Principles of Earthquake Engineering of Bridges Part 1

Principles of Earthquake Engineering of Bridges
Part 1: Principles & Approach
by Dr. Mark A. Ketchum, OPAC Consulting Engineers

for the EERI 100th Anniversary Earthquake Conference, April 17, 2006
Principles of Earthquake Engineering of Bridges Part 1
Presentation Outline
z Introduction
z Performance Criteria
z Fundamental Principles
z Ground Motions
z Structural Design
z Demand Analysis
z Capacity Analysis
z Detailing
z Advanced Topics

Introduction
z Part 1: Principles & Approach
z Part 2: Structural Analysis
z Types of Structures
z AASHTO Methodology
z California Methodology

Types of Structures
z Ordinary Bridges
z Special Bridges

Special Bridges

Ordinary Bridges

Typical California Bridge

AASHTO Methodology
z Force-based Conformance Checking
– Response Modification Factors
z Importance Classification
– I: Essential Bridges
– II: Other Bridges
z Seismic Performance Category (SPC)
– Importance Classification
– Peak Ground Acceleration
– Categories A, B, C, D

AASHTO Design Requirements
z Ground Motion
– Minimum 0.2g Static Lateral Load
– Seismic Response Coefficient
• Hazard, Soil, and Frequency dependent
– Site-specific spectra & motions allowed
z Analysis
– Static Lateral Load
– Single & Multi Mode Spectral Analysis
– Time History Analysis
z Force D/C with Response Modification Factors

California Methodology
z Displacement Ductility Approach
z Pre-determined Damage Locations
z Hazard & Soil Dependent Motions
z Category (influences performance requirements)
– Ordinary
– Important (Post-EQ Service Required)
z Classification (influences analysis / design requirements)
– Geometry (multilevel / curved / skew)
– Framing (stiffness / strength distribution)
– Geotechnical (near fault / soft soil / liquefaction)
California Performance Criteria
Safety Event Functional Event
Ground Motions 1000 - 2000 year Higher Probability

Return Period 150 - 500 year
“Maximum Credible” Return Period
Ordinary Bridge Limited Service Immediate Service
Significant Damage Repairable Damage
Important Bridge Immediate Service Immediate Service

Repairable Damage Minimal Damage

Fundamental Principles
z Strength
– Analysis under Design Ground Motions
– Capacity Controlled Components
z Redundancy
– Stiffness Balance with a Frame
– Frequency Balance between Frames
z Ductility
– Capacity > Demand
– Minimum Ductility regardless of Demand

Design Ground Motion
z Site Specific Assessment for Important Bridges
– Probabilistic Seismic Hazard Assessment
– Deterministic Seismic Hazard Assessment
– Rock Motion Spectra
– Rock Motion Histories
– Site Response Analysis
– Soil / Foundation / Structure Interaction Analysis
z Simplified Procedures for Ordinary Bridges

Design Ground Motion
z Simplified Procedures for Ordinary Bridges
– EQ Magnitude from Seismic Map
– PGA from Seismic Map
– Select Representative Standard Soil Profile
– Response Spectra for EQ / PGA / Soil Profile

California Seismic Map

California Seismic Map

Soil Profile Classifications

Response Spectra

Response Spectra

Response Spectra

Structural Design
z Proportioning
– Minimum Column Dimensions & Lateral Loads
– Balanced Stiffnesses and Frequencies
– Redundancy
z Demand Controlled Elements
– Columns / Bents
– Foundations
z Capacity Protected Elements
– Bent Caps / Girders
– Pile Caps / Footings

Structural Design

Seismic Demand Analysis

Seismic Demand Analysis
z 3D Model of Bridge
– Dynamic
– Elastic, Cracked (usually)
– Inelastic (special cases)
z Response Spectra Analysis
– Time History Analysis in special cases
z Directional Combination of Ground Motions
– Horizontal Motions
– Vertical Motions
z Displacement & Force Demands
Prototype Design

3D Elastic Dynamic Model

Displacement Demands

Displacement Demands

Force Demands
z Demand Controlled Elements
– Columns / Bents
– Plastic Capacities
z Capacity Protected Elements
– Column Shear / Bent Caps / Girders / Footings
– Overstrength Capacity of Controlling Elements

Seismic Capacity Analysis
z Inelastic (Nonlinear) Analysis
z Displacement Capacities of
– Components (e.g. columns)
– Subsystems (e.g. frames / bents)
z Curvature Capacities of Sections
– By Moment vs. Curvature Analysis
z Strain Capacities of Materials
– Steel
– Concrete

Strain Capacities of Materials
z Mild Steel
z Prestressing Steel
z Concrete
– Confined
– Unconfined

Mild Steel

Prestressing Steel

Concrete

Curvature Capacities of Sections
z Inelastic Analysis of a R/C Cross Section

Displacement Capacity
z Pushover Analysis – single-column system

z Pushover Analysis – multi-column system

z Pushover Analysis – complex system

Unified Demand & Capacity
z Global Inelastic Time History Analysis

Unified Demand & Capacity
z Global Inelastic Time History Analysis

Capacity Protected Elements
z Superstructures
z Bent Caps
z Footings
z Designed to direct inelastic damage into the

columns, pier walls, and abutments
z Strength > connecting elements

Detailing
z Column confinement
z Ductile column connections
z Joint shear
z Hinge restrainers
z Abutments

Influence on Cost
Deck Deck Bent Column Type Column
Model Structure Type Geometry Span Arrangement
Width* Depth Columns (Estimated Size) Height
1 CIP/PS box Straight 39’ 6’ 120’+150’+150’+150’+120’ 1 5.5’x8.25’ oblong 22’
2 CIP/PS box Straight 68’ 6’ 120’+150’+150’+150’+120’ 3 5.5’ circular 22’
3 CIP/PS box Straight 39’ 4’ 80’+100’+100’+100’+80’ 1 4’x6’ oblong 22’
4 CIP/PS box Straight 68’ 4’ 80’+100’+100’+100’+80’ 3 4’ circular 22’
5 PC/PS girder Straight 39’ 5’-2” 80’+100’+100’+100’+80’ 1 4’x6’ oblong 22’
6 PC/PS girder Straight 68’ 5’-2” 80’+100’+100’+100’+80’ 3 4’ circular 22’
7 PC/PS girder Straight 39’ 6’-2” 120’+120’ 1 4’x6’ oblong 22’
8 PC/PS girder Straight 68’ 6’-2” 120’+120’ 3 4’ circular 22’
9 CIP/PS box 1000’ radius 27’ 6’ 120’+150’+150’+150’+120’ 1 5.5’x8.25’ oblong 22’
10 CIP/PS box 30° skew 68’ 6’ 120’+150’+150’+150’+120’ 3 5.5’ circular 22’
11 CIP/PS box Straight 39’ 6’ 120’+150’+150’+150’+120’ 1 5.5’ circular 50’

Cost Trends
Cost/SF vs. PGA
for Magnitude 7.25 Earthquake
$180.00
$170.00
Type 1
$160.00
Type 3
$150.00 Type 4
Type 6
$140.00
Type 9
$130.00 Type 10
$120.00
Type 11
$110.00
Cost/SF ($/SF)
$100.00
$90.00
$80.00
$70.00
$60.00
$50.00
$40.00
$30.00
$20.00
$10.00
$0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00
PGA (%g)

Summary
z Documented Approach
z Supported by Theory & Testing
z Directly applicable to standard bridges
z Approach applicable to special bridges
z Performance basis
z Details

Part 2: Structural Analysis
z Robert Dameron

EERI 100th Anniversary EQ Conf. Tutorial
Principles of Earthquake
Engineering of Bridges
Part 2. Structural Analysis
PrinciplesR.
ofA.
Earthquake
Dameron,Engineering
PE, David forEvans
Bridgesand Associates, Inc., San Diego, CA
Dameron
OUTLINE
z Introduction / Historical Perspective
– Bridge Structural Analysis
– Bridge Seismic Research
z Structural Analysis Procedures

– Steps to follow
– Analytical Solution Methods
– Modeling of Components
z Example Projects
z Seismic Analysis Tutorial Example
– Martin Luther King Viaduct
z Conclusions and Lessons Learned

Principles of Earthquake Engineering for Bridges Dameron
INTRODUCTION / HISTORICAL PERSPECTIVE

• Distinction between “Design Analysis” and “Response
Prediction Analysis”
• “Prediction Analysis” was rare prior to 1989

– Bridge analysis mostly involved elastic and RSA techniques
• post-Loma Prieta Governor’s Board Report called for:

“comprehensive earthquake vulnerability analyses….of important
transportation structures….using state-of-the-art methods in
earthquake engineering.”
• 1990s: Caltrans conducted seismic research, bridge

seismic assessments, and state wide bridge retrofitting

INTRO - cont’d
• ULTIMATE GOAL: CALCULATE DEMANDS and CAPACITIES
• “DESIGN ANALYSIS”
– Global models to obtain member force and structure displacement
demands
– for routine projects, models are typically elastic
– EQ simulation is typically with response spectra
• LIMITATIONS (FOR EQ DESIGN)

– Member force demands often far exceed capacity (but are reduced with
ductile design procedures)
– Displacements (improved with cracked section assumptions) are

approximate
– All demands are “peak occurrence”; no info as to simultaneity, hysteretic

damping, low cycle fatigue
– Information limited on expansion joint behavior

INTRO - cont’d
• “PREDICTION ANALYSIS”
– Global Analysis models with material and geometric

nonlinearity
– EQ simulation with time history input (NTHA)
– Local Nonlinear Analysis models to assess capacity, failure

mode, or local displacement history

BRIDGE ANALYSIS CHALLENGES
San Diego - Coronado Bay Bridge

...different issues for new vs. old bridges
Proposed new East Spans of SFOBB

EVOLUTION OF “PREDICTION ANALYSIS”

• Essential Bridges; Importance of knowing
actual behavior
• Perf. Criteria: “Allowing Significant Damage”
– especially for retrofitting older bridges
• Need for more accurate predictions of:
– Dynamic character of response
– Displacements & low cycle fatigue
• Seismic device technology
– nonlinear dampers; iso bearings
• New, Major Bridges: Elastic Perf. Criteria

BRIDGE SEISMIC RESEARCH

• Extensive research in the 1990s
– on seismic performance of bridge components
– retrofitting measures
– devices: bearings and dampers
• Much of this was initiated/funded by Caltrans -
extensive literature/data now available from
reports by…
– UCSD, UCB, UCI, UCD, UNR, USC, and many other
universities

PHYSICAL TESTING OF CORONADO PILES

...TESTING OF CORONADO PILES, cont’d

TYPICAL HYSTERESIS TESTS
Good ductility and

energy absorption
Poor ductility and

semi-brittle failure
Budek, Benzoni,
Priestley: “Experimental
Investigation of Ductility
of In-Ground Hinges in
Solid and Hollow
Prestressed Piles”

STRUCTURAL ANALYSIS PROCEDURES
• Steps to follow
• Analytical Solution Methods
– Response Spectrum Analysis (RSA)
– Nonlinear Time History Analysis (NTHA)
• Modeling of Components

Steps to follow
(especially for
evaluating &
retrofitting
existing bridges)
by permission from Parsons Transp. Group

RSA METHODS
• Sometimes models analyzed frame by frame

• Elastic modeling of members and foundations
• Expansion Joints addressed with “Tension”
and “Compression” Models
• Foundations represented with Mass,
Stiffness, and Damping Matrices
• (shown in more detail later with an example)

NTHA METHODS - Solution Procedure

• NTHA now available in most FE programs
• Most common formulation:
– direct time integration using Newmark or other method
– Equilibrium iteration; similar to a static problem
– Choice of time step appropriate to structure and EQ input
• Need to conduct an eigenvalue study first
– for time step selection
– damping selection
– QA/QC of the model
• Superstructures can often be modeled elastically
• Linear studies are always recommended prior to
introducing nonlinearities
Unique Issues for some Bridge Tower Fdns.
• Stiffness
matrices for battered
pile groups can have negative off-
diagonal terms
• for NTHA, complex pile groups
usually require Nonlinear
Pushover Models with detailed
representation of Cross-Sections
and P-y Springs to accurately
simulate group behavior AND
predict Plastic Hinge Formation

NTHA Methods – cont’d

A Typical Challenge – Pile Modeling
PILE ELEVATION PILE SECTION
CONCRETE DECK
P/C OR R/C
M-Φ ELEMENT
HORIZONTAL “P-Y”
SPRINGS
(MODEL NONLINEAR STIFFNESS
AND RADIATION DAMPING)
VERTICAL “Q-U” SPRING

(INCLUDES PLUNGING FUSE)

NTHA cont’d - Pile (and column) Modeling

3
P-Y Curves
derived from 2
various programs
Force, "P"
1
Lateral Deflection, "Y"

NTHA cont’d
M-Ф Curves Varying with Axial Load
INC R
Curves Derived
EASIN
OR P R
from M-Φ
G AX
Moment
E STR Analysis Program

IA L L
ES S
OAD
for Different Axial
Loads
Curvature

NTHA METHODS - Damping

• Rayleigh Damping still most widely used
• Need capability to apply selectively to element groups
• Typical values (% of critical):
– Welded steel 2-3%
– Prestr. Concr. <1/2 yield 2-3%
– R/C < 1/2 yield 3-5%
– R/C with signif. cracking 5-7%
– Foundations 10%
• Need to verify that…
– damping is not double counted in element groups with
significant hysteresis
– Rayleigh damping is realistic (because of velocity reference
frame)
• Advancements in fdn. modeling, e.g. radiation damping
Nonlinear FE Analysis - Component Modeling

• Requires proven ability to predict
– nonlinear behaviors, hysteresis, energy absorption
– degradation and local “failures”
• Ultimately, most important test for selection of
program and modeling method
• Often requires specialized expertise and
– special elements and material laws not normally used
– key decisions in element layout / nodal geometry
• Difficult to find special research elements
implemented in general purpose FE programs

NTHA Project Examples:

San Diego-Coronado Bridge
• 8,000’ structure with a 90o
curve
• Steel girder superstructure;
concrete substructure
• P/S concr. piles supporting
marine footings
• Consultant team:
• Analytical studies
(Vulnerability Study)
performed by ANATECH and
OPAC, et al
• Retrofit Design performed by
T Y Lin / Earth Tech

Coronado Piles and Piers

NTHA Tower Modeling
Elasto-Plastic
Isolation Bearings
Joint
Nonlinear Elastic Superstructure
Col. Hinge
Nonlinear Contact
Dampers Springs or restrainers
F=CV1.0
Explicit Pile Elements
Nonlinear
Col. Hinge
(M-Φ behavior coupled
to axial force)
Ground Node

Modeling of Isolation Bearings

Elastic-Plastic
Material
Nonlinear-Elastic
Material
Superposition of Material Models

700
Shear
Force
(kips)
0
-700
-50 0 50
Bearing Displacement (inches)

Typical Pier - Plan View Tower Drift Trace
Flexure Retrofit
Col. Shear Retrofit

RSA Demand CapBeam Shear
Transv. Retrofit
Drift
(Inches)
NTHA
Response
Biaxial Flexure Env.
Longitudinal Drift (Inches)

Coronado Bridge Analysis - Highlights

• NTHA tailored to the bridge, the seismic hazard, the
retrofit solution, and to the performance criteria
• Key nonlinearities (typical of many large bridges):
– Joint opening / closing
– Pile hinging and SFSI
– Column hinging and P-∆ effects
– isolation bearings and dampers
• Laboratory component testing used to improve / verify
– hysteretic models of piles and columns
– failure criteria of piles and columns
– reliability of seismic devices
• Fault rupture also evaluated
– by combining static offsets with NTHA

Brooklyn-Queens Expressway
• Unique three-level (1948)

concrete cantilevers
• Pile supported retaining
walls
• Vital transportation link
• Consultant team:
– Analytical studies of
concrete structures by
ANATECH
– Project Lead:
Parsons Transp. Group

BQE Location Map

BQE “CR” Structures
PROMENADE
PROMENADE
EASTBOUND
EASTBOUND
DIAPHRAGM
DIAPHRAGM
WALL AT
WALL
2.54 AT
M O.C.
~
2.54 M O.C.
~
OPENING IN
OPENING IN
DIAPHRAGM
WESTBOUND DIAPHRAGM
WALL
WESTBOUND WALL
FURMAN ST.
FURMAN ST.
STEEL
STEEL
H-PILES
H-PILES

BQE Analysis Approach Summary

GOAL APPROACH TOOL
CAPACITIES CODE FORMULAE SPREADSHEETS

- STRENGTH
CAPACITIES DETAILED LOCAL ANACAP / ABAQUS
DISPL / DUCTILITY CYCLIC PUSHOVER
DYNAMIC BEH. ISSUES – SSI MODELS ANACAP / ABAQUS
SFSI, KI
“ “ SEMI-GLOBAL ADINA
SPATIAL VARYING MODELS
DEMANDS RSA & LTHA, SEMI- SAP-2000
MOMENT / SHEAR GLOBAL MODELS
DEMANDS NTHA, LOCAL ANACAP / ABAQUS
DISPLACEMENTS MODELS

Final Configuration for Applying Equivalent Uniform Motion

BQE – Shear Strain Contours (Damage Prediction)
Expected
Damage to
Wall
Shear Failure of
Roadway Slab

BQE - Time History Response of Three Levels

Nonlinear Analysis of Components

Example: New Tacoma Narrows Br.
• Suspension structure, parallel to
existing bridge, with hollow concrete
tower legs
• Engineering team and global NTH
analysts: Parsons Transp./HNTB JV;
• Local FE Modeling by David Evans
Associates (using ANACAP-U lic.
from ANATECH)
• Example of local FE analysis
combined with NTHA to confirm
seismic performance
FE Modeling of Tower Legs of New Tacoma Narrows

Modeling of Reinforcement

Contours of Vertical Strain - Strong Axis Push

Contours of Vertical Strain - Weak Axis Push


Martin Luther King Viaduct

Portland, OR
• Bridge Facts
– Replacement of existing 1936 structure.
– Total length of bridge, 340 meters
– 12 Spans, 3 CIP/PT Box Girder Spans and 9
PC/PT Box Beam with CIP Deck Spans
– Railroad right-of-way restrictions require
highly skewed bents.
• Performed by David Evans and Assoc. for

ODOT

• Analysis Details - Foundation Modeling
– Challenging soil conditions
– Core samples revealed up to 20M thick layer of sawmill waste
– LPILE was used to develop nonlinear lateral pile response
– Several iterations performed to determine final tangent
stiffness
Bridge 19667 - Lateral Load Parameters; Bent 4 West

Elevation γ′ φ′ C k
(meters) KSOIL (kN/m³) (deg) (kN/m²) ε50 (kN/m³)
14.38
Sand Fill 4 18.07 34 --- --- 24,430
13.01
Silty Clay Fill 4 15.71 30 --- --- 6,790
12.40
Woodwaste * 9.43 * --- --- *
7.98
Woodwaste * 1.19 * --- --- *
1.20
Silty Gravel 4 9.83 36 --- --- 16,300
-0.32
Sandy Gravel 4 11.40 38 --- --- 33,900
At depth
Table W5
* Use Custom Woodwaste PY Curves

• Analysis Details - Bridge Modeling

– Grillage model was developed using RM-2004.


– Sections were defined in geometric pre-processor using CAD
type interface with variations in girder depth, thicknesses, etc.
as a function of bridge stationing.


• Springs represent bearing pads at expansion joints and abutments

• Analysis Details – Bent Modeling

Bent 4 (Looking South)
Rigid Eccentricities
Rigid Eccentricities
Rigid Eccentricities Foundation Springs

• Analysis Details - Seismic Modeling

– Frequency Analysis
– Determine Mass Participation and Mode Shapes
Frequency Mass Participation
Mode (Hz) X Y Z
1 2.091 27.30 1.65 1.24
2 2.282 52.57 1.30 1.41
3 2.727 0.05 0.25 0.02
4 2.815 0.49 4.50 0.12
5 3.010 3.33 0.14 0.00
6 3.171 0.37 0.84 2.19
7 3.466 0.13 0.01 0.18
8 3.515 0.64 0.48 0.93
9 3.650 1.35 0.00 25.42
10 3.688 0.01 0.03 21.85
. . . . .
. . . . .
. . . . .
90 11.994 0.00 0.09 0.00
91 12.035 0.00 0.01 0.00
92 12.080 0.00 0.00 0.00
93 12.107 0.00 0.02 0.00
94 12.116 0.00 0.01 0.00
95 12.173 0.00 0.00 0.00
96 12.205 0.00 0.08 0.00
97 12.208 0.00 0.00 0.00
98 12.216 0.00 0.00 0.00
99 12.249 0.00 0.03 0.00
100 12.252 0.00 0.00 0.00
TOTAL 95.06 50.45 91.24


• Analysis Details - Seismic Modeling
– Site specific response spectra

Conclusions & Lessons Learned

• Modeling decisions: balance additional insights
gained versus increasing complexity, cost, and
uncertainties
• We always recommend a stepwise approach:
– Linear, including eigenvalues, for model checking (RSA can
generally be used in cases governed by a force-based criteria)
– Add geometric nonlinearities of contact opening/closing and P-∆
effects
– Add nonlinear seismic devices
– Add material nonlinearities where demands dictate they are
needed
• Some basic nonlinear procedures and elements

have proven reliability
Conclusions, cont’d: Limitations on Analysis
• Lack of reliable damage models

– especially for concrete: cover spalling, rebar buckling, shear
– but also for steel local buckling
– need experimental validation for acceptance
• Continued conflict between structural, geotechnical
engineers, and seismologists; too much engineering
effort applied to aspects which
– have little influence on the final structure design
– incorporate more detail than is warranted by the source info
• Codes need to resolve issue of multiple ground motions
• NTHA should continue to be augmented by linear
dynamic and nonlinear static pushover methods

Acknowledgements
We wish to thank:
• California Dept. of Transp., Washington Dept. of Transp., New York

State Dept. of Transp.: project owners and sponsors
• T Y Lin / Earth Tech (JMI); Parsons Transportation; PTG/HNTB JV:
project design engineers and project leads for the Coronado, BQE,
and Tacoma Narrows projects, respectively
• UC San Diego, Professors Seible and Priestley: Coronado pile
testing and other information
• ANATECH Corp.: nonlinear analysis for the Coronado and BQE
projects; authors of ANACAP program
• David Evans and Associates: nonlinear analysis for the New
Tacoma Narrows tower leg evaluation and MLK project
• Mr. Brian Hansen: analyst for the Coronado, BQE, and MLK projects

Principles of Earthquake Engineering of Bridges Part 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Principles of Earthquake Engineering of Bridges Part 1

Uploaded by

Copyright:

Available Formats

Principles of Earthquake Engineering of Bridges

Part 1: Principles & Approach

by Dr. Mark A. Ketchum, OPAC Consulting Engineers