamic loading according to the Morison equation is.g. Coupled staticand dynamic analysis of the complete system is in general required to establish the floater motions of deep water systems in terms of mean position and combined wave.Guidance . helps to avoid instability problems often encountered during application of prescribed displacements for such systems. still included. A typical application is analysis of tensioned risers with moderate transverse excursions. Time domain analysis is consequently the primary method of analysis. Combined WF and LF forced vessel motions should be considered in riser analysis if riser dynamics is significantly influenced by low frequency excitation. The analysis represent a fundamental check of the dynamic properties of the riser system and should always be considered as the first step in the dynamic system analysis. 503 Treatment of nonlinearities is the distinguishing feature among available analysis techniques. This combined use of catenary/FE analysis gives a significant reduction in computation time for compliant riser systems and. Engseth et al 1988.note  structure model. The basic idea behind this approach is that the overall influence from bending stiffness is moderate for compliant riser configuration. stationkeeping system and floater comprise an integrated dynamic system responding to environmental loading from wind. treatment of nonlinearities and main area of application are summarised in the following. structural and load linearization). Static equilibrium is ensured by equilibrium iteration at each load increment.or simultaneous with 1 to avoid instability problems if the riser is modelled to be free to translate vertically at upper end. In this approach. These examples illustrates that application order of the defined load components can be decisive for the efficiency and stability of the static solution procedure.DNVOSF201 Dynamic Risers. especially concerning prediction of extreme response. nonlinear time domain analysis based on step by step numerical integration of the incremental dynamic equilibrium equations. user defined reference configuration defining the state of no stress) to obtain the static riser configuration. (i. however. C 500 Dynamic finite element analysis 501 Risers. The situation is somewhat different for top tensioned risers where the effective weight of the riser system is carried by the applied top tension. Load application order should therefor be considered carefully to avoid instability problems. rotations or tension variations or in situations where description of variable touch down location or material nonlinearities are important. damping. perhaps most important. 303 Some computer programs tailor made for slender structure analysis offers an alternative strategy utilising the catenary solution as starting point for FE analysis (e. The nonlinear approach will give an adequate description of all nonlinear effects and will consequently give a good representation of a possible nonGaussian response.and low frequency motions. Nonlinear simulations will typically be needed for systems undergoing large displacements. A representative mean floater position accounting for average environmental forces as well as low frequency (LF) motions is usually applied in riser system analysis.e. linearized time domain analysis based on linearization of the dynamic equilibrium equations with regard to stiffness. The linearized approach is far more efficient than nonlinear analysis and is hence an attractive alternative when hydrodynamic loading is the major nonlinear contributor. inertia. it is obvious that the response characteristics of riser systems in general are nonGaussian.e. it will normally be sufficient to apply a rather crude slender DET NORSKE VERITAS . It should however be noted that the described approach only is applicable to riser systems that do not respond dynamically to LF floater motions. A NewtonRaphson type of equilibrium iteration is applied at each time step. Guidance note: The load components are for compliant riser configurations often applied one by one in the order 134 (2 is irrelevant for compliant riser configurations) or alternatively 1 followed by simultaneous application of 3 and 4. This requires that 2 is applied before. Stochastic linearization for combined wave/current loading is required for irregular analysis. which means that the FE and catenary solutions are close. The WF floater motions are computed in the frequency domain. see D for an introduction to coupled analysis 502 Global riser system analyses are. Frequency domain analysis will always give a Gaussian C 400 Finite element eigenvalue analysis 401 Eigenvalue analysis is used to determine the eigenfrequencies and eigenmodes of the riser system. Eigenvalue analysis is of particular interest in the early stage design of deep water tensioned risers operated from Tension leg and Spar platforms to avoid unwanted resonance dynamics. . Based on the identified nonlinearities. however. structural linearization). damping and inertia forces at static equilibrium position. This means that the system stiffness. the procedure described by Peyrot and Goulois (1979). normally performed considering forced excitation due to wave frequency (WF) floater motions as well as direct wave and current loading. and external forces at static equilibrium position (i.g. frequency domain analysis based on linearization of stiffness.of . application of volume forces (1) and prescribed displacements (3) is replaced by one equilibrium iteration starting from the catenary solution established by e. O’Brien and McNamara 1989).end . 504 Commonly used dynamic FE analysis techniques. In such analyses. waves and current in a complex way. damping and mass matrices are kept constant throughout the analysis and that system displacement vector can be found by a simple back substitution at each time step. October 2010 Appendix A Page 57 302 Each of these load components is in a standard FE approach applied in one or several load increments starting from an initial stressfree configuration (i.
g. see e. The computation time is small when compared to time domain analyses.and slender structure analyses are carried out separately.DNVOSF201 Dynamic Risers. moment. tension in mooring lines as well as tension. This section is dedicated to discussion of analysis methodologies for the complete system involving fully coupled system analysis. Ormberg and Larsen (1997). Computed floater motions are then applied as loading in terms of forced boundary displacements in subsequent slender structure analysis. Nonlinear timedomain analysis considering irregular wave frequency (WF) and low frequency (LF) environmental loading is generally required to give an adequate representation of the coupled floater/slender structure dynamics on nonlinear systems. 202 The computational efforts required for coupled system analysis considering a detailed model of the slender structure including all mooring lines and risers are substantial and should therefore mainly be considered as a tool for final verification purposes. Ormberg et al (1998) and Colby et al (2000). as this contribution is automatically included in the slender structure response. D 100 Combined floater/slender structure analysis General 203 101 Floater. D 200 Coupled system analysis 201 All relevant coupling effects can be consistently represented using a fully coupled analysis where the floater force model is introduced in a detailed Finite Element (FE) model of the complete slender structure system including all mooring lines and risers. 102 The discussion of coupled analysis is mainly focused on application to nonlinear systems (e. For a more comprehensive discussion of techniques for analysis of the complete system. shear. see for instance Ormberg et al (1997). Consistent treatment of these coupling effects is decisive for adequate prediction of floater motions as well as slender structure response for some deep water concepts such as slack. Current loading and damping due to the slender structures (i.g. slender structure response at every time instant. The first step is always a floater motion analysis. October 2010 Page 58 Appendix A response and is therefore in general not recommended for extreme response prediction. waves and current in a complex way. coupled floater motion analysis as well as traditional decoupled floater motion analysis. The LF motions of these systems may be significantly influenced by slender structure coupling effects. The output from such analyses will be floater motions as well as a detailed slender structure response description (e. By careful modelling.g nonlinar restoring characteristics from slender structures) with significant LF floater motions. 302 The most direct way to proceed is to apply time series of combined WF and LF floater motions computed by the floater motion analysis as boundary conditions in the slender structure analyses as shown in Figure A1 (branch a). It will therefore be no need for assessment of the low frequency damping from the slender structure. D.and riser designs. The main application area is fatigue calculations and longterm response statistics to identify design conditions to be applied in time domain analyses. risers and mooring lines comprise an integrated dynamic system responding to environmental loading due to wind. curvature and displacement in risers). Such effects may be of importance for some deepwater mooring line. Floater motion analysis and detailed slender structure analysis are carried out separately in the two latter approaches to achieve computations that are more efficient. It should be noted that this approach yields dynamic equilibrium between the forces acting on the floater and DET NORSKE VERITAS . These techniques can give significantly different results depending on the actual system characteristics. All strategies have in common that the floater motion. D 300 Efficient analysis strategies considering coupling effects 301 Several strategies can be proposed to achieve computational efficiency. The intended application area is typically coupled analysis of FPSO’s and Spar platforms.e. Rooney et al (1990) for application examples for a top tensioned TLP production/injection riser. risers. Risers and critically loaded mooring lines are analysed one by one in the slender structure analyses contributing to computational flexibility as well as a significant reduction in computation time. This approach will also capture possible LF slender structure dynamics as well as influence from LF response (possibly quasistatic) on the WF response.and taut moored FPSO’s (Ormberg et al 1998) and Spar platforms (Colby et al 1999). this approach is capable of predicting floater motions and detailed slender structure response with same precision as the complete coupled system analysis. tethers and mooring lines) may significantly influence the low frequency floater motions of deep water mooring systems. Coupled floater motion analysis in combination with subsequent slender structure analysis as discussed in the sections D 300 – D 400 is generally recommended to achieve a more efficient and flexible analysis scheme.
damping. D 400 Coupled floater motion analysis 401 The primary purpose of coupled floater motion analysis is to give a good description of floater motions. October 2010 Appendix A Page 59 Advanced vessel mode l Simplified slender structure model Vessel Motion Analysis identical to the approach applied in coupled system analysis as described above. The slender structure is consequently assumed to respond quasistatically to LF floater motions. see Figure A1 (branch b). This information can be extracted from model tests of the complete system or from coupled floater motion analysis as discussed in 303 It has been experienced that time histories covering roughly 2025 LF motion cycles is required to obtain adequate damping estimates. 503 Efficient analysis and consistent treatment of coupling effects can hence be achieved by splitting the floater motion analysis in a rather ‘short’ coupled floater motion simulation and ‘long’ decoupled floater motion simulation.DNVOSF201 Dynamic Risers. The numerical solution technique as well as floater force model is however E. the computational efficient decoupled approach allows for long simulations to achieve the required statistical confidence. The LF damping estimated from the coupled floater motion analysis is applied in the decoupled floater motion analysis to obtain consistent treatment of coupling effects.g. Floater motions can be simulated using a coupled – or decoupled approach as described in sections D 400 and D 500. section 6. no bending stiffness etc. LF motions are computed by stepwise numerical integration in time domain.) in the coupled analysis still catching the main coupling effects (restoring. environmental condition is crucial for the LF floater motion analysis. The computation time is small when compared to the coupled floater motion analysis. LF & WF vessel motions D 500 b Select vessel motion representation Decoupled floater motion analysis a Establish "representative" offset (mean and LF) LF & WF vessel motions The purpose of this approach is to compute rigid body floater motions considering static. detailed slender structure response is secondary. E 100 Hydrodynamic loading on slender structures General 101 The hydrodynamic loading on slender structures can be expressed by the Morison equation in terms of the DET NORSKE VERITAS . Ormberg et al (1998) for further details.g. low frequency and wave frequency environmental loading. It should however be emphasised that the damping contribution from the slender structures for some systems is sensitive to the environmental excitation. It can therefore be proposed to apply a rather crude slender structure FE model (crude mesh. The damping estimate should therefore preferably be based on the same environmental condition as considered in the decoupled floater motion analysis. Case studies of deep water mooring systems verify that the floater motions can be accurately predicted with acceptable computation time using a simplified modelling of the slender structures (Ormberg et al 1998). 501 The floater force model is identical to the model applied in coupled analyses as described above.2. 402 This approach gives a significant reduction in computation time due to a reduced number of degrees of freedom in the coupled analyses. and mass). 305 Treatment of coupling effects in the floater motion analysis is decisive for the validity of this approach. Vessel WFmotion transferfunction (RAO) Advanced slender structure model (Model and analyse selected risers and mooring lines one by one) Slender Structure Analysis Slender Structure Analysis 502 Assessment of the LF damping for the actual WF slender structure response LF and WF slender structure response Figure A1 Vessel motion analysis 303 Traditional assumptions can alternatively be applied considering WF floater motions as dynamic excitation while LF floater motions are accounted for by an additional offset. see e. reference is made to e. The slender structures are represented in a simplified way in terms of static restoring force characteristics and a constant LF viscous damping. respectively. it is possible to account for current loading on the slender structures in terms of an equivalent constant force acting on the floater. Furthermore. 304 For guidance regarding calculation of representative floater offset. The restoring force characteristics may include effects from current loading on the slender structures.2 in API RP 2SK. while WF motions normally are computed in frequency domain. If not.
The latter calls for special attention and is discussed separately below.vt & r rt . which cover most situations of practical interest. It is however also most important to keep in mind that the eigenmodes and eigenvalues of the system are influenced by the added mass term in the Morison equation. v n & & rn . Hence. undisturbed or disturbed). it is important to observe the eigenmodes and eigenperiods of the system will be influenced by the added mass term. The crossflow principle is in the latter case somewhat questionable. z Dh y Dhz Dhy & vn . The fluid velocities and acceleration vectors can be found by considering relevant contributions from wave kinematics (regular or irregular.DNVOSF201 Dynamic Risers. added mass of aircans for Spar riser systems). section A.e. The two first terms are in FE implementations included in the external load vector while the latter term (added mass term) is included in the mass matrix. see the SINTEF handbook (1992).1 ) 1 πD2 πD2 f n = ρCn Dh  vn − &n  (vn − &n ) + ρ b Cn & n − ρ b (Cn − 1)&&n r r r D Mv M 2 4 4 πD2 πD 2 1 & & f t = ρCtDDh  vt − rt  (vt − &t ) + ρ b CtM vt − ρ b (CtM − 1)&&t r r 2 4 4 where: fn ft ρ Db Force per unit length in normal direction Force per unit length in tangential direction Water density Buoyancy diameter (i. 103 Formulation of the normal load component is dependent on the actual shape of the crosssection. &rn Figure A3 Definition of local cross sectional axes DET NORSKE VERITAS . see C. E 200 Morison equation for circular crosssections The Morison equation for circular crosssections can be expressed as: (A. Drag and inertia coefficients in tangential direction C tD . equivalent diameter for description of resulting buoyancy on a general riser cross section) Hydrodynamic diameter Fluid velocity and acceleration in normal direction Structural velocity and acceleration in normal direction.g. The fluid and structural velocity and acceleration vectors are in a FE approach known in the global reference frame.C n D M & vt . The Morison equation is therefore discussed separately for circular and doublesymmetric crosssections. These local symmetry axes are denoted y and z in the following and are shown in Figure A3. Added mass contributions should therefore be carefully evaluated as a part of eigenvalue analysis in order to do an adequate assessment of the governing eigenmodes and eigenperiods of the system (e.2. treatment of the Morison type of loading is an essential issue when selecting method of analysis.g. and special attention should be paid to the selection of hydrodynamic coefficients.4. current (constant velocity or velocity and acceleration) or moonpool kinematics. 104 Hydrodynamic loading according to the Morison formulation is a major source to nonlinearities in the response characteristics of slender structures. Figure A2 Pipe with attached buoyancy elements E 300 Morison equation for double symmetric crosssections 301 The Morison equation can be extended to double symmetric cross sections by decomposing the normal velocities and accelerations in direction of the local crosssectional symmetry axes. October 2010 Page 60 Appendix A relative fluidstructure velocities and accelerations.2. C tM 102 Hydrodynamic loading in normal and tangential pipe directions is normally computed independently according to the socalled crossflow (or independence) principle. C n . 201 In addition. Hence. This approach requires that the fluid and structural acceleration vectors are decomposed in the instantaneous normal and tangential pipe directions. CA = C M – 1 is defined as the added mass coefficient. &&t Drag and inertia coefficients in normal direction Fluid velocity and acceleration in tangential direction Structural velocity and acceleration in tangential direction. 202 The described formulation is applicable to bare circular pipes as well as equivalent circular pipe models (e. equally spaced circular cylindrical buoyancy elements on a circular pipe as illustrated in Figure A2). though.
e. &&ny 402 An extensive discussion on the dependence of the hydrodynamic coefficients on several of these parameters can be found in Sarpkaya and Isaacson (1981) and Sumer and Fredsøe (1997). Reynolds number Re = UD/ν. reference is made to the Sintef handbook (1992).145 in Newman 1977) and Ω is the angular velocity. The force resultant will in general not follow the direction of DET NORSKE VERITAS . it is often sufficient to use a simplified consideration to select the coefficients in Morison’s equation. 304 The described model is applicable for modelling of hydrodynamic loading on more complex crosssections such as pipe bundles. combined pipeumbilical cross sections etc. Nevertheless. Cnz D M C tD . Therefore. KC and roughness ratio. relative current number Uc/UM.&rt C ny . the inviscid moment per unit length about the longitudinal axis (i.DNVOSF201 Dynamic Risers. where UM is the free stream velocity amplitude of the oscillatory flow and T is the period of oscillation. October 2010 Appendix A Page 61 302 The force components in tangential and normal yand zdirections can be expressed as: 1 ny (A. The tangential forces can be equally important to the normal forces. projected area) in normal yand zdirections Hydrodynamic reference diameter in tangential direction Fluid velocity and acceleration in normal ydirection Fluid velocity and acceleration in normal zdirection Fluid velocity and acceleration in tangential direction Structural velocity and acceleration in normal ydirection Structural velocity and acceleration in normal zdirection Structural velocity and acceleration in tangential direction Drag and inertia coefficients in normal ydirection Drag and inertia coefficients in normal zdirection Drag and inertia coefficients in tangential direction & v ny . v ny & v nz .2 ) & & ρC D Dhy  vny − rny  (vny − &ny) + ρAb Cnyvny − ρAb (Cny − 1)&&ny r r M M 2 1 & & f n z = ρCn zDh z  vn z − rn z  (vn z − &n z) + ρA bC n zvn z − ρAb (Cn z − 1)&&n z r r D M M 2 1 & & & f t = ρC tDDht  vt − rt  (vt − rt ) + ρAb CtM vt − ρAb (C tM − 1)&&t r 2 f ny = the velocity. where k is the characteristic dimension of the roughness on the body. for instance due to varying flow conditions or lack of information. A more detailed discussion is found in the SINTEF handbook (1992) For riser with buoyancy elements. Note that this model is not applicable for rotational symmetric cross sections. where fn is the natural frequency of the riser.vt & r r ny . where: fny . Keulegan Carpenter number KC = UMT/D. it is important to distinguish between areas where the drag term act as excitation (e. use values for steady flow.g.g.3 p. fnz ft ρ Ab Dhy . tangential direction) can be expressed as & & M t = −m66 Ω + ρA b (Cny − C n z )(v ny − rny )(v n z − &n z ) r M M (A. The last term is the Munkmoment. where Uc is the current velocity and UM is the velocity of the oscillatory motion.2. respectively. A sharp distinction between areas with excitation or damping is not always possible. D is the diameter and ν is the kinematic viscosity. as the correct Morison formulation will not be retrieved.3 ) where m66 is the added moment (see Table 4. C tM 403 Focus should always be given to selecting hydrodynamic coefficients slightly on the conservative side.e. it can be difficult to decide the coefficients based on the abovementioned criteria. In such considerations. As a first approximation. v nz & vt . parts of the riser system insignificantly influenced by wave loading). For circular bare pipes natural choices are CM = 2 and CD = 0. &&nz & & r t . The inviscid force and moment for a general crosssection is discussed by Newman (1977) p 139 E 400 Principles for selection of hydrodynamic coefficients 401 The hydrodynamic coefficients are dependent on a number of parameters: body shape. Their presentation is concentrated on circular cylinders with different Re. & r r nz . 303 In addition. where U is the free stream velocity.7 – 1. a sensitivity study should always be performed to support rational conservative assumptions when a high level of confidence is required. C ny D M C nz . wave zone) and areas where the drag term act as damping (e. For practical purposes. Slightly high/low values should be selected for areas where the drag term act as excitation/damping.Dhz Dht Force per unit length in normal y – and zdirections Force per unit length in tangential direction Water density Crosssectional buoyancy area Hydrodynamic diameter (i. reduced velocity U/fn D.2. section A4.0. roughness ratio k/D.
accelerated uniform flow and slug flow. This means that the effective tension is the overall governing stiffness parameter for the vast majority of slender structures. see e.and internal fluid densities Acceleration of gravity External. 102 The hydrostatic pressure is treated by the effective weight/effective tension concept as described in section G for static and dynamic analyses. H 200 Steady flow 201 Steady flow corresponds to e. Patel and Seyed 1989): the effective tension is not affected by the steady flow. Hence. submerged weight of pipe including content) Mass of pipe External.4 ) True wall tension (i. The effect of a steady flow on a static riser configuration can be summarised as (Fylling et al 1986. the nonconservative pressure force model is replaced by an equivalent conservative volume force model). Te = Tw − A i Pi + A o Po We = m p g + A i ρ i g − A oρ o g Where : Te Tw A o . use of anti fouling coating) as well as structural behaviour (e. In dynamic analyses. Skomedal (1990) for further details.ρi G P0 . the effective weight corresponds to the submerged weight of the effective mass). Marine growth 101 Marine growth on slender structures will influence the loading in terms of increased mass. the mass of the internal fluid is included in the effective mass defined as the crosssectional mass including content m = e mp + ρ i A i . Hydrostatic pressure loading 101 Loading due to external and internal pressure acting on a pipe section is normally treated in terms of the effective weight/tension concept (Sparks 1984): 101 Loading due to internal fluid flow is normally included in terms of the hydrostatic pressure and mass contribution in global riser analyses.g.4 depending on the type of organisms.g.DNVOSF201 Dynamic Risers. the hydrodynamic coefficients should be assessed with basis in the roughness specified for the marine growth. less marine growth is normally considered for slender structures with significant dynamic displacements). the effective tension formulation is applicable to any general shaped volume stiff cross section (i.g.e.Ai We Mp ρ o . it is recommended to increase mass. cross sectional volume is not influenced by the hydrostatic pressure). H. 102 Hydrostatic pressure is acting normal to instantaneous orientation of the pipe and can hence be classified as a nonconservative loading (follower load).e.and Internal pressure 103 Additional loading (centrifugal and coriolis forces) is however introduced due to the internal fluid flow through the pipe. October 2010 Page 62 Appendix A F. global buckling and stability is governed by the effective tension. In addition. The relative density of marine growth is in the range of 11. H 100 General Internal fluid flow G. 103 The physical significance of the effective tension can be summarised as: the geometric stiffness is governed by the effective tension.Pi Effective Tension (A. main advantage of the effective weight/tension formulation is that loading due to hydrostatic pressure is represented by vertical conservative forces (i. 105 In FE analyses.e. normal production with a homogenous fluid flow with constant velocity through the pipe. diameter and hydrodynamic loading. axial stress resultant found by integrating axial stress over the crosssection) External.e. The DET NORSKE VERITAS . pipes inside pipes) by summation of effective tension and effective weight contributions from all pipes. no fluid flow through the pipe is considered).g. roughness and depth variation of thickness. The marine growth characteristics are basically governed by the biological and oceanographic conditions at the actual site. The significance of the additional loading is briefly discussed in the following for three fluid flow categories: steady flow.e. 103 The thickness of marine growth to be included in design analyses will in addition be dependent on operational measures (e. buoyancy diameter and drag diameter according to the specified depth variation of marine growth. This is of vital importance for efficiency and stability in computer implementations as pressure integration over the deformed pipe geometry is avoided. 104 Field measurements at the actual location combined with field experience regarding the extent of marine growth on similar structures will therefore be the reference for specification marine growth to be considered in design analyses 104 The effective tension formulation is also directly applicable to multipipe crosssections (i. 102 Site dependent data for marine growth are normally specified in terms of density. regular cleaning.e. the effective mass is consistent with the effective weight (i. This model is formally correct for static and dynamic analyses of risers conveying a hydrostatic internal fluid (i.and internal cross sectional areas Effective weight (i.e.
TLP. close to supports.e. 202 The effective tension including the effect of steady internal fluid flow can be expressed by: 403 The load model accounting for the slug flow must in (A.g. Severe slug flow is typically related to the latter condition. touch down area).g Semi. Furthermore. slug frequencies close to governing eigenfrequencies of the riser system should be considered carefully. e. centrifugal force and the coriolis force as the slugs travel through the pipe (the latter term is often considered less important and consequently omitted). In addition. Model tests of a submerged Ushaped flexible hose considering a range of fluid velocity and support motion combinations have been conducted to investigate this effect (Sintef 1992). Fylling et al (1986).g. or frequency domain results. weight. i. The global response to this excitation can be predicted by nonlinear time domain analysis using the load model as described for slug flow conditions in H 400. This effect is negligible for compliant riser configurations. H 300 Accelerated uniform flow 301 Sudden stop or start of the internal fluid flow will cause an accelerated uniform flow. All relevant data characterising the slug flow is input to the global response analysis and is typically established by numerical multiflow simulations and/or laboratory measurements supported by field experiences. 102 Riser support motions may be obtained in several ways. forced support displacements. it is convenient to parameterise the slug flow in terms of velocity. but may have some significance for calculation of the vertical upper end position of top tensioned risers. see e. Response due to slug flow is expected to be most pronounced for deep water compliant gas riser configurations (e. and the static configuration is only affected by the axial elongation caused by the steady flow. Transient slug conditions can in addition occur during startup. For further details and examples of practical application see e. Forced Floater Motions 401 Slug flow is characterised by an alternating flow of liquid slugs and gas pockets.g using time series from model testing. 203 The steady fluid flow will introduce additional loading on a curved pipe exposed to dynamic excitation by e. Implementation of an adequate load model due to such flow conditions is needed for prediction of the global riser response due to slug flow by nonlinear time domain analyses. Passage of large slugs through a gas riser may also introduce significant quasistatic change in riser configuration due to change in effective weight by the slug mass. Steady flow is hence seen to modify the wall tension in a similar manner to internal pressure. length and density of each individual slug as well 104 If LF displacements/rotations are of importance for determination of resulting riser responses the optimum DET NORSKE VERITAS .. 103 The frequency domain solution implies selecting an appropriate quasistatic offset (mean + slowly varying) and heel/tilt (e. the velocity. “Steady state” slug flow can be classified as either hydrodynamic slugging or terrain induced slugging governed by the site specific elevation profile of the flowline. Special care has to be exercised when transferring the RAO's from e. the slug flow through the pipe is not influenced by the motion of the riser). 101 Forced floater motions are defined as displacements imposed on the riser due to motions of the surface floater. WF floater motions are usually given as RAO's (response amplitude and phase angle). Possible slug flow excitation should therefore always be carefully evaluated for deepwater compliant riser configurations. or coupled/decoupled analyses. It was concluded that dynamic response was insignificantly influenced by the steady fluid flow.g a motion analysis program to a purpose made riser analysis program. depending on riser system and floater being considered. by increasing the wall tension and leaving the effective tension unchanged. as the slug frequency defining the time interval between successive slugs. metallic lazy wave configurations). For a more detailed discussion of slug flow characteristics as well as review of numerical simulation techniques to predict the slug flow.DNVOSF201 Dynamic Risers. For such applications. This flow pattern will result in a transient inplane excitation of the riser system.g.g. October 2010 Appendix A Page 63 the steady flow causes an increase in the true wall tension and a corresponding axial elongation. The most appropriate method (time/frequency domain) has to be selected. Spar) and superimposing the wave frequency (WF) motions. 402 Slug flow can hence be considered as a time dependent variation of internal flow velocity and density at any location along the riser. Ship). These forced displacements may be introduced at several elevations on the riser depending on type of floater (e. All these quantities may in general be considered as stochastic variables. 404 Dynamic effects due to slug flow is normally most pronounced in areas along the riser with high curvature due to the centrifugal load component (e. Spar. hog/sag bend. operational changes of flow rate and pigging operations.5 ) general include contributions due to mass. It should be observed that the increase in wall tension ∆ TW = ρ i A i v i2 is independent of curvature and flow direction. Furthermore. H 400 Slug flow I. Patel and Seyed (1989) and Sanderson et al (1999) Te = Tw − Ai Pi + Ao Po − ρi A v 2 i i Where νi is the steady internal fluid velocity. indicating that this loading type is of less importance for riser systems. density and length will in general change (according to deterministic considerations) as the slug passes through the riser. Burke and Kashou (1995).g. Slug volumes seen during pigging are usually the largest slug volumes seen during normal operation.g. it is assumed that the slug flow and riser motions can be treated independently (i.e.
Typically. high ringing accelerations of a TLP may have to be evaluated in case of heavy riser components (trees) at deck level.e. is governing for the excitation forces.e. This approach requires that the entrapped water is included in the hydrodynamic model used to compute the floater motion characteristics. the inertia contribution due to fluid acceleration.and stiffness proportional damping coefficients. Deep draft floaters and shell Spar platforms) J 200 Moonpol kinematics 303 Sensitivity studies considering riser dynamics and eigenvalues should always be performed to support decisions regarding choice of hydrodynamic coefficients to yield conservative response estimates. It should also be observed that the drag term will act as damping. i.g.e. α1 and α2 are denoted the mass. The outputs from such programs are RAO’s for disturbed kinematics (velocity components) consistent with the floater motion RAO’s. (A. A simplified model for the moonpool kinematics can be obtained by assuming that the entrapped water follows the hull motion rigidly. K. This disturbance is most easily determined by use of radiation/ diffraction analysis programs. Fluid velocity and acceleration components can then be found at any location in the moonpool by straightforward transformations of the hull motions (i. This formulation is applicable for FD as well as TD analysis.7 ) This means that the global damping matrix (C) is found as a linear combination of the global mass (M) and global stiffness (K) matrices. 302 The riser motions relative to the moonpool are to a large extent governed by how the riser is supported inside the moonpool. in terms of transfer functions for moonpool kinematics consistent with the hull motion transfer functions). translations and rotations at a specified location on the floater. For large volume floaters and risers located close to e. The excitation force is hence not very sensitive to the CD and CM values due to the small relative motion between the fluid and the riser (see equation above). which allows for a simple expression of the modal damping ratio ξ (i. C = α1 M + α 2 K (A. J. Special attention should be focused on possible resonant modes of the entrapped water. damping relative to critical) at angular frequency ω: DET NORSKE VERITAS . Simplified frequency domain analyses can however be performed to assess the LF riser responses/ importance. Assuming that the entrapped water follows the hull motion rigidly.g. 301 Uncertainty is connected to the hydrodynamic coefficients applicable inside a moonpool. The latter approach allows for consistent treatment of moonpool kinematics due to simultaneous WF and LF floater motions. The motivation for adopting this model is mainly due to computational conveniences. Such calculations will however require a very careful modelling to achieve a realistic picture in case of complicated moonpool geometry and/or multiple risers in the moonpool.e. shell Spar). J 100 Hydrodynamic loading in moonpool General 101 Treatment of hydrodynamic loading on slender structures in the moonpool may be of vital importance for some floater concepts (e. 201 Kinematics of the entrapped water in the moonpool area can in principle be treated in the same way as the disturbed wave kinematics described in section I (i.DNVOSF201 Dynamic Risers.6 ) where v H . respectively. centre of gravity is typically selected as motion reference point). K 100 Structural damping Global Rayleigh damping model 101 The Rayleigh damping model commonly adopted for description of structural damping: 202 Due to the complexity of the problem. ringing and springing for TLP's) will usually not be of concern for the riser system. This has to be evaluated on a case by case situation. this disturbance must be determined and accounted for in design. it is often desirable to apply a simplified model in practical calculations. A slightly low value is recommended as a conservative estimate. J 300 Hydrodynamic coefficients 105 High frequency (HF) motions (e. The “Froude Krylov”term. It is recommended that models of moonpool kinematics are verified against model tests. acceleration and direction). The riser motion in transverse moonpool direction will typically be constrained at several locations along the riser (e. October 2010 Page 64 Appendix A solution is to perform time domain analyses (WF and LF included). Accurate assessment of the drag and inertia coefficients may therefore not be necessary for adequate modelling of hydrodynamic loading for this support condition.g columns/pontoons. v H are the hull velocity and acceleration & components normal to the riser. One of the advantages is that the damping matrix is orthogonal with respect to the eigenvectors of the system.g. The CM value will however influence the eigenvalues of the riser system. the hydrodynamic loading in the normal direction can be expressed as: fn = 1 n & ρC D  v − r  ( vH − &n ) r 2 D h H n πD2 πD2 & + ρ b v H + ρ b (C n − 1)(v H − &&n ) r M 4 4 106 The presence of the floater gives rise to changes in the fluid kinematics (velocity.
8 ) stiffness proportional formulation would hence give a low damping when most needed to stabilise the solution. For combined WF and LF response. α2 is selected to give a realistic damping level at the dominating dynamic frequency. respectively. However. The following extensions of the Rayleigh model have been proposed: local spatial damping to facilitate specification of different damping levels in different parts of the structure. the damping matrix is kept constant throughout the analysis). Kashou S F (1995) “Slug Sizing/Slug Volume Prediction. June 1998.DNVOSF201 Dynamic Risers. applies to all global degrees of freedom and is hence limited to specification of an overall realistic structural damping level. In practical implementation. e. for some applications it is desirable to have more flexible damping models. ISBN no 8259572664. see Bech et al (1992).g. API RP 2SK “ Recommended Practice for Design and Analysis of Station Keeping Systems for Floating structures” Second Edition.9 ) 104 The stiffness proportional damping is seen to increase linearly with the frequency. This is in particular the case for low effective tension problems using a stiffness proportional damping model. axial. For further details. Guidelines and Handbooks API RP 2RD “Design of Risers for Floating Production Systems (FPSs) and Tensionleg Platforms (TLPs)” First Edition. For such problems. In practical implementation. The simple closed form expression for modal damping presented above is therefore formally not applicable for specification of the damping level. L. However. dynamic frequency range) and for all practical purposes no damping in the low frequency range (i. Sødahl N. as damping is not considered to be of significance for prediction of the quasistatic low frequency response. 201 The global Rayleigh damping model. 105 The Rayleigh damping model is applicable to all relevant global analysis strategies (nonlinear and linearized time domain and frequency domain analyses). 103 The proportional damping model applies to all global degrees of freedom and will hence give the same damping level in all parts of the structure. the structural damping at angular frequency ω is given by: ξ = α2 ω 2 (A. bending and torsional deformation modes).e.e. this can be achieved by specification of damping coefficients for element subsets before assembly of global system matrices. State of the art review and simulation” OTC paper no 7444. specification of different damping levels in different local deformation modes (i. The main drawback with the local damping models is that the orthogonality with respect to the system eigenvectors is lost. Okkenhaug S (2000) “Coupling Effects for a Deepwater Spar”. bending. L 200 Technical references 106 The latter approach is in general recommended due to better performance regarding numerical stability and computational efficiency. SINTEF (1992) “FPS 2000 Flexible Pipes and Risers: Handbook on Design and Operation of Flexible pipes “. Katla E. this can be achieved by specification of local damping coefficients related to axial. Sødahl N (1992) “Structural Damping in Design Analyses of Flexible Risers” Marinflex’92. α2 is selected to give a realistic damping at the dominating wave frequency (typically peak period in the wave spectrum). December 1997. OTC 12083 DET NORSKE VERITAS . Explicit expressions are given in standard textbooks on structural dynamics. In practical applications. Skallerud B. and torsional deformations in local element system before transformation and assembly of global system matrices. It should also be observed that the mass proportional damping would give damping due to rigid body motions.e. L 100 References Standards. α1 and α2 can be selected to give required modal damping ratios ξi and ξj at specified angular frequencies ωi and ωj . November 1992. London. application in nonlinear time domain analyses call for special attention. K 200 Local Rayleigh damping models 102 In practical calculations. damping is typically needed to stabilise the solution when geometric stiffness is lost due to low dynamic effective tension. Burke N E. The mass proportional damping is therefore normally neglected for compliant structures undergoing large rigid body motions (Sintef 1992). quasistatic frequency range). as discussed in the previous section. constant formulation which means that the Rayleigh damping matrix is based on mass and stiffness matrices at static equilibrium position (i. This is considered to be a realistic model.e. The global mass and stiffness matrices will in a nonlinear time domain analysis be a function of the instantaneous nodal positions and rotations. 202 The local models can be applied in combination and used in updated as well as constant damping formulations as discussed in the previous section. Sintef report STF70 A92006. October 2010 Appendix A Page 65 1α ζ = 1 + α 2ω 2 ω (A. The updated Bech A. The following implementations of the Rayleigh damping model can hence be applied: updated formulation which means that the Rayleigh damping matrix is based on instantaneous mass and stiffness matrices. Colby C. Clough and Penzien (1975). The proposed model will hence give damping in the wave frequency range (i. For a stiffness proportional damping model.
Patel M H. Engseth. Houston. Ormberg H. E (1990) “Buckling Calculation of Offshore Platform Conductors” 1st European Offshore Mechanics Symposium. Isaacson M (1981) “Mechanics of Wave Forces on Offshore Structures” Van Nostrand Reinhold Company 1981. Vol 11. Nordsve. A. O’Brien.(1994): ”Design and Installation of Auger Steel Catenary Risers. Schnittker. C.. Trondheim. Fylling I. Seyed F B (1989) “Internal flowinduced behaviour of flexible risers” Engineering Structures.Vol 106.. Los Angeles.. OTC. ISOPE 1996. BOSS 1997. 7620. 1989. Computers and Structures. Offshore Technology Conference. 1979 Phifer. OMAE 1997. and Langner. D. Skomedal. Sødahl N. Houston. Karunakaran.. Larsen C.1989. Bech. C M (1988) “Efficient Method for Analysis of Flexible Risers” Proc.. Rooney P. Proc. October 2010 Page 66 Appendix A Clough R W. Engebretsen K. Passano E (1995) “Reducing Simulation Lengths in Extreme Response Estimation of a Flexible Riser” Proc. Steinkjer O (1998) “ Efficient Analysis of Mooring Systems using Decoupled and Coupled Analysis” OMAE 98. (1989) “Significant Characteristics of Threedimensional Flexible Risers Analysis” Engineering Structures.. R. World Scientific Publishing. No 4. Larsen. J F. 1990. a Case Study” OTC 6435. ASME. Sanderson N. Sødahl N.M (1979) “Analysis of Cable Structures”. Penzien J (1975) : “ Dynamics of Structures” McGrawHill 1975. P P. Peyrot. Ormberg H. P J. Vol 10. Proc.DNVOSF201 Dynamic Risers. Kopp.” Journal of Energy Resources Technology. Goulouis. McNamara. J (1997) “Hydrodynamics around cylindrical structures”.. Fylling I J .F.” Paper No. Pressure and Weight on Pipe and Riser Deformations and Stresses. Olufsen A (1996): “An Efficient Metal Riser Configuration for Ship and Semi Based Production Systems” Proc. Rio de Janeiro. Houston..C.W.G. Sarpkaya T . J. Fredsøe. Sparks. Cambridge: The MIT press. BOSS 1988. D.M (1989) “ Design Procedure for Bending Stiffeners in Flexible Riser Systems” PRADS 89. Ormberg H. Newman J N (1977) “Marine Hydrodynamics”. Larsen K. E. of 2nd workshop on subsea pipelines. DET NORSKE VERITAS . Sødahl N. Moros T. A. Vol 11. Allen. Pettersen D J (1990): “TLP Rigid Riser. Bech A (1986) “ On the Effects of Slug Flow on the Response of Flexible Risers and Submerged Hoses” IBC Technical Services. A. B M. D A and Sturdevant. Fairhurst P. OMAE 1995. Sødahl N. (1997): “Coupled Analysis of Floater Motion and Mooring Dynamics for a Turret Moored Tanker” Proc. N T. London 1986. OTC. L A (1989): “Design and Application of a Dynamic Flexible Pipe Array for Placid's GC 29 FPS” Paper No 6164.H. C P (1984): “The Influence of Tension. Swanson.H. Espinasse. (1997): “Coupled Analysis of Vessel Motions and Mooring and Riser System Dynamics” Proc. Sumer. Larsen K. A.Harrison G (1999): “Design and Practical Considerations of Steel Catenary Riser and Pipeline Systems” Proc.
B. These behave as pre DET NORSKE VERITAS . the component is virtually at the end of its useful life and will normally be withdrawn from service if repair is not possible. Structural discontinuities. Identify locations to be assessed.. Consequently. Identify fatigue strength data. Improve fatigue capacity using. Extensive references are given to the DNV RPC203. Local joint stress analysis. Although crack propagation rates can change from one material to another and from one environment to another.g. the crack initiation period represents the bulk of the total fatigue life. Table B1 Task Define fatigue loading.C 200. the fatigue strength is relatively unaffected by material tensile strength because the bulk of the fatigue life of a welded joint is spent in the propagation phase.E on fatigue assessment of risers subjected to repeated fluctuations and provide details on fatigue analyses methods recommended in Section 4. construction detail and fabrication among others. bolts). Calculate shortterm nominal stress range distribution at each identified location. In the case of welded joints however. Fatigue Capacity SN Curves E 100 A. connectors. October 2010 Appendix B Page 67 APPENDIX B Contents A. 303 The difference in the crack initiation phase of parent material and welded joints has significant effects on overall fatigue performance. repairs. weld toe/root discontinuities are generally present. For a general introduction to methodology for fatigue damage assessment. 302 In the case of welded joints. Narrow Band Fatigue Damage Assessment A 400 Methods for fatigue damage assessment D. Fatigue Strength Analysis. This is particularly noticeable at high fatigue lives where the fatigue crack initiation period may exceed 95% of the fatigue life. Identify thickness correction factor Fatigue analyses Summary of a typical fatigue assessment procedure Comment Based on operating limitations including WF. joints (girth pipe welds. Global riser fatigue analysis. the bulk of the fatigue life of a welded joint can be attributed to fatigue crack propagation. once a fatigue crack has grown to a detectable size. etc. In general.: more refined stress analysis fracture mechanics analysis. reference is made to DNV RPC203 Sec 1. change detail geometry change system design. A 200 Application 201 The assessment procedure assumes that the riser has been designed in accordance with all other requirements in this standard. Fatigue analysis procedures C. In the case of unwelded components (e. the fatigue life of a component can be broken down into two phases: Crack initiation and propagation. anode attachment welds. A 100 Objective General 101 This objective of this Appendix is to support section 5. Apply thickness correction factor to compute resulting fatigue stresses Calculate accumulated fatigue damage from weighted shortterm fatigue damage. LF and possible VIV load effects. there is no consistent trend with regard to tensile strength. Determination of the hotspot SCF from parametric equations or detailed finite element analysis. General A 100 A 200 A 300 A 400 B 100 B 200 B 300 C 100 C 200 C 300 D D D D 100 200 300 400 Objective Application Fatigue design Methods for fatigue damage assessment General Basic fatigue damage methodology Global fatigue analysis procedures General Narrow Band Gaussian Fatigue Damage Narrow Band NonGaussian Fatigue damage General Cycle counting Semiempirical Solutions Analytical Solutions for Bimodal Spectra General FATIGUE ANALYSIS existing cracks. A 300 Fatigue design 301 In general. Wide band Fatigue Damage Assessment 401 A typical sequence in fatigue design of a riser is shown in Table B1. Further actions if too short fatigue life. seamless pipes and machined components). In the case of machined components. SNcurve depend on environment.3. the fatigue strength of an unwelded component increase with material tensile strength due to the increased initiation life associated with higher strength materials. weld profiling or grinding improved inspection /replacement programme E.DNVOSF201 Dynamic Risers.
4) Where n(Si ) is the number of stress cycles with range Si and N(Si ) is the number of stress cycles to failure as expressed by B. The probabilities of occurrence for all seastates within the block are lumped to the selected seastate.E+03 NSW 1 E+04 1.E+08 1. for a given constant stress range.m1 ) Stress Range. is a function of the actual structural design and hence also related to SN curve. The thickness exponent.3) Figure B1 Basic definitions for twoslope SNcurves 204 The Where a and m are empirical constants established by experiments.E+07 1. peak period and wave direction. k 203 Bilinear (twoslope) SN curves in loglog scale are also frequently applied for representation experimental fatigue capacity data. N.DNVOSF201 Dynamic Risers. The expected fatigue damage per unit time can for a linear SN curve in loglog scale be expressed as: DET NORSKE VERITAS .5) a ⋅ S− m 1 N = 1 −m 2 a 2 ⋅ S S > Ssw S ≤ Ssw D L = ∑ Di Pi i =1 Ns (B. 102 A general approach for calculation of wave. the fatigue damage is computed for each selected shortterm seastate for all the blocks.E+06 1.1) m1 and m2 are fatigue exponents (the inverse slope of the bilinear SN curve) and a 1 and a 2 are characteristic fatigue strength constant defined as the meanminustwostandarddeviation curve. Ssw is the stress at intersection of the two SNcurves given by: Where DL Ns Pi Longterm fatigue damage Number of discrete sea states in the wave scatter diagram Sea state probability.E+05 1. t ref =25mm. For further details reference is made to DNV RPC203. The former two are addressed in the following. i. see DNV RPC203 Sec 2. MinerPalmgren rule is adopted for accumulation of fatigue damage from stress cycles with variable range: 202 The stress range to be applied in fatigue damage calculations is found by application of a stress concentration factor as well as a thickness correction factor to the nominal stress range : D=∑ i n( Si ) N(S i ) (B.7) t S = S0 ⋅ SCF ⋅ 3 t ref k (B. while the latter is described in Appendix E. S 100 SSW (a2.e.Tp . S: 10 N = a S−m or equivalently: (B.6) Nsw is the number of cycles for which change in slope appears. P(Hs . and section E for further details. a single seastate is selected to represent all the seastates within the block. i. N log(N) = log(a ) − m log(S) (B.and lowfrequency fatigue damage contributions is based on application of the following procedure: the wave environment scatter diagram is subdivided into a number of representative blocks.e. B 100 Fatigue analysis procedures General Where: S0 Nominal stress range SCF Stress concentration factor 101 Three different contributions to fatigue damage should be addressed: The waveinduced.θ ) i Short term fatigue damage Ssw = 10 log( a1 ) −log( Nsw ) m1 (B. k.E+09 1 E+10 No of cycles. the lowfrequency and the vortexinduced stress cycles. the weighted fatigue damage accumulation from all seastates can be expressed as: t3 t Thickness correction factor ref The thickness correction factor applies for pipes with a wall thickness t3 greater than a reference wall thickness. of the (B. within each block.3. Log(Nsw ) is typically 67. 1000 Di (a1. Normally parameterised in terms of significant wave height. October 2010 Page 68 Appendix B B.2) 1 1.m2 ) B 200 Basic fatigue damage methodology 201 The basic fatigue capacity is given in terms of SN curves expressing the number of stress cycles to failure.
Fatigue close to upper termination is normally governed by WF stress cycles while LF response may be of significance close to the seafloor termination. The relative DET NORSKE VERITAS . 305 Adequate fatigue life shall be documented for all parts of the riser system. the splash zone is normally a critical area for top tensioned as well as compliant riser configurations mainly due to WF bending stress cycles. hydrodynamic loading. and considerations regarding resonance dynamics and combined WF and LF fatigue damage are of special importance for spar risers (in particular for integral aircan solutions). 302 The shortterm fatigue conditions should be selected carefully to give an adequate representation of the stress cycles for the lifetime of the riser system. C. Description of wave loading up to actual wave elevation is of vital importance for accurate prediction of fatigue damage Due regard should also be given to possible disturbances in the wave kinematics caused by the presence of the floater.9) Equation (B. results are sensitive to mesh size as well as wave kinematics). Soil properties. Accurate modelling of boundary conditions and stiffness properties is required (e. The adequacy of the mesh applied in the touchdown area should also be confirmed by sensitivity studies. fatigue damage of metallic risers are given in the following: the areas close to upper/lower termination of top tensioned risers will normally experience significant dynamic bending stress variation. taper joints.as well as low frequency (LF) stress cycles. B 300 Global fatigue analysis procedures 301 The basis for fatigue damage calculations is global load effect analyses to establish the stress cycle distributions in a number of stationary shortterm environmental conditions. Adequate results can hence be obtained by use of linearized time domain. the degree of nonlinearity involved is generally smaller. Critical locations are typically close to riser supports in the hull area. However. Time domain analyses supported by sensitivity studies to confirm adequacy of load model is recommended (i. Compared to extreme response analysis.or frequency domain analyses in many cases.9) can also be applied to compute the longterm fatigue damage directly from the longterm distribution of stress cycles. reference is made to Annex C.to moderate seastates with high probability of occurrence rather than a few extreme seastates. resonance dynamics and floater motion characteristics. It is always recommended to do an assessment of the relative contributions from WF and LF stress cycles to the fatigue damage to support rational decisions regarding choice of method of analysis.g. (B. For a bilinear SNcurve in loglog scale the corresponding expression becomes: importance of WF and LF fatigue damage is strongly system dependent and will in addition vary significantly with the location along the riser.8) and (B. The expected fatigue damage is hence directly related to the mth order moment. mesh size and mean floater position are important for prediction of fatigue damage. Special attention should be given to possible LF stress cycles at the keel joint. This is because the main contribution to the total fatigue damage in most cases comes from low.to moderate seastates. LF fatigue damage may be disregarded if it is documented by proper analyses that the LF fatigue damage is negligible when compared to WF fatigue damage. any use of simplified analysis methodology shall be verified against nonlinear time domain analyses. The selection must be based on a thorough physical knowledge regarding static. Each cycle’s range is assumed to be twice the value of the corresponding value of the local stress maximum.and dynamic behaviour of the riser system with special attention to FE modelling. Time domain analyses are generally recommended together with sensitivity studies to support rational conservative assumptions regarding soil properties. yielding: 304 The fatigue damage will generally have contributions from wave frequency (WF). For an introduction to methodology for establishment of longterm response distributions. D= f0 a2 Ssw ∫ 0 s m 2 f S ( s) ds + f0 a1 ∞ Ssw m ∫ s 1 f S (s) ds (B.DNVOSF201 Dynamic Risers.e. E[S m] (or µ m) of the stress cycle PDF.8) and (B.9) constitutes the basic formulation for assessment of the shortterm fatigue damage in each stationary environmental condition as expressed by (B. seafloor touchdown area is a critical area for steel catenary risers and other proposed compliant riser configurations. C 100 Narrow Band Fatigue Damage Assessment General 101 The basic assumption in narrowband fatigue damage estimation is that the stress cycles (S) can be determined directly from the stress maxima (Sa).g. soil properties for fatigue analysis in the touchdown area of SCR’s) 303 Fatigue analysis will normally involve global load effect analyses in a number of low. Sensitivity studies should be performed to support rational conservative assumptions regarding identified uncertain parameters (e. The general principles for selection of analysis methodology and verification of simulation model as outlined in Annex A and Annex D respectively should be adhered to.1). The WF floater motions as well as direct wave loading on the riser govern WF fatigue damage.8) Where f0 is the mean number of stress cycles per unit time and fS(s) is the probability density function (PDF) for the stress cycles. October 2010 Appendix B Page 69 f ∞ f D = 0 ∫ s m f S ( s) ds = 0 E Sm a 0 a [ ] (B. stiffness characteristics of flexjoints etc). Examples of critical areas wrt. while the LF floater motions govern LF fatigue damage.
16) (B. ⎜ sw ⎟ ⎬ ⎜ ⎟ 2α a2 β ⎠⎝ ⎪⎝ ⎠ ⎪ ⎩ ⎭ (B.10) The standard deviation.18) where sa is the local stress maximum and σ is the standard deviation of the stress response process. and where log(a1 ) and m1 are the intercept and inverse slope of the right leg of the bilinear SN curve. σ = λ0 1 2π (B. ⎜ a2 2 ⎠ ⎜2 2σ⎟ ⎪ ⎪⎝ ⎝ ⎠ ⎭ ⎩ f0 ⋅ 2 2 σ m1 2 Note that the Rayleigh distribution in (B. β: shape) are linked to the ˆ ˆ statistical moments μ. Sa. x) = ∫ e −t t ϕ −1dt G2 (ϕ . f0 of the stress response process.14) ( ) (B. of the stress response process. The Weibull probability density function is given by: ( ) (B.20) where G1 and G2 are the complementary incomplete Gamma function and incomplete Gamma function. October 2010 Appendix B Page 70 S = 2 ⋅ Sa (B. 302 The fatigue damage per unit time in the general case of a bilinear SNcurve can then be expressed analytically as follows: β m f 0 ⋅ (2α ) 1 ⎧⎛ m1 ⎞ ⎛ S sw ⎞ ⎫ ⎪ ⎪ ⎟. respectively G1 (ϕ . the number of stress cycles per unit time is given directly by the zero crossing frequency. 202 For a linear SNcurve (in loglog scale) the fatigue damage per unit time can be expressed as: D= m f0 ⎛m ⎞ 2 2 σ Γ ⎜ +1⎟ a ⎝2 ⎠ C 300 Narrow Band NonGaussian Fatigue damage 301 For time domain analyses. the twoparameter Weibull distribution model is frequently employed as a generalisation of the Rayleigh distribution for the local maxima (i. ⎜ sw ⎟ ⎬ ⎜2 2σ⎟ ⎪ a1 2 ⎠⎝ ⎪⎝ ⎠ ⎭ ⎩ 2 m2 ⎧ ⎫ f0 ⋅ 2 2 σ ⎪⎛ m 2 ⎞ ⎛ Ssw ⎞ ⎪ ⎟ ⎬ + ⋅ G 2 ⎨⎜1 + ⎟. DET NORSKE VERITAS .11) λn = ∫ ω n Sσσ (ω )dω 0 ∞ (B. σ from time domain simulations.11) is obtained for The Weibull distribution may be fitted to the shortterm (or longterm) distribution of the local maxima.19) 203 For an SN curve which is bilinear on loglog scale.13) ⎛ ⎛ s ⎞β ⎞ β f S ( sa ) = α −β β sa −1 exp⎜ − ⎜ a ⎟ ⎟ ⎜ ⎝α ⎠ ⎟ ⎝ ⎠ β=2 and α = 2σ (B. is defined by a Rayleigh probability density as: λ2 λ0 where λn is the nth order spectral moment of the response. σ and zero crossing frequency f0 are hence given as: Furthermore.15) These equations can be used to establish moment estimates of the distribution parameters with basis in sample estimates ˆ ˆ μ.12) where Γ(⋅) is the gamma function given by Γ(ϕ) = ∫ e− t t ϕ −1 dt 0 ∞ (B.. σ for the local maxima as follows: ⎛ 1⎞ ˆ ⎜ μ = αΓ⎜1 + ⎟ ⎟ ⎝ β⎠ ⎛ 2⎞ ⎛ 1⎞ ˆ ⎜ ⎟ σ = α Γ⎜1 + ⎟ − Γ⎜1 + ⎟ ⎜ β⎟ ⎝ ⎠ ⎝ β⎠ 2 ( ) (B. given by ⎛s f S (s a ) = ⎜ a2 ⎜ ⎝σ ⎛ −s2 ⎞ ⎟ exp ⎜ a ⎟ ⎜ 2σ 2 ⎠ ⎝ ⎞ ⎟ ⎟ ⎠ (B.DNVOSF201 Dynamic Risers. The Weibull distribution parameters (α: scale. the distribution of local stress maxima.21) 204 The fatigue damage is hence directly expressed by the standard deviation and zerocrossing frequency of the stress response process This formulation is of special convenience for frequency domain analyses where results from the global analyses are expressed in terms of the autospectral density.17) C 200 201 Narrow Band Gaussian Fatigue Damage f0 = If the stress response process is assumed to be narrow banded and Gaussian. Sσσ(ω). the fatigue damage becomes D= ⎧ ⎞ ⎫ ⎪⎛ m ⎞ ⎛ S ⎪ G 1 ⎨⎜1 + 1 ⎟. log(a2 ) and m2 are the intercept and inverse slope of the left leg of the bilinear SN curve. and SSW is the stress range at the knee of the bilinear SN curve.e. x) = ∫ e −t t ϕ −1dt 0 x x ∞ (B. for NonGaussian stressresponse processes). ⎜ D= G1 ⎨⎜1 + ⎜ ⎟ 2α ⎟ ⎬ a1 β ⎠⎝ ⎪⎝ ⎠ ⎪ ⎩ ⎭ β m2 ⎧⎛ f ⋅ (2α ) m ⎞ ⎛S ⎞ ⎫ ⎪ ⎪ + 0 ⋅ G 2 ⎨⎜1 + 2 ⎟.
sample estimate of fS(S) and of the average number of stress cycles per unit time. D 100 101 Wide band Fatigue Damage Assessment General D RFC = D NB κ RFC (B. (B. An often used approach is based on the assumption that the true damage DRFC (i. Time domain simulation and cyclecounting procedures will accordingly be relevant.22) For marine risers.e. In case the process may be assumed to be composed of two independent Gaussian stress response processes an upper bound on the estimated fatigue damage can be obtained by adding the variances of the contributions directly. Accurate analytical solutions to fatigue damage estimates can be obtained for wellseparated bimodal stress spectra (e. Sensitivity studies should therefore be conducted to document that adequate fatigue damage estimates have been obtained.23) where ε is the bandwidth parameter defined by (note that ε=1 for a broad banded process and ε=0 for a narrow banded process): 102 Wide band fatigue assessment is of special importance for fatigue assessment of combined WF/LF stress response. the stress response is normally neither narrowbanded nor completely widebanded. FFTsimulation) ε = 1− λ2 2 λ0 λ4 (B. Wirshing and Light. using a rain flow counting technique) can be established from a corrected narrowband result: 101 The fatigue design is based on use of SN curves obtained from fatigues testing. detail and environment under consideration. Barltrop & Adams proposed the following expression: κ RFC (m) = a + (1 − a )(1 − ε) b where a = 0.033m b = 1.g. Dirlik. Barltrop & Adams proposed an empirical closed form expression for the stress probability density function. Extension to more general SN curves (e.323 (B. 402 E.22) have been derived by analytical means assuming two independent narrowbanded Gaussian process. For a linear SN curve. bilinear) is straightforward. E 100 Fatigue Capacity SN Curves General 204 D 300 Semiempirical Solutions 301 A number of semiempirical expressions have been proposed in the literature to correct the narrow band fatigue damage calculation for the effects of a broad bandwidth. The recommend method is the Rain Flow Counting (RFC) method. Fatigue capacity data for joint classifications of relevance for risers are given in Table B2.g. Special DET NORSKE VERITAS .DNVOSF201 Dynamic Risers. Specials purpose counting algorithms have been developed with techniques applicable to nonGaussian stress time histories. 203 The response process due to combined wave. each with a corresponding SN curve. In a wideband response a strict relationship between the stress cycles and stress maxima and minima do not exist. or ⎯ simplified analytical solutions where DNB is the narrow banded Gaussian fatigue damage given by C 200 and κRFC is a correction factor. Reference is given to Jiao & Moan (1990). The joint classifications apply to typical joints/details for risers subjected to cyclic bending moment and tension.587m − 2.g. ⎯ semiempirical solutions. large ‘m’). This is of special importance for combined WF/LF stress time histories or in cases with SNcurves with large (inverse) slope (i. For further details. see e. The following procedures exist to describe fatigue damage for a wide band process: ⎯ cycle counting algorithms. Sec 2. see e. 202 The RFC method provides an estimate of the stress probability density function (i. For practical fatigue design.24) 302 As a promising alternative. For this reason the distribution of stress cycles can not be evaluated accurately from the distribution of stress maxima.g. Cycle counting methods represent time domain estimates of fatigue damage.g.e. D 400 401 Analytical Solutions for Bimodal Spectra D 200 201 Cycle counting The fatigue damage may be obtained by counting the stress cycles in the actual or simulated stress time histories. Statistical uncertainties will therefore always be present in the estimates. reference is made to DNV RPC203. The zerocrossing frequency may be expressed as a combination of the respective zerocrossing frequencies based on expressions for the sum of two independent Gaussian processes.).and low frequency excitation is generally widebanded. October 2010 Appendix B Page 71 D. It is in general applicable to results from time domain analyses but can also be applied in connection with frequency domain analyses through a transformation of frequency domain results to time domain (by e.926 − 0.8) can subsequently be applied for estimation of fatigue damage in each stationary shortterm condition. 102 If fatigue data does not exist for the material. SN curves should either be developed by testing.e. use of fracture mechanics assessment or by use of lower bound SN curves.3 and Appendix 1. welded joints are divided into several classes. a process with a combination of low frequency and wave frequency Gaussian component). where a correction function on a form similar to (B.
3e exp − (D / t 3 ) −0. Assessment of the representative eccentricity shall be based on detailed information regarding production tolerances and installation/welding procedures supported by rational conservative assumptions as appropriate for the actual design. October 2010 Page 72 Appendix B care shall be taken for chemical environments not covered by DNV RPC203. The SCF may be calculated by detailed FE analyses or by closed form expressions for the actual structural detail.5 t3 SCF = 1 + ( ) (B. The following closed form expression applies for welded riser joints /1/: where e is the representative eccentricity due to geometrical imperfections to be applied in fatigue design.DNVOSF201 Dynamic Risers. DET NORSKE VERITAS .25) 103 A stress concentration factor (SCF) applies to account for possible stress magnification due to imperfect geometry of two adjacent joints (e.g. due to fabrication tolerances and installation procedures).
0 F1 0.g.0 NOTE 1 The thickness penalty applies only for thickness greater than 25 mm. (E.00 0. (E.1) Double side e ≤ min(0.00 0.00 SCF Single side e ≤ min(0.151) 1. Third Edition.1t 3 . 2 mm) e ≤ min(0.00 1. 3 mm) e ≤ min(0. 4 mm) D Eq. T.15t 3 . No.00 1.0 Steel bolts and threaded joints in tension F1 (coldrolled) W3 (cut threads) 0. 4 mm) D 0. whichever is the smaller. NOTE 2 For girth weld eccentricities greater than 0. G. 4 mm) F 0. Vol. No benefit can be taken for sections thinner than 25 mm. the reference thickness is the stress diameter.1t 3 .0 Single side on backing e ≤ min(0. engineering critical assessment. For bolts. F.1t 3 .00 0. DET NORSKE VERITAS .4 mm) 1. October 2010 Appendix B Page 73 Table B2 SN curves for Risers Description Welding Geometry and hot spot Tolerance requirement2) SN curve. 3 mm) e > min(0.5.0 FEManalysis 1.15t 3 . Jiao. 2 mm) e > min(0.1t 3 .15t 3 . Butterworth & Heinemann.0 F3 0. “Probabilistic Analysis of Fatigue due to Gaussian Load Processes”. according to DNV RPC203 F1 Thickness exponent k 0.1) Seamless pipe Machined components Automatic longitudinal seam welded pipes B1 B1 B2 0. References Barltrop & Adams “Dynamics of Fixed Offshore Structures”.DNVOSF201 Dynamic Risers.151) Eq. e.00 1.1t 3 .401) 1. special considerations apply. Probabilistic Engineering Mechanics. 1990. & Moan.2.15t 3 or 4 mm. 5.0 Single side e ≤ min(0.
The relative importance of these nonlinearities is strongly system and excitation dependent. The ‘associated’ return periods must be assessed with basis in environmental statistics for the actual location. 36 hours).DNVOSF201 Dynamic Risers. wave and current yielding the same return period for the combined environmental condition are typically applied. Tp etc) and a given duration (e. section 6. For a more detailed discussion. Different combinations of wind. DET NORSKE VERITAS . reference is made to e. The main focus is on consistent implementation of the LRFD design format. Reference is given to Appendix A for further discussion of analysis strategies. A 100 Objective General A B C Wind [year] 100 Associated Associated 101 The objective of this Appendix is to provide an introduction to practical implementation of LRFD and WSD design checks for combined loading based on the generalised load effect formulation introduced in Section 3. 102 Two fundamentally different methods can be applied for assessment of the characteristic load effects: Based on environmental statistics Based on response statistics The purpose of this document is to give an outline of these strategies with emphasis on the computational efforts involved in practical applications as well as inherent limitations. The offset accounting for LF motions is additional to the mean floater offset governed by mean environmental loading. Each design condition is hence described in terms of a limited number of environmental parameters (e.2. large rotations in 3D space and possible material nonlinearities as well as seafloor contact. Implementation of the WSD design format B 200 Design based on environmental statistics 201 It has traditionally been common practice to adopt the extreme response found by exposing the system to multiple stationary design environmental conditions as the characteristic extreme response. References Examples of typical 100years design environmental conditions Waves [year] Associated 100 Associated Current [year] Associated Associated 100 A. Longterm load effect statistics G. Design principles General Design based on environmental statistics Design based on response statistics General Generalised load effect Shortterm acceptance criteria Long term acceptance criteria ULS Analysis Procedure Post processing procedures Computer implementation General Implementation in design analyses General Envelope statistics Extreme response estimation Statistical uncertainty and simulation planning General Response surface approach Standards. Examples of relevant combinations to obtain 100years design conditions are given in Table C1. Shortterm extreme load effect estimation F. geometric stiffness. Hs. 202 Wind loading is indirectly included in the global riser system analysis as an important contributor to mean floater position and low frequent floater (LF) motions. Guidelines and Handbooks Technical references 101 Riser systems in general are highly nonlinear structures due to nonlinearities introduced by hydrodynamic loading. For guidance regarding calculation of representative floater offset.g.2 in API RP 2SK. General A 100 B 100 B 200 B 300 C 100 C 200 C 300 C 400 C 500 C 600 C 700 D 100 D 200 E 100 E 200 E 300 E 400 F 100 F 200 G 100 G 200 Objective B. October 2010 Page 74 Appendix C APPENDIX C ASSESSMENT OF EXTREME LOAD EFFECT FOR COMBINED LOADING Contents A. Waves and wave frequent (WF) floater motions are included as dynamic loading in the global riser system analysis while LF floater motions normally are included as a representative static offset. B 100 General Design principles B.g.g. Implementation of the LRFD design format D. Time domain finite element analysis therefore constitutes the primary method of analysis for slender structures. Table C1 Case E. reference is made to Appendix A. C.
However. waves and current consistent with the environmental conditions at the actual site should be applied for permanent installations. 90% confidence interval) should be considered. denoted xD . Alternatively. Main focus is placed on implementation of design equations for ULS conditions as this is the most general approach.e. Rk is a vector of crosssectional capacities and Λ is a vector of safety factors (i. The importance of this formulation is that the combined time dependent action of bending moment and tension is transformed into a scalar process expressed by the generalised load effect. Hs. This will in most cases imply that waves. Furthermore. design based on response statistics is in general the recommended procedure and should be considered whenever practical possible for consistent assessment of characteristic load effects (especially for verification purposes when shortcomings in the traditional approach based on environmental statistics have been identified). 302 The main challenge related to this approach is to establish the longterm load effect distribution due to the nonlinear dynamic behaviour experienced for most riser systems. 205 A wave period variation shall in addition be performed to identify the most unfavourable loading condition. Ted denote design values for bending moment and effective tension. ∆p. DET NORSKE VERITAS . This is of particular importance for ULS conditions which normally are associated with the most pronounced nonlinear response characteristics.2 ) B 300 Design based on response statistics Where g(t) is the generalised load effect (or utilisation function) at a specific location on the riser and Md . Analyses are performed for assumed critical directions (typically ‘near. Relevant simplifications in case of SLS and ALS conditions are briefly discussed. current and floater offset are (conservatively) assumed to act in the same direction. far and cross’ conditions) and final characteristic response is identified as the most unfavourable from the analyses. Acceptable results can however be expected for quasistatic systems with moderate nonlinearities. Tp) as described by environmental contours can be applied for more consistent identification of critical conditions.g. i. The load effect with a return period of Dyears. For this purpose it is convenient to consider generalised loading expressed by the following generic equation: g( t ) = g( M d ( t ). material. At least 3 different periods covering a realistic variation range (e. This section gives an introduction to consistent implementation of LRFD capacity checks for combined loading considering global time domain analysis. C 100 Implementation of the LRFD design format General 206 The main problem related to design criteria based on environmental statistics is that the return period for the characteristic load effect is unknown due to the nonlinear dynamic behaviour of most riser systems. This will in general lead to an inconsistent safety level for different design concepts and failure modes. Guidance to possible computational strategies is further outlined in F. simultaneous variation of waveheight and period (e.DNVOSF201 Dynamic Risers. 301 Consistent assessment of the Dyear load effect will in general require a probabilistic response description due to the longterm environmental action on the riser system. Ted ( t).total number of load effect maxima during D years.e. R k . Λ ) (C.and log term design approach are discussed separately in E and F respectively. The described procedures have been applied for assessment of design loads for riser systems in research activities but are yet not established as standard industry design practise.g. A verification of design criteria should be performed in the following situations: New concepts Systems with significant nonlinear response characteristics Dynamically sensitive systems The verification should be based on longterm extreme load effect assessment as discussed in section F for critical conditions.1 ) . FX (x) . Acceptance criteria are established for design based on environmental statistics (short term approach) as well design based on response statistics (long term approach). C. safety class and condition factors). Statistical techniques for extreme load effect estimation for a short. respectively. C 200 Generalised load effect 201 Consistent treatment moment/tension correlation is a key issue for efficient capacity checks for combined loading.longterm peak distribution of the (generalised) load effect C 300 Shortterm acceptance criteria 301 The code checks for combined loading in a stationary design condition is hence reduced to extreme value prediction of the generalised load effect. can formally be found from the longterm load effect distribution as: FX ( x D ) = 1 − where : ND 1 ND (C. ∆p denote the local differential pressure. 204 The most severe directional combination of wind. The former is denoted design wave approach while the latter is denoted a design storm approach. October 2010 Appendix C Page 75 203 Either regular or irregular wave loading is considered in the global response analyses.
effective weight and nominal floater position).environmental loading. Expected extremes of nonGaussian time histories are in practical applications normally estimated from a parametric probabilistic model (e. 305 Conservative estimates always could be obtained by separate estimation of design values for effective tension and resulting bending moment disregarding correlation effects which formally may be expressed as: max g M max . dynamic components from environmental loading as well as static components due to functional and environmental loading. This approach may yield acceptable result when the design is driven by one dominating dynamic component (typically bending moment for top tensioned risers with well functioning heave compensation system). i.e. allow for maximum utilisation). Techniques for establishment of the longterm load effect distribution are discussed separately in F.e. The only additional effort needed from the analyst is hence separate storage and treatment of the static response due to functional loading. This analysis is restarted from 1) considering additional loading due to steady current and mean floater offset due to environmental actions.DNVOSF201 Dynamic Risers. this analysis sequence is convenient for application of static and dynamic loading and is used in the vast majority of design analyses. This is because the bending moment components and effective tension normally are nonGaussian response processes and because the limit state function defines a nonlinear transformation of these time series.6 ) M = M . peaks of g(t)). max [ ] (C.3 ) C 500 ULS Analysis Procedure Where g max is a representative extreme value of g (t ) .M F 302 The standard framework for response processing of results from time domain analyses can therefore be directly applied for code checks. The distinction between static and dynamic environmental loading is always a key issue that must be evaluated carefully in view of the actual concept (e.4 ) r M ( t) = M y ( t). DET NORSKE VERITAS . The acceptance criterion can hence be expressed as: xD ≤ 1. see below 501 Separation of global response into components due to functional and environmental loading is an additional key issue for ULS analyses. This analysis is restarted from 2) considering additional relevant dynamic environmental loading on the system (e. C 400 Long term acceptance criteria Consistent extreme load effect estimate for combined loading can be found as a percentile in the longterm distribution of the generalised load effect. Λ ≤ 1 d ( ) (C. 3) Dynamic time domain analysis . Weibull) fitted to the simulated realisation of individual response peaks (i. This will typically include application of response envelopes in case of regular wave analysis and statistical extreme value prediction in case of irregular wave analysis 303 Statistical estimation of the expected extreme value (or most probable extreme value) for a given duration (e. ∆p.e.g. Reference is made to E for a further discussion [ yF zF ] TeF 2) Static analysis .g. Separation of these quantities into components requires that the static configuration due to functional loading is determined separately.e. The analysis is typically started from an initial stressfree configuration with incremental application of functional loading to reach the final solution.g. loading due to wave action and floater motions. static vs. 36 hours) is hence required in case of irregular analyses. dynamic current and LF floater motions). R k . October 2010 Page 76 Appendix C g max ≤ 1 (C.7 ) Te ( t ) 503 The referred global response quantities are assumed to contain the total response. The purpose of the 1st step in the analysis sequence is to establish the static equilibrium configuration due to functional loading (i. The static force output is two axial bending moments and effective tension due to functional loading: r (C. The maximum value of g (t ) applies in a design wave approach (excluding the startup transient) . possible slug flow etc) The force output is simultaneous time histories of two axial bending moment and effective tension: 304 This approach will automatically account for the correlation between effective tension and bending moment components and is hence capable of optimal design (i.g. (C.environmental loading. which require due consideration of analysis strategy as well as response post processing. The following analysis sequence can be applied: 1) Static analysis . Ted . This is in accordance with the storage and output conventions applied in the majority of tailor made computer codes for slender structure analysis In fact. while statistical extreme value prediction is required in a design storm approach. It should however be noted that g(t) always will be a nonGaussian response process.5 ) Where xD is the percentile in the long term (generalised) load effect distribution corresponding to a return period of Dyears. 502 The basic force output from global time domain analyses are simultaneous time series of bending moments and effective tension. M z ( t) where indices indicate extreme values. These response quantities contain contributions due to functional as well as environmental loading.functional loading.
∆p.of .9 ) 2 = γ FMyF + γ EMy E(t) + (γFMzF + γ EMzE(t)) 2 ( ) D. The resulting bending moment is computed as: M ( t) = M 2 ( t) + M 2 ( t ) x y (D.Guidance .1 0. local differential pressure and usage factor. Establish time history of the generalised load effect Implementation of the WSD design format General g( t ) = g Md ( t). η) (C. D 200 Implementation in design analyses 201 The generalised load effect for the WSD design C 700 Computer implementation format for combined loading can be expressed as: 701 The key to efficient LRFD capacity checks for combined loading is a computer implementation of the procedures described in the previous sections. The main technical features needed in to perform capacity checks in a stationary design condition can be summarised as: Separation of global load effects into E. the extreme load effect can be estimated as the expected.10) 101 Practical implementation of the WSD design format for combined loading is simpler when compared to the LRFD ULS design checks because no separation of the load effect into F and E components is required.77 1.8 ) M E ( t ) = M( t ) − M F 702 A computer program with the described functionality is capable of performing all relevant capacity checks for combined loading automatically with a minimum of input from the analyst.E.F. Te . Ted ( t). Implementation of the WSD design format is hence similar to the LRFD SLS and ALS design checks as discussed in section C 602.1 1.11) Where M.12) 202 Evaluation of acceptance criteria based on the generalised load effect is identical as outlined in sections C 300 and C 400 for the LRFD approach E. effective tension. October 2010 Appendix C Page 77 γf 1. Establish response components due to environmental loading: r r r (C. E 100 Shortterm extreme load effect estimation General 101 For a design storm approach. Application examples are presented by Sødahl et al (2000) TeE ( t ) = Te ( t) − TeF 2.3 .91 C 600 Post processing procedures γe 1.end .3 0. ∆p. The generalised load effect can hence be computed directly from the effective tension and bending moment components given as output from the global analyses. 602 SLS and ALS LRFD capacity checks can be based directly on time series for resulting moment and effective tension given as output from the global analyses.loads Efficient communication with FE global analysis program Graphical presentation of results as a function of location along the riser. R k . Establish design values r r Md(t) = γFMF + γ EME(t) (C. D 100 Ted(t) = γFT + γETeE(t) eF 3.and Fcomponents Generate time series of the generalised load effect Processing results from regular/irregular dynamic analysis Analyse several E. A brief introduction to implementation of WSD design checks is however given in the following for completeness.note  601 The post processing to compute the generalised load effect based on output from the ULS analysis procedure described in the previous section can be summarised in the following steps: 1. Consistent treatment of correlation requires that steps 2) and 3) in the post processing procedure discussed in the previous section is considered. ∆p and η denote bending moment. Guidance note: The following combinations of partial coefficients need to be checked for LRFD ULS conditions g( t ) = g( M( t).DNVOSF201 Dynamic Risers.F.or most probable largest response peak for the specified duration of the design DET NORSKE VERITAS . respectively. Λ ( ) (C.safety factor combinations Evaluate utilisation by nonGaussian extreme value statistics Evaluate statistical confidence in extremes Evaluate contribution from P. Te ( t ). Rk .
This “brut force” approach will yield unbiased extreme response estimates at any location along the structure. In most practical applications. σT = σE NR (C. The average envelope will hence represent expected extreme load effect along the structure.DNVOSF201 Dynamic Risers.g. A problem often encountered in practical applications is that fitted parametric model fails to describe the ‘true’ upper tail behaviour resulting in biased extreme value prediction. excitation level as well as location along riser. which is of vital importance for estimation of extreme values. Bury (1975). by use of engineering judgement or formal statistical hypothesis test) Compute estimate of characteristic extreme load effect based on the fitted model (i. For prediction of characteristic extreme response it is hence required that the simulation time must be identical to the duration of the design condition (e. the choice of probabilistic model is (at least to some extent) empirical. Special attention must be placed on description of the upper tail of the distribution. The probabilistic distribution of the load effect process is in general is nonGaussian. but is in most situations too time consuming to be of practical use.. reference is made to statistical textbooks. Thus.g. Furthermore.g.g.g. A significant variation of the nonGaussian response characteristics must in addition be foreseen along the riser. regression. The statistical uncertainty can be expressed in terms of the standard deviation of the estimated expected extreme value. For a more detailed discussion. The process of obtaining extreme load effect estimates from time domain analyses will typically involve the following issues : Envelope statistics. regular wave loading). and for verification of more sophisticated statistical methods for prediction of extreme response based on one realisation. 102 Output from irregular time domain analyses using the analysis/post processing procedure as described in section B are time traces describing one realisation of the (generalised) load effect. Gumbel extreme value distribution. and ALS design conditions. σ T : 303 The selected parametric model is fitted to the simulated peak sample using an appropriate statistical estimation technique (e. method of moments.g. which is the wanted output from shortterm global response analysis. Rayleigh. This approach applies to SLS. Special estimation techniques (tail fitting techniques) have been designed to improve the fit in the upper tail region at the expense of a somewhat increased statistical uncertainty (Sødahl and Larsen 1992). ULS. The choice of a proper distribution model will hence depend on the riser system. Normal. Furthermore. E 200 Envelope statistics 302 The major challenge is often related to selection of an adequate probabilistic distribution model for the individual peaks of the load effect process. 3 – 6 hours). extrapolation is often involved in practical estimation of characteristic extreme load effect. percentile in fitted peak distribution or expected extreme peak value) Quantify statistical uncertainty of the estimated characteristic extreme load effect.13) where NR is the number of realisations and σE is the standard deviation of the extreme response estimated from all realisations. Estimation of extreme values from nonGaussian load effect time series Estimation of simulation length required to obtain extreme load effect estimates with sufficient statistical confidence E 300 Extreme response estimation 301 The main steps involved in statistical processing of stochastic load effect time histories to produce characteristic extreme load effect can be summarised as: Select probabilistic distribution model (i.e. e. based on previous experience and physical knowledge of the dynamic behaviour of the actual riser system of concern (see also Appendix A for a discussion of governing nonlinearities). This concept is very useful to establish design values in case of deterministic loading (e. 202 A more careful interpretation is however needed for application of the envelope concept in for application in a design storm approach. see Gumbel DET NORSKE VERITAS . Exponential) are frequently applied. parametric probabilistic model for individual response peaks or extreme peak for a given duration) Estimate parameters in the selected model based on the available response time history realisation Accept/reject the selected model (e. Simple parametric models (e. October 2010 Page 78 Appendix C condition. It can however be applied for simple systems 304 Mathematical arguments in terms of limiting asymptotic distributions can in addition be applied to establish models for extreme peaks within a specified time window (e. 203 Improved statistical confidence can be achieved by considering the average envelope found by averaging over several realisations. 201 Response envelope is defined as extreme response values (minimum and/or maximum) attained during the time domain simulation as a function of location along the structure. etc). The choice of distribution model is complicated by the fact that the nonGaussian response characteristic in general is strongly system and excitation dependent. maximum likelihood. probability weighted moments. the duration of the simulated time record will in many situations be shorter than the specified duration of the design condition due to practical limitations related calculation time on the computer.g. Envelopes from irregular time domain analyses will represent realisations of the extreme load effect for the duration considered in the time domain simulation. the extremes predicted in this way will have low statistical confidence as they only represent the extreme load effect found for one realisation.e. Weibull.
The sampling distribution can consequently be completely described by the mean value and variance of the estimator.g.g.e. bootstrap estimation) assuming a sample of independent stochastic variables is normally used to establish the variance.15) An important consequence of this equation is that an increase of the simulation length with a factor of 4 is required to reduce the standard deviation with a factor of 2. generalised Pareto distribution.g. The independence assumption is normally an acceptable approximation for the peak sample. The longterm environmental loading process can be divided into time intervals with stationary conditions. a probabilistic description of the response from the longterm environmental action. The sampling distribution will also be the basis for selecting the most efficient estimator among several possible candidates.14) 102 The main challenge related to this approach is to where c is a (unknown) constant. asymptotic expressions) or numerical simulation techniques (e. σi t ≥ t T σc T c S i S 2 (C. If target confidence is not obtained (i. see Davison and Smith 1995).16) where : FX(x) M w(M) FXM(xm) Longterm distribution of load effect peaks Vector of parameters describing shortterm environmental conditions Weight function accounting for variation in load effect mean level crossing frequency Short term distribution of load effect peaks for a stationary environmental condition (i. commonly denoted the sampling distribution. 404 The following procedure can be applied for practical planning of simulations to obtain a target confidence specified in terms of the standard deviation σ Tc : Perform time domain analysis with initial duration t iS Estimate extreme response and associated standard deviation of the estimate σ Ti based on the initial duration t iS establish the shortterm load effect distribution FXM as nonlinear irregular time domain analysis in general will be required to give an adequate description of the response process. Different realisations will consequently give different estimates of the extreme load effect.g. DET NORSKE VERITAS . wave and current action on the coupled floater/slender structure system i. Approximate techniques (e. F 100 Longterm load effect statistics General 101 The longterm load effect distribution is a result of the combined wind. estimators that can be expressed as a function of sample moments) the following relation between simulation length tS and standard deviation of the estimator σT can be established by asymptotic approximations: fM(m) σT = c tS (C. These methods have in common that simplifications are introduced in the longterm load effect description to enable practical computations.e. The estimation variability can be expressed in terms of the probabilistic distribution of the applied estimator. FX ( x ) = ∫ w ( M ) FX M ( x  m) f M (m )d m M (C. denoted shortterm conditions. It is further assumed that each shortterm condition can be completely described by a limited number of environmental parameters (Waves will for example typically be described by significant wave height. F. The sampling distribution can hence be applied to express the confidence of the estimated characteristic extreme response as a function of simulation length for each particular estimator of concern. Cramer (1971) for moment based estimators). This assumption is justified by theoretical results showing that the sampling distribution of most estimators of practical interest will approach the Gaussian distribution asymptotically as a function of sample size (see e.g. peak period.e.DNVOSF201 Dynamic Risers. Simplifications are typically based on rational conservative assumptions regarding system behaviour with respect to e. spreading.e. This information can be applied directly in practical planning of computer simulations to estimate the simulation length needed to give estimates of characteristic extreme response with a specified confidence (Sødahl and Larsen 1992). 103 Discrete approximations to this general formulation form the basis for approximate techniques for assessment of the longterm load effect distribution in practical applications. σ Ti > σ Tc ) the c increased simulation length t S needed to fulfil the confidence requirement can be estimated as: E 400 Statistical uncertainty and simulation planning 401 A fundamental problem related to estimation of characteristic extreme load effect is that statistical uncertainties are introduced because estimates are based on simulated time series realisations of finite lengths. October 2010 Appendix C Page 79 1958) and for peak excesses over high thresholds (e. The longterm response distribution can hence formally be expressed as: 402 The exact sampling distribution is in general very difficult to establish for finite samples and is in practical calculations normally approximated by the Gaussian distribution. mean direction etc). conditional on M) Distribution of environmental parameters 403 For moment based estimators (i.
ISOPE.g.g. For practical application. Possible simplifications and conservative assumptions introduced to ease the design process of deepwater risers must therefore always be evaluated very carefully for each riser concept of concern.e. Tp Number of discrete sea states in the wave scatter diagram Weight factors accounting for variation in level crossing frequency Sea state probability Longterm distribution of load effect maxima Short term distribution of load effect maxima Standards. 5 or less). These seastates will serve as ‘interpolation points’ and should hence be selected very carefully Perform global response analysis for the basic seastates considering irregular time domain analysis..of . 3. floater position. Sødahl N. 1958. Floater offset and current are expected to be governing for the global response in lower parts of tensioned risers. 4. Kirkemo F. Columbia University Press . Establish probabilistic models for the shortterm load effect distributions for all basic seastates.13). Tp ) i (C.DNVOSF201 Dynamic Risers. DET NORSKE VERITAS . Sødahl N. Weibull distribution) to the simulated sample of load effect peaks is a typical procedure. Floater type. October 2010 Page 80 Appendix C environmental directionality.end . 1975. G 100 References Where: N w(Hs. 202 The response surface approach can be formulated in the following steps: 1. 5. The computational efforts involved in such techniques are establishment FXM by global load effect analyses in a limited number of carefully selected stationary environmental conditions considering irregular wave excitation. 203 The response surface will hence enable computation of the longterm load effect distribution considering a possible nonGaussian shortterm load effect characteristics. Tp ) i FX ( x  H s . pp 393442. operation of the system etc. Wave induced floater motions will normally be of some importance all along compliant riser configurations. Leira B.17) i =1 N G. Larsen C M and Olufsen A (1992): ” Extreme Response Estimation of Flexible Risers by Use of Long Term Statistics” Proc. Farnes K. riser configuration and boundary conditions will determine how the external loading is transformed into deformations and internal reaction forces in the riser.T p)i P(Hs. John Wiley. wave/current correlation. Farnes & Moan (1993) for a flexible riser application example and Sødahl et al (2000) for a SCR application example. Guidelines and Handbooks API RP 2SK “ Recommended Practice for Design and Analysis of Station Keeping Systems for Floating structures” Second Edition. s FX ( x ) = ∑ w ( Hs . ISOPE 1992. the commonly applied discrete formulation for environmental statistics described in terms of a HsT p wave scatter diagram can be expressed as: 2.1993. 52. December 1997. G 200 Technical references Bury K V (1975) “ Statistical Models in Applied Science”. Davison A C and Smith R L (1990): “ Models for Exceedances Over High Thresholds” Journal of the Royal Statistical Society Series B. (1993): ” Extreme Response of a Flexible Riser System using a Complete LongTerm Approach”. Igland R (2000): “Design and Analysis of Metallic Risers. Gumbel E J (1958) “ Statistics of Extremes”. station keeping system.g. Global analysis Procedures” OMAE 2000 . Larsen C M (1992): “Methods for Estimation of Extreme Response of Flexible Risers” Proc. Establish longterm response distribution by use of the discrete approximation to the general formulation defined in F 101. Waves and floater motions will always be crucial for the response in the upper part of the riser.Proc. Establish shortterm distributions for all relevant seastates by interpolation/extrapolation techniques using results obtained by analysis of the basic seastates as interpolation points. and Moan T.. combined environmental conditions). A. Interpolation/extrapolation techniques can then subsequently be applied to establish FXM for all relevant environmental conditions required for assessment of the longterm load effect distribution. ISOPE 1992. Fitting a parametric model (e. (C.T p)i FX(x) s FX (x  Hs .note  F 200 Response surface approach 201 The response surface approach can be thought of as a direct numerical approximation to the general formulation as defined by Eq. As an example. current and floater motions to the response of deepwater riser systems is strongly system specific. Cramer H (1971):” Mathematical Methods of Statistics” Princeton University Press.Guidance . Tp ) i P( H s . The situation is more complex in lower parts of deepwater riser systems. it is however crucial that acceptable precision can be obtained by use of relatively few basic seastates (e. A significant variation in response characteristics along the riser must also be anticipated. See e. 1971. Guidance note: The relative importance of waves. Mørk K. Select a limited number of basic ‘representative’ seastates (i.
numerical approximations of the theoretical models are needed to facilitate computer solution. 203 Furthermore. Verification of numerical procedures C 200 C 300 C 400 Spatial discretisation Frequency discretisation Time discretisation 102 However. Simplified modelling in terms of adjustments of hydrodynamic coefficients must be evaluated by more advanced techniques considering transfer functions for wave kinematics consistent with the floater motions.and dynamic finite element analysis. D. detailed design. A 200 Introduction 201 The computer model of a riser system represents two fundamentally different types of approximations to the physical system: Theoretical models. References A. Decoupled floater motion analysis should be supported by coupled floater motion analysis when significant coupling effects are identified. Section 4.g. Such studies should at least be carried out for new deepwater concepts. Verification of theoretical models 101 Global analyses should in general be performed with welldocumented and verified computer codes for analysis of slender structures.g. Furthermore. quasistatic model for bend stiffener response). The numerical approximation will typically involve spatial discretisation of the structure into a finite number of elements as well as time. load models (e. 202 The theoretical models represent the fundamental assumptions in terms of idealised models for the physical system.g. wave spectrum. global crosssectional models. Examples of typical situations are given in the following: Dynamic analyses should be considered to verify quasistatic assumptions. Many riser concepts are sensitive to wave loading in the splash zone. B. Numerical approximations. Verification of theoretical models C.). Airy wave kinematics etc). Examples of theoretical idealisations are environmental models (e. October 2010 Appendix D Page 81 APPENDIX D Contents A.DNVOSF201 Dynamic Risers. early design. Morison equation. Frequency domain analyses should be validated against time domain analyses. Floater/slender structure coupling effects should always be assessed by coupled analysis and/or model tests for deepwater mooring systems. Rayleigh structural damping. Effects from simplified treatment of LF floater motions in terms of an additional offset should be evaluated for deep water concepts.and/or frequency discretisation of the dynamic loading. This is in particular important for systems that may be subjected to resonance dynamics. The effect of disturbed kinematics due to the presence of the floater should be carefully evaluated. it is crucial to have a basic physical understanding of the applicability and limitations in commonly used theoretical models. 204 Hence. General A 100 A 200 Objective Introduction VERIFICATION OF GLOBAL ANALYSIS MODEL B. Regular wave analyses should always be verified by irregular analyses. A 100 Objective General 103 Any use of simplified analysis strategies will in general require benchmark validation by comparison to more advanced analysis procedures. solution strategy etc. Any structural modelling simplifications to gain computational efficiency should be validated against a 101 The purpose of this Appendix is to give an introduction to principles for verification of the computer model applied in global static.g.g. the key issue involved in verification of the computer model is to ensure that the theoretical models and numerical approximations represent the real physical behaviour of the riser system.g. This is of particular importance for a critical assessment of modelling and analysis of new concepts and to ensure that adequate results are obtained when simplified modelling and analysis strategies are applied. accumulated experience expressed in terms of recommended practice for modelling and analysis should always be consulted. soil model etc) and models for structural behaviour (e. LF variation of effective tension) should be addressed. Statistical correlation as well as effects from LF response on WF response (e. and final verification) DET NORSKE VERITAS . This is of particular importance for turret moored ships at deep water locations. feasibility studies. As discussed in Appendix A C 200 the required accuracy is closely linked to the purpose of the analyses (e. ref. Verification of combined use of global and local quasistatic response models by comparison to a complete response model (e. Linearized time domain analyses should be validated by nonlinear time domain analyses.
DNVOSF201 Dynamic Risers. Furthermore. The latter formulation will however require a very careful modelling of the exposed outer area for complex riser systems with variable outer diameter (e. omission of bending stiffness. Areas with high load intensities (e. 106 Independent analyses of selected critical conditions are in addition highly recommended as a part of the design process of riser systems. components. hog and sag bend). simplified modelling of boundary conditions etc). The primary purpose of this check is to verify mass and buoyancy modelling. Simple equilibrium iteration is however required in obtaining the static configuration (e. Two important examples are discussed below: Input of floater motion transfer functions in terms of amplitude and phase angle (or alternatively on complex form) as function of wave frequency and direction related to a local floater coordinate system. buoyancy modelling (pipe. 107 Sensitivity studies are also recommended to investigate the influence from uncertain system parameters (e.g. This check represents a verification of the mass (pipe. internal fluid etc). Examples of simple analytical checks are given in the following: Verification of static effective tension distribution of top tensioned risers.g. hull supports). Investigation of convergence in the solution by repeated analyses considering successive refinement of the discretisation is the basic principle to verify that the discretisation is adequate. The discretisation is considered adequate when the change in response between two successive discretisation is acceptable seen in relation to the purpose of the analyses. Special attention should be given to the following parts of the riser system: Areas with high curvature (e. The catenary configuration solution will in most situations represent a close approximation because the effect of bending stiffness to the overall static configuration normally is negligible. motion of a point on the floater at some distance from origin of vessel coordinate system) generated in global riser analysis should in particular be verified by analytical calculations for different wave directions and floater directions. Terminations to fixed structures. 104 Analytical verification should be performed whenever possible to verify modelling and input parameters. splash zone). wave direction and floater coordinate system differ from program to program. soil data etc). taperjoint. waves and riser deflections is a very useful tool for verification of floater motions. Buoyancy can be treated in terms of effective tension as discussed in Appendix A or alternatively by integration 101 Numerical approximations will typically involve spatial discretisation of the structure into a finite number of elements as well as time. Approximate solutions are given in terms of closed form expressions for tensioned beams and cables with uniform crosssectional properties. calibration against model test). C 200 Spatial discretisation 201 Repeated static and dynamic analyses considering successive refinement of the element mesh can be applied to assess the adequacy of the spatial discretisation. use of average of crosssectional properties. support rational conservative assumptions and identify areas where a more thorough investigation is needed to achieve an acceptable modelling (e. but it will also give a verification of the shape of the static configuration.g. Animation showing floater motions. In this situation. The independent analyses should in principle always be carried out using a different recognised computer program. Verification of numerical procedures 105 It has been experienced that surprisingly many modelling mistakes can be traced back to a few common problem areas. Conversion between different definitions is usually required to apply output from hydrodynamic floater motion analysis (e. Spar risers. systems with attached buoyancy elements etc).g. The static configuration of single line compliant riser configurations can be verified by use of catenary equations disregarding the effect of bending stiffness. using the socalled ‘shooting’ approach).g. October 2010 Page 82 Appendix D more comprehensive structural model (e. Both formulations are correct and will hence give the same riser response when applied correctly.e. Areas with significant change in crosssectional properties (e.g.g.g. simplified modelling of components. of the hydrostatic pressure acting on the outer riser surface. Eigenmodes of top tensioned risers can be verified by analytical calculations. phase angle. equivalent multipipe model. there will be no practical gain by further refinement of the discretisation. hydrodynamic coefficients in moonpool.and/or frequency discretisation of the dynamic loading. The effective tension distribution of top tensioned risers can be found by accumulation of effective weight along the riser. It is therefore recommended that use of computer programs based on pressure integration for representation of hydrostatic pressure should be validated against other codes using the effective tension formulation. Such operations should be performed very carefully with emphasis on thorough verification. Floater terminal point motion (i. Definition of amplitude. bend stiffener etc).g. Contact areas (touch down. diffraction/radiation approach) as input in global riser analyses. The main purpose should be to quantify model uncertainties. calibration and enhancement of computer analysis of riser systems. C. it is crucial to utilise information from model tests as well as fullscale measurements whenever possible for validation. additional buoyancy components etc) and tensioner modelling of the system. DET NORSKE VERITAS .
β and HilberHughes. The relative change in length between adjacent elements with uniform crosssectional properties should not exceed 1:2. (relevant for e. Possible cancellation frequencies should be identified and covered by the discrete representation. slower convergence is normally observed for shear forces and bending moments when compared to effective tension). heave.α methods are frequently applied. momentcurvature hysteresis). beam elements based on conventional displacement formulation may display a significantly different numerical behaviour when compared to hybrid elements used in a mixed formulation. while the phase angle is assumed to follow a uniform probabilistic distribution over the interval (02π). e. It should also be clarified how the actual computer program handles possible excitation outside the frequency range of the floater motion transfer function (this is a well known source to erroneous excitation). This is because the rate of convergence normally will be different for different response quantities (e. contact problems and significant nonlinear material behaviour (e. Discretisation of wave direction with a spacing in the range of 1530 deg. The discrete frequencies and directions must be selected carefully to obtain an adequate description of the floater motions: The frequencies should be selected to cover the resonance peaks in vessel motion transfer functions (e. some aspects are discussed in the following: The time step required to obtain a stable numerical solution is to a large extent governed by the highest eigenmode present in the discrete structural model.e. 402 Time domain analyses considering stochastic wave loading will typically require generation of discrete time histories for floater motions and wave kinematics according to a specified wave spectrum. C 300 Frequency discretisation 301 Floater motion transfer functions are represented in terms of amplitude and phase angle as function of a number of discrete wave frequencies and directions.DNVOSF201 Dynamic Risers.g. Choice of time step is crucial for the stability and accuracy of direct time integration methods. Quality checks of response time histories should always be considered to identify possible unphysical noise reflecting an inaccurate numerical solution.g. instability problems. systems with significant displacement dependant nonlinearities such as low tension problems including snap loading. This is because all eigenmodes need to be integrated accurately to obtain a stable solution (i. This DET NORSKE VERITAS . Use of variable time stepping procedures should at least be validated against constant time step algorithms when unphysical noise is detected in response time series. Furthermore. Important aspects regarding load discretisation is discussed in the following: C 400 Time discretisation 401 Numerical time integration is applied in time domain analyses to produce discrete response timeseries. The overview statistics discussed in Appendix A is a very useful tool for detection of possible unphysical response peaks. roll and pitch resonance frequencies). 302 Results from frequency domain analysis are given in terms of auto. noise introduced by change of time step. October 2010 Appendix D Page 83 Areas with change in element lengths. Unconditional stable. is normally sufficient to give a good representation of the floater motions. This is in particular the case for numerical sensitive systems. Study of convergence considering successive refinements of the time discretisation is a useful exercise to determine the required time step to obtain an adequate numerical solution.g. The adequacy of the frequency discretisation can be assessed by repeated analysis considering successive denser frequency spacing. It is therefore recommended to apply constant time step algorithms when analysing numerically sensitive systems.4s for numerically wellbehaved systems. The frequency spacing will hence be decisive for the variance and covariance found by integration of the corresponding response spectra. The amplitude of each harmonic component is normally computed from the specified spectral representation of the load process. This is because e. Variable time step integration methods may introduce high frequency noise when applied to numerically sensitive systems. also modes that are of no significance for the response description) Typical time step is in the range of 0.1.g. 203 The convergence study must be performed for the actual element used in the analyses. Identified suspicious locations along the riser should be subjected to closer examination by spectral and statistical analyses as well as visual inspection of the response time histories. semisubmersibles and TLP’s).0. The main advantage of this approach is that almost no additional cost is related to use of many frequencies to describe the load processes. 202 The convergence should be assessed for all relevant response quantities. static as well as dynamic analyses should be considered in the evaluation studies. The latter approach is normally preferred in variable time step algorithms due to explicit control of numerical damping to suppress possible high frequency 403 The generation of load time histories can be carried out very efficiently by use of the FFT (Fast Fourier Transform) technique using equidistant frequency representation of the load process. Nonlinear analyses will in general require a shorter time step to obtain a stable numerical solution when compared to linearized analyses.g. The load time histories are represented in terms of a finite number of harmonic components. A lower relative change may be required in case of nonuniform crosssectional properties.g. single step integration procedures such as Newmark.and crossspectral densities at a number of discrete frequencies. The frequency range should cover relevant frequencies in the wave excitation.
J F.g. Several strategies have been proposed. Benchmark validation by successive increase of number of interpolation points is recommended to verify the spatial interpolation. 106. A time step in the range of 0. McNamara. Watters A J (1995) “ Frequency Content Selection for Dynamic Analysis of Marine System” OMAE 1995. 405 An additional practical problem related to use of variable frequency spacing is that it is more complicated to assess the repetition period of the generated time histories. resonance peaks in the vessel motion transfer function and peak period in wave spectrum). Furthermore. Interpolation in time and space is hence necessary during the simulation. The main advantage is that wave kinematics can be calculated at instantaneous spatial position allowing for consistent representation of wave kinematics in case of large riser displacements (e. Benchmark validation by successive increase of number of frequencies is recommended. These frequencies must hence be selected very carefully to give an adequate representation of the loading (e. This approach is however far more time consuming than the FFT approach and will only be applicable when relatively few frequencies are considered for representation of the load processes (typically 100200).g. use of variable frequency spacing is required to cover the relevant D. The spatial interpolation should in particular be considered carefully to obtain an adequate representation of the loading close to sea surface.251s is typically sufficient to facilitate adequate time interpolation of WF excitation. Vol. 404 Direct accumulation of harmonic components representing floater motions and wave kinematics can alternatively be performed during the simulation to overcome the interpolation problem related to the FFT approach. References Garrett D L. 36 hours). Garrett et al (1995) and McNamara and Lane (1984).e.DNVOSF201 Dynamic Risers. Lane M (1984) “Practical Modelling of Articulated Risers and Loading Columns” journal of Energy Resources Technology. The repetition period of the generated load time history is also uniquely determined by the frequency spacing of the harmonic components (see Appendix A) The main drawback is however that time series must be generated prior to the simulation at fixed locations along the riser.g. The numerical behaviour of the random generator may depend on the actual computer used in the analyses. Variable spacing of interpolation points (i. Gu G Z. 406 The quality of the generated floater motions and wave kinematics depends on the ability of random number generator to produce statistically independent phase angles. points where load time histories are pregenerated) along the riser is normally considered to obtain efficient analyses. October 2010 Page 84 Appendix D is of particular importance to describe the relevant frequency content of vessel motion transfer function and wave spectrum as well as the response process in case of resonance dynamics. Quality checks of generated wave realisations are recommended in connection with new computer installations to ensure that the generated realisations are Gaussian. frequency range with as few harmonic components as possible. DET NORSKE VERITAS . see e.g. Approximate closed form expressions are available for some algorithms. combined LF and WF floater motions). Judgements based on the autocorrelation function estimated from the generated realisation can alternatively be applied to assess the repetition period (Garrett et al 1995). Statistical properties of the process and individual peaks should be considered for several realisations with rather long duration (e.
(E. Multimodal response analysis based on empirical hydrodynamic coefficients (and tests). φ(s)) to be calculated as: 101 A simplified estimate of the induced fatigue damage can be computed by neglecting the influence of the waves.DNVOSF201 Dynamic Risers.25) c) For each mode. the crossflow vibration will generally be perpendicular to the current direction. Blevins (1990).Guidance .note  B.note  Where U is the local tangential flow velocity and D is the outer riser diameter. and γ is the mode participation factor. A 100 Objective General 101 This Appendix proposes a fourstep method for assessment of VortexInduced riser response amplitudes and corresponding fatigue damage. B 100 Fatigue Assessment Simplified Assessment of Fatigue Damage where E is the modulus of elasticity and SCF is a stress concentration factor . If the simplified analysis indicates insufficient fatigue capacity. Fatigue Assessment Simplified Assessment of Fatigue Damage Multimodal Response Analysis Based on Empirical Models Methods Based on Solution of the NavierStokes equations Modification of Riser Properties Vortex suppression devices C. refined assessment methods preferably supplemented with tests are required.3) B.end .of . Typically. General A 100 B 100 B 200 B 300 Objective VIV ANALYSIS GUIDANCE following procedure can subsequently be applied (B 102. The DET NORSKE VERITAS . 32 = γD 0.14 to 0.g. . f s = St U D (E.of . Guidance note: Often. These steps of increasing complexity is defined as follows: Simplified assessment of fatigue damage. b) Define a band of local vortex shedding frequencies fs along the riser using: D.Guidance .. Computational Fluid Dynamics solving the NavierStokes equations. October 2010 Appendix E Page 85 APPENDIX E Contents A. assuming undisturbed current velocities to apply. see e. κ is the curvature of the mode shape φ(s) at the point (s. Laboratory test.B 106).2) 105 Compute corresponding stress range: S = A SCF⋅ E ⋅ κ⋅ (Dt) (E.e. Guidance note: For rotationally symmetric riser systems. the main design focus is to evaluate if the fatigue capacity is sufficient. St is the Strouhal number where upper and lower bound values should be checked (Typically St = 0. this will be the mode with the highest frequency among the “probable modes” 104 For a given flow velocity compute the vibration amplitude for the anticipated mode according to Sarpkaya (1979): A 0. a simplified (i. d) Identify the most likely mode shapes to be excited by VIV and select the one with the highest curvature for a unit modal amplitude. References A. Methods for reduction of VIV C 200 C 300 103 Identify dominant mode shapes and natural frequencies as follows: a) Determine the natural frequencies and mode shapes for bending in the crossflow direction based on analytical models or by numerical FEM analysis. conservative) VIV analysis will suffice if the resulting fatigue damage is within the tolerated limit. Accordingly.06 + (2π ⋅ S 2 ⋅ K S ) t Where Ks is the stability parameter. For nonsymmetric systems. the crossflow vibration is assumed to occur in the plane of the relevant modeshapes . The method should be chosen according to the specific case investigated. check for which parts of the riser the natural frequency for the mode is within the limits of the local shedding frequency. 102 Identify the planes of vibration for the relevant mode shapes in relation to the specified current directions.1) 102 The fundamental principle is that for cases where VortexInduced Vibrations (VIV) are likely to represent design problem. more sophisticated should be applied.end .
i. according to Vandiver (1993). The resulting stress range is subsequently obtained by multiplication with 2 A SCF. Blevins (1990) states that a reduced damping greater than 64 reduces the peak amplitudes to less than 1 % of the diameter. the presence of shear flow in the region of the higher modes greatly reduces the probability for lockin.Guidance . Introduce vortex suppression devices.end . the reduced damping is usually lower than one and it is very seldom possible to increase the damping to such an extent. tension.e. and a response spectrum can be retrieved. C 200 Modification of Riser Properties 201 There are several different ways of reducing the amplitude of vortex induced vibration. structural damping. Subsequently. S is stress range and m and a are constants defining the SNcurve. for marine risers the wake will be turbulent. Larsen and Halse (1995) conducted a comparison between programs showing considerable discrepancies concluding that at present no generally accepted program exist for calculation of VIV response. see Appendix B. 1996). October 2010 Page 86 Appendix E 2 κ( x) = ∂ φ 1 + ∂φ ∂s 2 ∂s 2 −3 / 2 (E. Calculate response in the time domain.e. It is usually possible to avoid the resonant crossflow region when the highest reduced velocity is below 3. requiring very small timesteps or a good turbulence model. B 200 Multimodal Response Analysis Based on Empirical Models 201 If significant VIV induced fatigue damage is likely. 106 The fatigue damage is estimated by application of the relevant SNcurve as: DF = f n ⋅ TL a ⋅S m (E.note  Guidance note: If a finite element model is applied. below the resonant region. One way of achieving this is by application of a generalisation of the procedure described above. Validation of the numerical results by sensitivity studies with respect to key parameters should accordingly be performed. In marine applications. A correlation function for the loading process at two points along the riser is introduced. Zdravkovich (1981) classifies the DET NORSKE VERITAS . a double integration is performed. The next level of refinement is typically methods for multimodal response analysis based on empirical or semiempirical values of the hydrodynamic coefficients. There are also two other main approaches for calculating the response: Calculate modal response in the frequency domain. also including modelling of the dynamic boundary conditions.. . Proper force coefficients for vibrations including more than one frequency are still lacking. Otherwise. There will always be a higher natural mode with a frequency that corresponds to fs .g. C.Guidance . diameter. the computations generally become increasingly complex and time consuming. Comparison with results obtained from fullscale or model experiments is also essential for calibration and finetuning of the numerical algorithms. Even if this approach at present stage is very time consuming and possibly not correctly modelled for high Re. The excitation is directly dependent on the response.of . This approach can incorporate general current profiles. more thorough calculations should be conducted. see e. i.of . The parameters entering the calculation of load and response generally requires calibration with model field data. To be well above the resonant area is much more complicated. frequencies and phase angles for various combination of incident velocity and crosssection vibration. the stresses corresponding to unit mode shape amplitude is first computed based on the stiffness matrix for the relevant element.5) where fn is the frequency of the relevant mode. In the end such a simulation would hopefully stabilise or maybe repeat.4) Jumps from one mode to another may happen.DNVOSF201 Dynamic Risers.note  B 300 Methods Based on Solution of the NavierStokes equations 301 The analysis based on solution of the full NavierStokes equations implies a set of twodimensional fluidflow analysis for sufficiently many crosssections along the riser. Methods for reduction of VIV 101 If the calculated VIVresponse is a problem. it is likely that this will be a feasible approach in the future. there are two main approaches: Modify the properties of the riser. However. The direct solution of the complete flow equation is until now restricted to low Reynolds numbers (no turbulence in the near wake). Guidance note: For vertical risers in wellknown environmental conditions recognised semiempirical programs may be applied. T1 is the design life of the riser. 302 For compound multipipe riser geometries. 202 A different approach is to increase the reduced damping. For screening purposes a 1year velocity with associated velocity profile is considered conservative. Here one must have a considerable database of crosssection tests giving force coefficients. SHEAR7 (Vandiver & Li. C 300 Vortex suppression devices 301 A second possibility is to add vortex suppression devices to the cylinder. .end . a weighted summation of computed damage over the longterm current distribution for velocities and direction must be performed. However.
DNVOSF201 Dynamic Risers, October 2010 Appendix E Page 87 means of suppression to three categories according to the way it influence the vortex shedding: surface protrusions (wires, helical strakes etc.) triggering separation; perforated shrouds, axial slats etc. (breaking the flow into many small vortices); and near wake stabilisers, preventing the building of the vortex street. In Blevins (1990), eight different devices are shown, and comments on their use and effects are given. Common for all (except the ribboned cable) is that they increase the cost of the riser, and that they will complicate handling during installation. Some of the devices also reduce the drag coefficient, especially the streamlined fairing. However, in most cases the inline drag coefficient is increased rather than being reduced by introducing vortex suppression devices.
D.
References
Blevins, R. D. (1990). FlowInduced Vibration (Second ed.). New York, USA: Van Nostrand Reinhold. Sarpkaya, T. (1979). Vortexinduced oscillations, a selective review. J. of Applied Mechanics 46, 241258. Larsen, C.M. and Halse, K.H.(1995):”Comparison of models for vortex induced vibrations of slender marine structures. In Proceedings of the Sixth International Conference on FlowInduced Vibration, London UK, pp. 467482 Vandiver, J. K. (1993). Dimensionless parameters important to the prediction of vortexinduced vibration of long, flexible cylinders in ocean currents. J. of Fluids and Structures 7, 423455. Vandiver & Li (1996). User Guide for SHEAR7 Version 2.0. MIT, September 1996. Zdravkovich, M. M. (1981). Review and classification of various aerodynamic and hydrodynamic means for suppressing vortex shedding. J. of Wind Engineering and Industrial Aerodynamics 7, 145189.
DET NORSKE VERITAS
DNVOSF201 Dynamic Risers, October 2010 Page 88 Appendix F
APPENDIX F
Contents A. General
A 100 A 200 B 100 B 200 B 300 B 400 B 500 B 600 B 700 B 800 Objective Application
FRAMEWORK FOR BASIS OF DESIGN
miscellaneous.
B 200
General design requirements
B. Design basis
General General design requirements Internal fluid data Environmental data Data for Floater and Stationkeeping System Riser system and interfaces Analysis methods and load cases Miscellaneous
201 The operator should specify project specific design
requirements, e.g.: riser location; general requirments; description of the riser system including extent, main interfaces, configuration, boundary conditions, main dimensions and main components; choice of applicable design codes, standards and regulations; nominal and minimum internal diameter of equipment bores interfacing with the riser; length of each component type; number off, for each component type; required service life; testing ; fire protection ; material selection, coating, corrosion protection and corrosion allowances.
A.
A 100 Objective
General
This Appendix defines the items normally to be included in the design basis document.
A 200
Application
Design basis shall be prepared for all risers.
B.
B 100 General
Design basis
B 300
Internal fluid data
301 The operator should specify all relevant internal
fluid parameters. As relevant, the parameters listed in Table F1 should be specified. For uncertain data, the parameters should be specified as realistic ranges (min/normal/max). Expected variations in the internal fluid parameters over the service life should be specified.
101 A design basis document shall be created in the
initial stages of the design process to document the basis criteria and analysis methodology to be applied in the structural design of the riser system.
102 When the design has been finalised, a summary
document containing all relevant data from the design and fabrication phase shall be produced, i.e. a Design, Fabrication and Installation (DFI) résumé.
302 If temperature and pressure is correlated, extreme
combinations of temperature and pressure may be provided in the form of a design envelope diagram.
303 If rapid decompression of internal gas may occur,
the corresponding adiabatic temperature drop inside should be calculated by the supplier, and reflected in the minimum design temperature.
103 This section presents the essential of the
information that must be available to the designer, in order to be able to design the riser according to this standard. This information is normally included in a design basis document.
104 Typical information needed to perform a riser
design includes as a minimum: general riser system design requirements; functional requirements of the riser system; operational requirements of the riser system; internal fluid data; environmental data; floater data; interface requirements and equipment/component data; structural analysis methodology including load cases to be considered; verification procedures;
DET NORSKE VERITAS
DNVOSF201 Dynamic Risers, October 2010 Appendix F Page 89 Table F1
Parameter Internal pressure
Internal fluid parameters
Comment The following internal pressures should be specified: maximum internal pressure including operating, design and incidental pressure with possible pressure profile through service life ; mill and system test pressure requirements ; minimum internal pressure (including vacuum condition if applicable). The following temperature should be specified: operating temperature or temperature profile through service life ; design maximum temperature ; design minimum temperature ; Including produced fluids, injected fluids, exported fluids, and continual and occasional chemical treatments (dosages, exposure times, concentrations and frequency) ; all parameters which define service conditions, including partial pressure of H2S (sour) and C02 (sweet) ; fluid density range corresponding to relevant pressure and temperature ; fluid/flow description including fluid type and flow regime. ; sand or particle erosion data ;. Sweet or sour in accordance with fluid composition. Fluid type and flow regime including slugs. Annulus fluids for multipipe systems Flow rates, fluid density, viscosity. Fluid heat capacity.
Table F2
Parameters Location Water depth Seawater data
Environmental parameters
Comment Geographical data for planned fields of operation. Design water depth (minimum and maximum), tidal variations, storm surge and subsidence. Density, pH value, and minimum and maximum temperatures. Minimum and maximum during storage, transportation, installation and operation. Description, shear strength or angle of internal friction, friction coefficients, seabed scour and sand waves (soil/well and/or soil/pipe structure interaction characteristics). To be used for analysis/design riser base foundation, soil restraint for conductors and soil/structure interaction evaluation for touch down region for catenary risers. Maximum values and variations along length of thickness, density and surface roughness. Current velocity as a function of water depth, direction and return period, and including any known effects of local current phenomena. In terms of significant and maximum wave heights, associated periods, wave spectra, wave spreading functions and wave scatter diagrams as function of direction and return period. Wind velocity as function of direction, height above water level and return period. Maximum ice accumulation, or drifting icebergs or ice floes. Ground motions described by means of spectra or time series.
Air temperature Soil data
Temperature
Fluid composition
Marine growth Current data
Wave data
Wind data Ice Earthquake data
Service definition Fluid/flow description Flow rate parameters Thermal parameters
B 500 Data for Floater and Stationkeeping System 501 The operator shall specify all data for the floater
and stationkeeping system of relevance for design and analysis of the riser system.
B 400
Environmental data
401 The
operator should specify all relevant environmental parameters. As relevant, the parameters listed in Table F2 should be considered. Combined wind, wave and current conditions should be specified for relevant return periods (e.g. 1, 10 and 100 year return periods).
502 The following general floater data should be
included as relevant for the actual installation:
Main hull dimensions; Detailed hull geometry, draughts, mass, radii of gyration etc required to required to perform hydrodynamic motion/excitation analysis of the floater; Detailed moonpool geometry, if relevant; Location of riser supports and riser supporting structures/devices (e.g. tensioner, moonpool supports etc)
402 For temporary (retrievable) risers, the operator
should specify the required range of environmental conditions (weather window) and planned field locations for which the riser should be suitable.
403 For environmental conditions at the limits of the
weather window, it should either be possible to safely retrieve the riser, or it should sustain being hangoff throughout a design storm specified by the operator.
Specification of possible interference areas, including other risers, mooring lines, platform columns, floater pontoons, keel, surface equipment and deck, surface jumper and deck, etc. and definition of allowable interference/clashing if any.
503 Floater motion characteristics should normally be
specified in the design basis. The following information is required for documentation of the floater motion characteristics:
DET NORSKE VERITAS
mechanical connectors. seafloor conditions including characteristic soil properties (e. October 2010 Page 90 Appendix F WF floater motion transfer functions in 6 degrees of freedom with a clear cut definition of amplitudes and phase angles as well as wave directions. as relevant: DET NORSKE VERITAS . B 700 Analysis methods and load cases 602 An overall layout of the riser system should be provided together with a clear definition of scope of design.and current coefficients for the floater.g. midwater arch and distributed buoyancy modules. pulling capacity. motion reference point) and directions of coordinate axes. Description of buoyancy modules such as aircans. etc. amplitudes and phase angles etc) to allow for implementation of these data in the actual software for stationkeeping analysis. The actual water depth at the location and together with the slender structure restoring force for the actual mooring/riser system design shall be applied in calculation of WF floater motion transfer functions. etc) 603 A general description of the top interface between riser system and adjacent structure should include information. stiffness. stress joints etc). jumpers. stroke. the maximum permissible deflection angle should be defined for the relevant tension and pressure ranges. Detailed description of the tethers/mooring system. The floater attached coordinate system used as reference for floater motion transfer functions shall be documented in terms of origin (i. subsea equipment like BOP. suspended line lengths. attachments etc. . friction coefficients etc). The following should be included: 702 Design criteria for all relevant temporary phase conditions including.e.DNVOSF201 Dynamic Risers. description of possible clump weights/buoys. draughts). LWRP. such as: Floater support boundary conditions.g. mooring line breakage shall be specified.). Description of additional external lines . tension and bending moment) of the wellhead equipment and the top suspension. if relevant Mean position and second order motions for relevant design conditions including intact as well as damaged conditions due to e. material quality. Wind.e. 601 The customer should provide the required information on any interfaces between riser pipe and adjacent structures. etc. design of temporary and permanent riser top suspension systems (spiders. specification of which parameters/components of the riser system that are subject to design (typical examples are wall thickness. subsea template dimensions and stiffness . pressure. EDP. All applicable limit states for all relevant temporary and operational design conditions shall be considered.g. The following additional information is required to conduct coupled and /or decoupled stationkeeping analyses WF and LF transfer functions for hydrodynamic excitation on the floater. load/displacement characteristics (linear/nonlinear) and failure tolerance of tensioner systems. LMRP. flex joints. The following information may be included as relevant for the actual installation: wellhead datum relative to sea level.g. connectors. 606 For temporary top tensioned risers. 504 The design basis document may include relevant data for evaluation of the global performance of the installation.g. i. stress joints. Indications of preferred solutions should be given to the extent possible. etc. the maximum allowable disconnect angle of the emergency disconnect package (EDP) should be defined by the operator for input to the operating condition limits for the riser analysis. reference coordinate system. B 600 Riser system and interfaces 607 For risers equipped with flexjoints. surface equipment like surface flow tree. Riser joints including cross section data. geometry. location of anchors and floater attachment points etc) DP system characteristics in case of DP assisted mooring systems Detailed description of the riser system A clear cut definition of must be provided for transfer functions and coefficients (e. tension joint. position tolerances and capability curves). Frequency dependent added mass and damping for the floater. Arrangement of risers. riser joint length. DP system performance (e. in case of more that one riser . 604 A general description of the bottom interface and subsea equipment should be included in the design basis document. Riser configurations. material data.g. equipment and component data. The floater motion transfer function shall be given for relevant loading conditions (i.g. For slack/semitaut/taut mooring systems this will typically include layout pattern of the mooring lines and detailed mooring line composition (e. directions. Description of structural components of relevance for the actual installation (e. buoyancy modules. umbilical etc. annulus content. 605 The operator should provide information on the permissible loading (e. subsea tree. ball joints emergency disconnect package. if any. to which the riser is connected.e. Examples of information that may be included in the design basis document are: 701 The intended procedures to be adopted in the design of the risers shall be documented. conductor stiffness and soil restraint.
tensioner failure. functional and environmental load criteria and design load combinations (cases) . retrieval. internal pressure and related internal fluid density. installation. vessel offset. transportation. weak links (if relevant). SNcurves.) . including description of : global analysis model(s) including modelling for wave and current loading and floater motions.g. relevant SLS criteria for the riser pipe and structural components B 800 Miscellaneous 801 A general description of other essential design information. top tension. Procedures/scope for verification of the riser design (e. connection and disconnection . 703 Design criteria for all relevant operational phase conditions including. dropped objects etc. local analysis model(s). as relevant for the actual installation: limiting pressure. criteria for limit state checking. e. relevant ALS criteria . essential design parameters and analytical procedures associated with temporary phases e. 705 A general description of the structural evaluation process. including: description of procedures to be utilised for considering global and local responses. 704 A general description of analysis models to be utilised. lifting/handling. DET NORSKE VERITAS . maintenance and repair/replacement.g. riser abandonment. collision. description of procedures to be utilised for combining global and local responses. basis for stress concentration factors (SCF’s). drive/drift off. load cases to be analysed.g. fire. testing and independent review/analyses of the design).DNVOSF201 Dynamic Risers. explosion. etc. description of procedures to be utilised for code checking. functional and environmental load criteria and design load combinations (cases) . relevant ALS criteria. October 2010 Appendix F Page 91 limiting pressure. including: inservice inspection criteria general philosophy for inspection.g. description of fatigue evaluation procedures (including use of design fatigue factors. essential design parameters and procedures associated with operational phases e.