This action might not be possible to undo. Are you sure you want to continue?

Prop-Imperfection Subsea Pipeline Buckling

Neil Taylor & Vinh Tran School of Construction. Sheffield City Polytechnic, Pond Street, Sheffield, UK. SI IWB (Received !6 October 1991: revised version received i July 1992;accepted 14July 1992)

ABSTRACT In-service buckling of subsea pipelines can occur due to the introduction of axial compressive forces caused by the constrained expansions set up by thermal and internal pressure actions. Proposed herein is a mathematical model relating to a pipeline, the otherwise horizontal and straight idealised lie of which is interrupted by an encounter with an isolated prop or point irregularity. The overbend produced can serve, in the presence of enhanced topologies involving trenching, burial, discrete or continuous, and fixed anchor points, to trigger vertical or upheaval buckling of the pipeline under inservice conditions. The results of a series of case studies are contrasted with data appertaining to alternative models available in the literature: experimental support is additionally noted. By questioning the implicit stress-freewhen-straight assumption present in these alternative models, it is considered that a consistent, imperfection-prone isolated prop formulation is hereby provided, suitable for design application. Key words: in-service buckling, subsea pipelines, isolated prop, trenching, burial, fixed anchor points.

NOTATION A D E Cross-sectional area Pipe d i a m e t e r Elastic m o d u l u s S h e a r force at prop

KFi

325 Marine Structures 0951-8339/93/$06.00 © 1993 Elsevier Science Publishers Ltd, England. Printed in Great Britain.

326

Neil Taylor. Vinh Tran

Fap

h, hj, h2 I ki(i = 1-6) L

Lfap

Li Lo

L~, Lsl,

Lu L*

Ls2

7l

N, Ni P P

Po

q q'

t

T T'

Us

U

O Ui Om Oom

v.',.

x CZ

Anchor shear capacity Cover depths Second moment of area of cross-section Constants Buckle length Anchorage spacing Buckle length of the isolated prop imperfection topology Buckle length of the contact undulation imperfection topology Slip lengths Buckle length at upheaval state Lower limit on buckle length re axial friction force response through slip length Bending moments vP/EI Maximum bending moments Internal pressure rise Buckle force Pre-buckling force Buckle force at quasi-idealised state Buckle force at upheaval Submerged self-weight of pipeline per unit length Submerged self-weight of pipeline cover per unit length Wall thickness of pipe Temperature rise Pressure-equivalent temperature rise Resultant longitudinal movement at buckle/slip length interface Resultant flexurally induced end-shortening Vertical displacement of the pipe Vertical displacement of the imperfection topology Maximum vertical amplitude of the buckled pipe Maximum vertical amplitude of the imperfection topology dv/dx etc. Spatial coordinate Coefficient of linear thermal expansion Trench angle Poisson's ratio Yield stress Axial friction coefficient Axial friction coefficient of overburden Lateral friction coefficient

0

O'y

0;

eL

Subsea pipeline buckling

327

INTRODUCTION The increase in demand for hydrocarbon deposits has led, during the past two decades, to the development of substantial offshore infrastructure in the North Sea. More recently, marginal offshore fields have been exploited employing unmanned satellite facilities. Hydrocarbon export frequently employs subsea pipelines which can either simply rest on the seabed or lie in excavated trenches, with or without burial. A subsea pipeline laid at ambient temperature and subsequently employed to transport high-temperature hydrocarbons under pressure is thereby subject to the introduction of axial compressive forces caused by the constrained thermal and pressure actions and buckling can ensue. '-3 With hydrocarbon transportation temperatures up to 100°C above that of the water environment and operating pressures over 10 N/mm2 these forces can be substantial given the ability of the pipeline/seabed interface to generate the necessary frictional resistance to axial movement. Pipeline installation is both sophisticated and expensive and investment is substantial. Failure of a pipeline is Costly both in terms of lost production and repair, and actual in-service buckling failures have recently been recorded in the literature. 4-6 With the later employment of smaller bore pipes for in-field hydrocarbon transportation from marginal fields employing satellite technology, the vertical or upheaval buckling mode has become of paramount importance as such pipes must be trenched and/or buried to protect them, for example, from damage by anchors and/or trawling gear -- the latter can weigh up to 100 tonnes. Trenching/burial largely obviates alternative lateral mode buckling failure. 4-7.8 Three basic types of initial imperfection can be identified as illustrated in Fig. 1. In the first case, the pipeline remains in continuous contact with some vertical undulation in an otherwise idealised horizontal and straight lie. The isolated prop alternatively features a sharp and distinct vertical irregularity such that voids (sea-filled) exist to either side. The third case occurs where the above voids become infilled with leaching sand and represents a special sub-case of the first. The initial imperfection is denoted by amplitude Yoreand wavelength Lo or Li as shown. Whilst L~ is determined from simple statics, Lo is subject to individual engineering judgement.3 Present interest is centred on the isolated prop case of Fig. l(b). The prop represents the undercrossing of a non-parallel pipe or the presence of an intervening rock; stop-start trenching procedures can also be responsible. The overbend of the pipe serves to trigger upheaval buckling wherein the pipe lifts off the prop, resisted in these attempts by the

.r ' - ± ." '" " " Prop..'')~illliiiiTiii... self-weight. " " ".. " " . Essentially.. Finh Tran Cover level if p i p e l i n e -~. together with indefinitely small deformations and linearly elastic constitutive properties. D .. : : '" " buried 2.~m h ' : " " : " " " " : " " " " " " " " ? DI L Loll ~ Lo....: -I ] Isolated Prop " 'h :' . four sets of equations are generated appertaining to: . •. The following study presumes system symmetry and seabed or trench-bottom rigidity. Typical imperfection configurations.~.2 d XTrench bottom / Seabed (a) Basic Contact Undulation l L (b) Prop Cp_. '. !.. burial overburden) on the pipe and the pipe's stiffness.e. . .328 Neil Taylor. L (C) LI/I Infilted Li/l _ Prop Fig...~. ".. .". " .. ~ ~ : " .t %.llllillllllillllilil.~:'l'. effective download (i. ? Sand Infill~ .-- -- -- •. ".

Indeed. whilst E and v are the appropriate elastic modulus and Poisson's ratio respectively.or post-buckling process. 5 . and (d) buckling relationships. As the temperature of the pipeline rises due to routine operation. regarding (a). The basic isolated prop subsea pipeline buckling model is now considered with emphasis being placed upon the respective buckling relationships. s-~llit is worth noting that most subsea pipeline buckling models largely agree regarding the composition of factors (a)-(c). the so-called prebuckling pipe force Po generated by a temperature rise T and a pressure rise p can be readily represented by ~ Po = A E a T + ~ ( 0 . Subsea conditions are assumed to preclude effective infilling of the adjacent voids with solid matter at any stage of the pre. with pressureequivalent T' applied as a back-end reduction as necessary. (c) longitudinal compatibility. With alternative isolated prop models available in the literature.Subsea pipeline buckling 329 (a) the interpretation of temperature and pressure rises over ambient in terms of axial compression so generated within the pipe. (b) longitudinal equilibrium. Merging the known action parameters T and p leads to computational convenience such that eqn (1) can be written Po = A E a ( T + T') (2) where T' = pD(0-5 . ISOLATED PROP TOPOLOGIES The proposed five key stages in buckling development are illustrated in Fig. m m units). action T alone is considered.v)/(2Eat) with T' ~ pD/(24t) for typical material values (N. The . it is within (d) that most models' idiosyncracies lie.v) (1) whereA denotes the net cross-sectional area of the pipe of outer diameter D and wall thickness t. the initial span or imperfection wavelength Li suffers a reduction as the pipeline tightens up under compressive action P (P < Po. The datum state refers to the initial lie adopted by the pipeline following laying operations whereby a vertical out-of-straightness is caused by the presence of a prop. see later). trenching and/or burial details together with the employment of fixed anchor points are treated later. 2. Here.

L~L ...p~... . Isolated prop topologies.... 2. pu c) Upheaval (L=Lu) . ...:. P(...:7. dj_ p ..~ v m d ~ P ~ Le'~ seTrbeen:lh bOttOm b) Pre-upheaval Flexure (Lu<L<Li) 0 (. .......d) Post-upheaval Buckling (Lu(L<Li) /////r/~/t.. ...u j~u l ~///////////It// e) Post-upheaval Buckling I f l l l I I I I i i I f l l f l l l f l l l I I I t l I I I I K I I E I I I I l l l l t t l t l l t l l l l f l f l f l l f -k P L Li L>Li Fig.330 a) Datum ( P = O) Neil Taylor.t//t// Li ..... 0 /.Pu -L L _ Lu(L<Li Li ~J. l/inh Tran [V°m ...

.+ ~ . equal to half the prop force.Subsea pipeline buckfing 331 wavelength L reduces to some specific value Lu (P = Pu) whereupon the pipeline lifts off the prop. DATUM ESTABLISHMENT The appropriate topology is shown in Fig./2 = vi x Io = t 0 (3) where/-)i denotes initial vertical deflection and vi x = d v ~ / d x etc.. Initial imperfection topology. With boundary conditions vi IL. Post-upheaval buckling initially involves wavelength Lu < L < L~. = . 3 with the pipeline effectively being under the contrasting actions of a prop imperfection of amplitude Uomand a submerged self-weight loading intensity ofq (to which can be added any overburden effect in the case of buried pipes m see later).computational manipulation gives q/unit length Ni__F. then equilibrium affords for general bending moment Mi Ix. ' FiLi qL2 qx2 Mi lx = Elvi . and a bending moment Ni acting at the crown together with a transverse reaction at the peel point. 0 < x < Li/2. . Reactions include a shear force F~.~ ._2 Fi = -E I d3vi dx 3 Vom Flexural Rigidity = El Peel point '1 I / f f f / f f f f f f f f f f f f f f f f f l / f l f f l l l f f l f f f l l ./2 t t ¢ = vi x IL./2 = vi xx IL.// fqLi / ~ Fi P:o O~ ~-x Li/2 Fig. 3.. with L > L~ ensuing if circumstances so dictate._. i qLi..2 (4) Noting u~10= Oom.+ F i x .~.

eqn (5) is employed as a kinematic imperfection of form. however.-. any prop buckling study which employs eqn (5) in conjunction with eqn (9) is effectively c o n d e m n e d to replicate established idealised studies. ~0Here.. imperfect moment-curvature relationship . whilst providing an initially curved datum v~(x) for ensuing stability studies. ~. Vinh Yran Li = 5"8259(V°qEI) TM (6) Fi _ E1 together with vi . Accordingly.332 Neff Taylor. PRE-UPHEAVAL FLEXURE Figure 4 illustrates the topology adopted upon the onset of in-service axial compression P which is constant through the wavelength Lu < L < Li. The foregoing argument leads to employment of the familiar. whilst eqn (5) is taken to be usefully true following field observations in the North Sea) eqn (9) is taken to relate to only a component of residual stress in the aslaid pipe.Io and -- .3EI (7) vi . ~2and that the 'isolated' inclusion of the stress data corresponding to eqn (9) provides an effectively imperfection-free formulation which would then be non-conservative these features are discussed further below n then the familiar engineering worst case scenario philosophy is invoked whereby the imperfectionnullifying idealised stress component given by eqn (9) is suppressed and a Perry-like datum assumption of stress-free-when-initially-deformed is employed/3 Hereafter.~x[~a× -- qL! 24EI (8) The foregoing equilibrium study. other components following from fabrication and laying operations/L ~2Given that any residual stresses are likely to be subject to in-service thermal stress relieving6-7. actually demands a supposedly previous hypothetical stress-free-when-straight datum with q initially relating to an empty pipe. in the absence of comprehensive and definitive as-laid residual stress data ~t~2. qLi vi ~xx[0 . q now allows for the pipeline containing hydrocarbons.

M~ _ E1 v ~ - ...trlt it r..r." h'/~ F L/2 Li/2 - I~--' I - - (a) Flexural Range Topology LulL x~Li q Initial Imperfection curve Prop Buckle under thrust P IVm=Vom reaction2F / I L <Li L _ (b) General F] =~ El2 =]: L/2 =L u. t H ~. however. with .'. rlttrt.. The respective boundary conditions take the form U [L/2 = U 'x [L/2 = V 'xx [L/2 = V'x ~ = 0 (11) together with v [0 = Uom. (10) where Mx represents the bending moment atx.Subsea pipeline buckling q/unit length 333 p N¢4F Flexural Rigidity = E l . vi xx . and v denotes the vertical pipe displacement at the deformed state (P ~ 0).trlrlll HIqqllllrHItt Illrt.H.= - LS J Topology (c) Axial Force Distribution Fig.r~/. Isolated prop .t.The presence of the bending moment at the peel point despite the zero curvature transversality requirement is to be noted.pre-upheaval: details of imperfect fully mobilised model. 0 < x < L / 2 . 4.

502 7.857 667 6.986 7.. (11) and (13) affords the characteristic equation M.0 2.Elvi'xx IIj2 = -Elvi'xx Iz.857 6. (10). with F representing half the prop force.298 091 1.713 8-039 8.221 515 1.30 1-20 1-10 1. bending moment M./2 (12) in accordance with eqn (10).5 3.0 6. Manipulation of eqns (5).259 967 1.P/E1.70 0-60 Upheaval F = 0 U p h e a v a l Vm = Vom Post-upheaval L < L~ L = Zi L = Li Post-upheaval L > L~ P -~ 80.5 4. IL/2 = EIv 'x~IL/2 .713 7.205 182 1.327 8.262 7.659 8.342 1 1.0 1.754 667 727 40 238 4 4 016 418 057 047 Remarks P~ 0 Pre-upheaval 1. is given by qx2 (13) 2 from equilibrium.0 6.342 1 1.1 (nL)21 1/4 (14) where n 2 -.212 541 1.4 4 . = e(Vom v) + N + Fx Li _ 5"8259 L nL (nL)2 cos(nL/2) + 2nL sin(nL/2) .~ cos(nL/2).232 263 1.0 5.. Vinh Tran M.334 Neil Taylor.90 0-80 0. N and F denoting the crown moment and shear force respectively. Also in conjunction with eqn (10).194 847 1.986 8 (L > Lu) .0 0.76EI/L 2 0-01 8.19931 1.0 3. Evaluating for nL in terms of L~/L (see Table 1) then vertical deflection v is given by TABLE 1 Typical Buckling Force Solution for Isolated Prop Model Li/L nL 1-5 2.

) qx2 M.P = [2~pAqAE(-us)]~/2 + ~pA(q~L2 . sin(nL/2) (16) with the crown shear force F being expressed as F E1 - ( . L~.Subsea pipelinebuckling 335 = nq-fl4Ei(-2cosn(L/2 . = 3 \ L .cos(nL/2)) (17) and general bending moment being given by Mx = P ( v o m . With the slip lenglhs Ls undergoing fully mobilised axial friction restraint ~ a q per unit length.V ) + ~ q (k3 +L~ 24 (nLi)4 1152 2 +Fx-~-. note the system topology and axial force distribution given in Fig.v % Io) - (-vi% Io) = q [2sin(nL/2) + (n~i. then familiar manipulation affords the equilibrium expression ~-3 Po . being generated in the adjacent lengths of pipe.k. 4(b) and (c).nL~i] EIn (1 . < N (18) noting F is available from eqn (17).x ) .nL )cos(nL/2) .x ) + klsinn(L/2 . Having established the buckling force P in terms of wavelength L and amplitude vm = V [0.3 nL { Li ) +q nF k2 = kl + nL k3 - (nLi)4 115~ + 2cos(nL/2) . the driving force behind the buckling mechanism.n2x 2 + k2nx + k3) (15) where k.F ) (19) where Us denotes the longitudinal movement of the peel point given by the equally familiar longitudinal compatibility expression . it is now necessary to employ longitudinal equilibrium and compatibility to relate P to the previously discussed temperature rise T = Po/AEa. Changes in wavelength are accompanied by frictional resistance to Po. . where OAis the axial friction coefficient between the pipe and the seabed.

)2dx More fully.F) + U = (24) where U is given by eqn (21)./2 (v'x)2dx = ( {q'~2 \ E I J n1 ~ (4 + k~) nL4_ + a0 (nL) 3 (k.nL] + 2k112 .1) .6 + ./2 ' 2 J0 (vix) dx = 1 / (22) (q)2 E-/ L~ 483 840 (23) Full solution for the pre-upheaval flexure stage is now available from eqns (14)-(23) although the familiar longitudinal fully mobilised friction modelling employed above fails to allow for the early phase of this stage in which all necessary frictional resistance is (theoretically) provided for by the peel point concentrated reaction 0A [qL/2 . eqn (21) is - 2 J0 1 (Li/2 (vi x) d x ' 2 (21) f/. u~ = 0 and Po = P+ (/)A(~--F) (25) The above formulation is valid for 0 < P < Pu where Pu denotes the buckle force in the pipe at the onset of upheaval from the prop. + . ..2 c o s ( n L / 2 ) . F o r L < L * . This circumstance has been accounted for elsewhere.1 (k~ .4)sinnL . with L = L* denoting the wavelength at which u~ = 0.k l ( c o s n L . 4(b) and eqn (19)). Vinh Tran (Po .4[(kl + nL)(l .F] (note Fig.P ) L U (20) u~ - 2AE in which U denotes the flexural end-shortening through the wavelength such that 1 c/2(o...cos(nL/2)) + 2 s i n ( n L / 2 ) .336 Neil Taylor.n L ) k . then L* is found from (PAL* ( ~ .(kl + n L ) s i n ( n L / 2 ) ] ] and t. it is simply necessary to indicate that eqns (19) and (20) are only valid for us < 0.14-15 Here. Prior to .

with F = 0.g.~.7. therefore. the idealised t solutions . E1 = 63%Pq~ Pu = Pit=0 = 42.e. it is pertinent to appreciate that the present analysis relates to in-service conditions. is defined as being that at which the prop reaction force (2F) reduces to zero. of crucial importance to the designer. The above are explicitly based upon the familiar moment-curvature expression given by eqn (10) which incorporates initial imperfection curvature v~'xx effects. the above lends further support to the adoption. 1) the pre-upheaval flexural regime represents an in-service capability for delaying the onset of upheaval. i. Given the above noted substantial degree of in-service movement herein concerned. eqn (9) is suppressed. it is contended that thermally induced residual stress-relieving is thereby similarly available. flexural and associated slip length movement can occur without upheaval being induced. Pu). it does share the residual stress relieving mechanism provided by the actually complex non-linear axial friction behaviour within the slip lengths.6. ~7ratcheting surely attending the cyclic nature of in-service activity. In-service stress-relieving has been conceptually propounded elsewhere with respect to infilled prop studies.027~u (26) where Pqi = 80"76EI/L 2 denotes the idealised buckling force value ~ (L -= Lu) and Lu = LJF=0 = 0"7451Li (27) Equations (26) and (27) are quite distinct from the upheaval values obtained in previous isolated p r o p models 8 ~0 and this factor requires particular consideration. However. as herein.Subsea pipeline buckling 337 consideration of the important upheaval state (e.16 Although the physical prototype presently under consideration lacks the self-weight relieving presence provided by the propattendent fill of the infilled case. 12. As discussed previously. If the internal stressing of eqn (9) were to be incorporated within eqn (10) a priori with M~ ix = EIv~'. In comparison with the infilled prop case (recall Fig. eqn (5) is taken to prescribe a stress-free-when-initially-deformed datum state. UPHEAVAL This state. From eqn (17). This important matter will be subject to further deliberation following presentation of the complete model. of a stress-free-when-initiallydeformed datum.

(-v'. it should be noted that given the imperfect force-deformation relationship of eqn (16) F E1 .[. These are therefore equivalent to idealised studies ~in which the pipeline has been 'disturbed' or propelled into the idealised buckling mode at amplitude ItimIPqi ~ Uom ]Pqi" (Regarding overall system modelling.- = L I& ` = 0.~° i. 10 = qL~ 3EI (F = 0) (32) This is true 7 for upheaval and beyond as described in the following. . it is a high risk assumption to be definitive about only that component which nullifies imperfect behaviour and is based upon a historically non-existent state. from eqn (7).~ I.775L~ (29) would ensue as eqns (5)-(9) represent the deformed state solution of a problem in which the (previous hypothetical) datum state was stressfree-when-straight. ! v . Second..-10) (30) then for F = 0. H~I = 80"76 77 = Pqi Lu where ''4 E1 (28) = .. 7-~2there is little doubt that the precise and componentonly elastic interpretation given by eqn (9) fails. in the absence of comprehensive as-laid residual stress data. Finally...:. a stress-free-when-straight pipeline has been subjected to displacement Vom under inertial loading q and then compressed by P. whilst the previous in-service considerations are not to be taken to suggest that complete relieving of all residual stress components is thereby provided.~.. First.~.. This is effectively implemented in previous isolated prop models.~. 8.. justification for the proposed prop model's conservative philosophy which results in the 37% loss in upheaval buckling resistance identified by comparing eqns (26) and (28) is twofold.) . ¢ = Vixxxl0 P (F = 0) (31) such that..e. thermal values may be only approximately idealised therein due to the employment of simplified compatibility assumptions?) Summarising.. there is the implicit kinematic requirement v.338 Neil Taylor.(-vi'x. Vinh Tran p~. Should definitive residual stress data become availableJ ~-~: this could be readily accommodated within the present model by suitable modification ofeqn (10). non-conservatively. to replicate a duly definitive in-service imperfect datum state.

boundary conditions taking the form ~! 1 N __ v ~. 8 ~ v N. As indicated in Fig. Figure 5 illustrates the initial post-upheaval stage with Fig.LLi Q jl q L . 5.initial post-upheaval:details of imperfectfully mobilised model (L < L 0. _ qLi V'xx" o--3-~ ~i ~ . 2. first with L < L~ and second with L > L~ (see below). ~ ". mathematical modelling of post-upheaval buckling requires a two-phase structure.q!'~ + PVrn_ ¢llllllllllllllllllllllllllllll~r'~lllll/lllllllA . the tightening-up of the wavelength is reversed with L now growing as buckling ensues with further rise in temperature.Mx v-v.._L Topology USURP OAqL/2 L/2 J~ LI= . 5(a) detailing the crucial flexural region. (a) F l e x u r a l v ~ ~ q / unit length . .Subsea pipeline buckling 339 POST-UPHEAVAL BUCKLING (Lu < L < Li) Upon upheaval.. x L/2 Li/2 L.~ Ls 'P <--__ Jl (c) Axial Force Distribution Fig. Isolatedprop ..=i Range Topology Buckle under thrust P Vm>Vom L<Li ~V ~ Initial Imperfection curve L Ls (b) G e n e r a l ~AqL/2= U s .

e.~-qx2+ N (35) noting that eqns (12) and (32) remain valid..2 H [ .7-.. 5(b) together with the axial force distribution shown in Fig...sinnL + nLi 3 n (Li .x) (37) for0 < x < L/2..v) ./2 t ¢ ¢ = v ..o 1( t'/2'2 (vi ....nL )sinn(L/2 . with F = 0. eqns (17) and (32)).3421.x) + (n-~ .3LLi + 3L 2) 3 .. Suitable manipulation of eqns (32)-(35) generates the characteristic equation 2sin ~ + ( n ~ _ n L ) c o s n~ nLi 3 = 0 (36) Equation (36) is evaluated for nL for given values of L~/L (recall the treatment ofeqn (14)) and key values are given in Table 1./2 = v . it is again necessary to relate P to the temperature rise T(Po)..3L)cosnL + n?5 (L~ . values for amplitude o m are determined in turn from eqn (37).' -.. I.745 = 1/1..)2dx .) dx l(q)21[ ~(Li- n2 3L)2(nL + sinnL)sinnL .vi'.q EIn4 ( .e.340 ]Veil Taylor.. with the pre-upheaval flexure modelling previously discussed at the upheaval state is available from Table 1. and (20) are again employed with U = 2 j 1 °l. the respective and alternative statements for upheaval being U m = U o m (i.:. Noting the system topology shown in Fig. 5(c) then eqns (19). Having related buckling force P to amplitude vm and wavelength L. eqns (34) and (37)) and F = 0 (i./2(v.-1. = EI(v'. Io = v x.) = P(vm ... That the present modelling smoothly interfaces. noting eqn (34)....2 c o s n ( L / 2 . note 0. Vinh Tran v I.2 = 0 (33) with vlo = Vm Equilibrium affords for 0 < x <L/2 (34) M. The deflection expression becomes v . as required.

proceeding as previously but noting that the transverse deflection v =f(x.# + ~-~ (Li .-.n L ~-s i n~ - (q)2 L~ ~ 967 680 Figure 5 indicates that fully activated slip lengths are tacitly assumed although should the pre-upheaval flexure stage have resulted in this not being the case.. equilibrium affords Mx = EI(vx. eqns (24) and (25) are employed subject to F --. 6.1 ) .Vi xx) P ( v m -....Subsea pipeline buckling 341 + 4 [~ ( c o s ~L .Li) . L/2.lt/2 = V'xxiL/2 = 0 (42) together with matching conditions at x = L~/2 Vx v .V) qx2 (39) subject to boundary conditions O]O = Om.1 ) (43) Manipulation of eqns (39)-(43) affords the characteristic equation sin n L 2 2 cos~-+sin nLi 2 n ~ i co s nLi 2 nLi = 0 3 (44) . POST-UPHEAVAL BUCKLING (Li < L) The key features of this stage of buckling are illustrated in Fig.0 in place of eqns (19) and (20).x EI~ Ein 2 --c°s (L-Li)-sm~(L-Li)~ (L - .2 s i n nL ~.. then for 0 < x < L i / 2 .+ nL] 2cos nL~-+ 2]) (38) n .- q ( sin nL nL Li) + cos ~ (L .+ N (41) subject to boundary conditions v [L/2 = V'. V'x Io = o (40) whilst for L~/2 < x <.. L) is not everywhere attended by the continuous imperfection vi = g(x.v) . equilibrium affords .3 L ) [ . r -. Li). qx 2 M = EIv xx = P(vm ..2 .

The equations of the deflected curve take the following form: for 0 < x <Li/2 . As can be seen therefrom.. Isolated prop--post-upheaval: details of imperfect fully mobilised model (L > L0. 6. L Ls o. Vinh Tran q/unit length i i VmVom~ ~j o r-(a) Flexural Range IIIII/ 4[ L/2 L>/ Li Topology q __ Buckle under thrust pIVm>Vom Vm I J--~C-. Values for nL are obtained in terms of L~/L as previously and key values are given in Table 1. ~ Initial Imperfection curve /Vom ~ .q' (c) Axial Force Distribution °. not only does the solution forL > L~ interface smoothly with that forLu < L < LI.us ~'/~ Ls d (b) General Topology i °. but also as Li/L decreases."'prop'~ L/2 _L Ivl "'<"~ ~ L/2 us.q.342 Neil Taylor.~-~.~_~ p. the imperfect (elastic) solution converges towards its idealised (elastic) envelope as anticipated./-7 .q' L Fig.

nLi ..cos 2 nL i nLi nLi] 6 cos ~ ..~ ...n2x 2) v .sinnLi] + 18 + ~(nLi)2 [nLi + sinnLi] + 2k4 - nLik46 [cosnLi .] c nL osnx + [2 nL cos 7 .cosnLi] .+ 2sin 2 nLi and nLi _ cosT 2]) (48) fLL/2 (V 'x)2dx (q)2.nLi + sinnLi] + ~n3 [L 3 . though by this stage it is unlikely that the slip length modelling would not have become fully e s t a b l i s h e d .j cosnx (45) nL~ sinnx + 2 + (nL) 2 3 8 and for Li/2 < x < L/2 (n2~)z + n~Li x .) 2 d x + 2 1 JL.or eqns (24) and (25) if required.1] cos ~ .L!] + k2~ [nL + sinnL .s i n ~ L .sin ~ .cos ~ .c o s ~ . -EI ~ + ~ ~/2 [nL ...sinnL .sinnLi] [cosnL .with flexural end-shortening now of the form ( ( 1 [ t~/2 ' 2 L/2 (u'x) 2dx _ 2 JO (vi x) dx (47) U = [2jol L./2 for where foLJ2 (0 'x)2dx L > L~ k7 q 21 [nLi.s i n ~_] sinnx+ 1 + (nL) 2 8 n2x 2) 2 (46) The basic isolated prop modelling is concluded by the incorporation of eqns (19) (F = 0) and (20) -.Subsea pipeline buckling _ 343 q ([ nL nL nL .E~ qn 4 ( [ .-~./2(V.

3 206 000 I-144 350 11 × 10 -6 0.) dx = in which constants k4. Note that from eqn (6) tilv..=lOOmm ---- 31.T sin ~ ..m o u n t e d h = 0) Parameter External d i a m e t e r Wall thickness Elastic m o d u l u s Effective s u b m e r g e d self-weight Yield stress T h e r m a l coefficient Axial friction coefficient Poisson's ratio a Symbol D t E q Cry a ~A v Value 219 14.. _ 2 ~- ~Li/2'2 (vi. Vinh Tran [ nL [~_ -~-cos~-+sin~-+TCOST-sin-sin~-+cos nL nL nLi nLi nLi] nL.=25Om m = 39. k4 __. 7(a) and (b) illustrate key characteristics.3 Unit mm mm N/mm 2 N/mm N/mm 2 /°C av e m p l o y e d for e v a l u a t i o n o f pressure c o m p o n e n t as required. Two magnitudes of imperfection yore have been employed to distinguish between stable and unstable responses.344 +2k5 -2k6 and Neil Taylor.. _ n Z (q)2 L~ ~ 6 2- 483 840 (50) 2 k5 and k6 are determined as follows nL ..53 0.cos nLi sin -22 2 k5 = ...7m and Li]v..8m (52) TABLE 2 Pipeline Parameters ( s e a b e d .]) cos(49) nL nL 2 nLi s i n n L . .cos nL nLi sin nLi .cos ~ k6 = T cos ~ .sin ~ - nL nL nL (51 ) nL nL nL BASIC M O D E L CASE STUDIES The parametric data given in Table 2 have been incorporated within the foregoing formulations and Fig.

5 .5 Force C h a r a c t e r i s t i c s = Fig. . 7.. I t I I ~ t/Idealised° t6oo r.7m "~ T v s L (Vom=0.25m) 40 o. R e f s 8. basic model (h 0).d 0 0.10 o ~z00 G} \\ m 800 / E ~ ~° I~ 40 0 .0 Amplitude 1. .4 17 2() 23 2.. Fully mobilised isolated prop models. 0 0. . o t ~--------_. .5 Buckle (b) Buckle 1.5 Vm (m) 2.L=Li=31..Subsea pipeline buckling 345 100 J 1 80 o IG) ¢o ¢t" I I j. 2. . L 3'2 (m) 35 318 4'1 44 Buckle Length (a) T h e r m a l Action Characteristics 2000. 0 I >° . o I-- E 20 / \ ' ~y.0 2.0 ?-.5 v m (m) 2. . . .0 Amplitude 1. ldealised=~ 60 ~ \ \ \. .5 Buckle 1.

the use of trenching and burial.15 (54) for ~L = 0"75. ~ (m/q)lo=2oo = 1.05 and (rn/q)]o=3oo = 1. Trenching Trenching serves to protect the pipeline.t5 The following considerations serve to expand the applicability of the present isolated prop imperfection model accordingly. The effect of trenching upon buckling resistance can be gauged by the fact that with 0 < 30 ° from a geotechnical standpoint. where m is given by m = q(sin0 + ~LCOS0) (53) with 0 denoting the trench angle and t~e representing the fully mobilised lateral friction coefficient. The onset of yield stress or finite rotations provides for an alternative. unlike elsewhere ~ ~0. purely vertical upheaval would actually dominate as per the basic model. together with the employment of fixed anchorages. less demanding limitation in the case of stable postbuckling configurations. the only effect should the pipe seek to follow the trench incline is to substitute effective inertial force rn. Vinh Tran The overall impression is considered to be consistent with system responses obeying the idealised envelope. and de-trenching due to in-service buckling is to be avoided. 4 Idealised burial and fixed anchorage scenarios have been published previously. in particular. then Fig. Recalling Fig. 8. in place of q. the submerged self-weight of the pipe. Whilst upheaval temperatures are theoretically enhanced. advances in offshore practice include. l(b) and Figs 4-6. Indeed.346 Neil Taylor. Further case studies follow. 7 involve yield stress occurring before finite rotations. ENHANCED CONSIDERATIONS The foregoing model is applicable to a basic seabed lie topology subject to the obviation of lateral mode buckling. Within the flexural or buckling wavelength L. due allowance being made for prop 'height" Vom as transverse deflections v and Vomare now inclined as suggested in Fig. the more likely the occurrence of (undesirable) snap buckling with designers maintaining operating temperatures/pressures below the upheaval values for the snap cases at least. being downgraded from the idealised case due to the presence of the prop imperfections. 8 illustrates an appropriate trenched section. . continuous or discrete. The smaller the imperfection (Vom). the cases illustrated in Fig. the imperfect loci thereafter being shown in broken/dashed form.

41 for h = 3D = 650 mm. for simplicity. ~. 10 with regard to burial type (a). It is to be recognised that continuous burial could result in the voids being in-filled to an extent that prevents pre-upheaval flexure (recall Fig. alternative contact undulation modelling is required.(f. 8' 19 however. extended post-upheaval buckling vertical displacement v will require q' = f(v) through the buckle wavelength L as opposed to the constant value given above. the effect of continuous burial upon imperfect pipeline behaviour is exhibited in Fig. two of these involve trenching as shown and. 8"~7Accordingly. cover h (or h i + h2) > D. For this circumstance. The isolated prop modelling is as given previously with the simple provision that q is replaced by q + q' throughout with the axial friction I coefficient numerically modified as required 17(OA = q~A. 9. 8.~a. additional insulation and enhancement of buckling resistance. the data of Table 2 again apply together with q'/q = 7. the basic model analysis actually corresponds to a suitably trenched lie. this constant value should suffice in the early and critical. not least to the designer.say). Herein. l(c)). Burial (continuous) Burial provides damage protection. 7-12 . Trench section.~ ~ . The submerged self-weight of the pipeline q is now artificially enhanced by an amount q' due to overburden pressure throughout the modelling and empirical formulae forq'/q in terms of cover (h) are available in literature regarding cases (a) and (b). i. generally. stages of upheaval itself.e. Clearly.~ d| i Fig. Three typical burial topologies are illustrated in section in Fig.Subsea pipeline buckling 347 Trench Vmax= Vm (d/cose Inclined transverse displacement V Illl//I l III q/unit run t)J-.

. 1 l(a) whilst Fig.. The topology is illustrated in Fig..53 115 ~ ii 0 ~ ..j .. Typical burial topologies. ~ Fill th o ~ Fill hF ~ " •~".. E ... Discrete rock dumping (intermittent burial) Continuous burial is very expensive." •'. .6 m . Snap ~ E 120 ..-"-1~ inertial loading 1 ~ 160 v F-o) \ "\ . ~.. 200- II Idealised under equivalent Ir....348 Neil Taylor. • ~.k'~/ram I Stable 0. ~k(q + q').) . 7oS~ "~'w''' in~.5 ~ ~ ~yid 110 Buckle Amplitude ' (m) 2.5 Vm Fig..e. ~.i. Vinh Tran C o v e r : h or hl+h 2 _/ Seabed Fil..~ ~ " ° " '~'~ ~6B @~ . .. ~'-~"£~.buried pipe (h = 3D).4m 80 E I-.b 2.....- Snap 40 L @l. 9. .. 1 l(b) shows the axial force distribution applicable upon full activation of the peel point friction reaction (pAqL/2 and of the slip length Ls~ distributed friction force. Thermal action characteristics -.'3-~7~ E O~ ...''" ' '.=0.¢:-~ \.. ~- L =Li=23.. • o.. 10. Costs can be reduced by the employment of intermittent burial whereby rock dumping is undertaken at judicious locations along the pipeline) s Cost-effectiveness is served by additional friction force generation within the slip length.. analysis proceeds as previously discussed for the basic topology unless the overburden slip length L~2 is activated for L < L~ whilst checks must also be made upon the pre-upheaval flexure analysis to ascertain whether the overburden is also therein involved.." " ~ t D (a) Basic Trench and Fill (b) Rock Dumping (c) Combined Fig. i. I L = Li = 18. (Prior to this stage.

-~ j (58) .~ laAq ~ --~ (L l a p .L ) / 2 .L 2 and Ls2 = (57) 1 1+ ~A + LL~] + L. for L > L~. L ~.IF'o b) Axial Force Distribution Fig.L~2){I +q'~O'AI l _ \ q J (PAJ (56) where U can be evaluated.x Li/2 / L / 2 "J ~IAclL/2 ~ ~ _. equilibrium affords Po . Isolated prop with discrete dumping (L > Li shown).. Here./2 ~ Fap ~1 a ) Topology ~AQLI2~--I~ACI( L f a p . + (L:2+2L~.lIIrlfzrlrrlr/t/rfl/rtylf117 .L). I 1. The mechanics of the system are only modified with respect to the longitudinal equilibrium and compatibility expressions. with L~l .LD .P = OA + OAqL~] + O~q 1 +q Ls2 (55) and longitudinal compatibility becomes (Po-P)L 2AE U = --~Aq[L~.Subsea pipeline buckling Cl / u n i t length 349 P • FAP I "I n /rt. from eqns (47)-(50).

13 together with the appropriate axial force distribution.. whilst longitudinal compatibility becomes -(FaP+cPAq(Lf~----L-)-)(Lt2PAEL) . eqns (57) and (58) remain valid for both stages. It is assumed. depending upon when the overburden slip length is activated: a program suite is strictly required for this purpose. that L < LD and L~ < Lb. Note that forL < L~. There are a variety of particular slip length configurations to consider when analysing these systems. the fully mobilised axial friction force 0Aq being generated throughout the slip length (Lt~p.41 as previously. The respective topology is shown in Fig. (~g = (~A for simplicity and LD = 500 m. The effect of intermittent burial is typified by Fig.P = (PA q~+ Oa q(Lfav2 L) + F.(P°-P)L-2~4E U (60) where evaluations for U are determined in similar m a n n e r to those relating to eqn (56).350 Neil Taylor.L)/2.o (59) where F~p denotes anchorage capacity. Both L D and q' can be varied with length of dump (>~2L~2)then determined. Comments regarding the need for a program suite as mentioned above to cater for the variety of possible slip length configurations involved again apply here. given the purpose of intermittent burial. The effect of employing fixed anchor points is exemplified in Fig. Fixed anchor points Fixed anchor points enhance buckling resistance by simply absorbing some proportion of the pre-buckling force Po = AEaT. Ucan be evaluated from eqn (38) whilst for pre-upheaval studies U is determined from eqns (21)-(23). the latter capacities being in excess of 250 kN. 12 which relates to the parametric data of Table 2 together with q'/q = 7. where Lf-.~p denotes the spacing of the fixed anchors. Both the spacing Lr~vand the capacity Fap of the anchors can be varied. Longitudinal equilibrium affords Po . . the figure relates to the case of the peel point friction force OAqL/2 being fully activated. 14 which relates to the data given in Table 2 together with Lfap = 500 m. and Lt~v > L > L~. Vinh Tran where L D is the intermittency distance.

478 N/mm 2.5 1.i n t e r m i t t e n t b u r i a l . 8 m i zo lO o 0. 5O t-- Modified .~ Idealised~2 60. ra 40 ~.5 1. T h e r m a l a c t i o n c h a r a c t e r i s t i c s ..0 z~ LD Buckle (b) Effect of Amplitude v m (m) varying i n t e r m i t t e n c y d i s t a n c e s Fig.0 1...Subsea pipeline buckling 351 Modified ~#..5 2.0 Buckle A m p l i t u d e 1. I d e a l i s e d 12 ~ ( L D = 50Ore) '\ _ ~-" . 12..~ ~" 40 a~ E I. ..5 (a) E f f e c t of varying 60t o..5 v m (m) overburden q' 2.0 2.oo?. v I- 50.0 10 I I I I I I I ! 0.~ 30 q~ = 8. 3o E ilJ ~ L = L i = 3 9 ..

Vinh Tran L D/2 I P q /unit length I overburden of I density q' • ~ // Z ZZfTI Z /[[I'lflflffI/f[Z/Z[//. typified by Figs 7. Lsl --I LS Ls2 _[ PI ~Aq LSl I bl Axial Force Distribution Fig. 12 and 14. Alternative modelling8 ~0actually suggests that for any prop (imperfection) amplitude Vom. the lift-off or upheaval state. DISCUSSION The basic isolated prop model proposed here is quite distinct from previously recorded formulations.352 Neil Taylor. the lift-off buckling force corresponds identically to that afforded by idealised (i. 13. Isolated prop with fixed anchor points (L > L~ shown).8 ~0unlike these alternative models. is shown to suffer a potential 37% degradation in this resistance if the existence of a supposedly previous yet totally hypothetical stress-free-when-straight . non-imperfect) studies for Vm= Yoreas identified by eqns (28) and (29)J 2 Here. f L a) Topology L/2 L. 10. the present proposal affords elastically imperfect behaviour.. consistent with the concept that conservative parametric convergence towards the corresponding idealised solutions should result as the relative effect of any initial imperfection decays with increasing system deformation.e. so important to offshore designers.

-'..... 6 However.::.. Thermal action characteristics ...Subsea pipeline buckling t I 353 O. 14.0 Buckle 115 ' 2. the similarity in the respective upheaval lengths Lu as suggested by eqns (27) and (29) belies more substantial differences in the appropriate action/response characteristics as typified by Fig. Not only does the acceptance ofeqn (9) in conjunction with eqn (5) require the existence of an historically fictitious idealised lie.. i t t | Modified .0 z~5 = Amplitude Vm (m) Fig.~ Idealised 12 (Lfa p=5OOm) I i O- 5O °.'" . Given the complexities attending the hostile environment involved.8m tn O-yld ( Fap =78.m o u n t e d / t r e n c h e d lie (h = 0)..- 0 01. 7.8 kN ) E 10 EZ0 I-. The deformation characteristics given by eqn (5) are accepted for the present model on the basis of the support provided for eqn (6) by field observations.fixed a n c h o r points s e a b e d . be safely ignored. state is questioned.¢' oooe~ N _.~" n- .5 i i I i 1.. by comparison.. Further. .7m .... ]2 it is considered inappropriate and high risk to construct the analysis other than in accord with that well-established principle of .. the precise stressing formulation given by eqn (9) is not considered to reflect an accurate assessment of the state of residual stress in the pipe in the as-laid state.~. P G \ \ Snap L = L i = 31.. it also requires that residual stress due to fabrication and laying operations n]']2 can.. 30L = Li = 39.4 I--'40" \ x //je.

.. is not computationally amenable assuming the development of slip .7.. in-service pre-upheaval flexural and axial movement occurs by design with this p r o t o t y p e . Further support for this approach is available from infilled prop studies which similarly suppress any supposed as-laid residual stressing.7. it is surely inconceivable to suggest these components will accurately reflect in-service residual stress levels following numerous cycles of in-service non-linear axial friction response. ~7 The prototypes corresponding to the isolated and infilled prop topologies share the c o m m o n features of actually complex nonlinear axial friction behaviour and the initial bending moments supposedly suggested by eqn (9). Whilst the basic seabed-mounted model essentially relates to a purely trenched lie. maximum operating temperature/pressure rises .7. Noting eqn (6). a closed-form evaluation ofT. Therein. Vinh Tran elastic stability whereby the datum is prescribed as being stress-freewhen-initially-deformed.t h e buckle length/ temperature rise locus of Fig.~2. Whilst a closed-form solution is available for the crucial upheaval buckling force Pu as given by eqn (26).c l e a r l y cannot exceed the temperature rise at upheaval. 7 is particularly relevant h e r e . 10. then the ratios corresponding to the case studies involving eqn (52) are 1/317 and 1/160 respectively and are considered typical of offshore practice.. T = Tu say. idealised theory indicates that 50% N~. The model could accommodate definitive and comprehensive residual stress data. the crown and maximum moment. the effect is duly conservative.and consequent as-laid stress relief due to the onset of localised plasticity under thermal loading must be considered highly probable in a manner similar to that discussed elsewhere. the effects of employing enhanced burial and anchorage techniques are clearly shown in Figs 10. ~7 Indeed.recall the arguments concerning pressure-equivalent parameter T' in eqn ( 2 ) .354 Neil Taylor. for unstable/ snap cases. ~5 17 Imperfection-based data are thereby made available for design purposes. max < 0" 1r) delimits the stable post-buckling cases studies as shown in Figs 7. is due to self-weight considerations.~3As noted above. Although lacking fill support to assist in cyclic thermal stress relieving. 12 and 14. such stressing is considered to be relieved under in-service conditions due to the interaction of non-linear fill accretion and slip length axial friction behaviour with thermal cyclic loading. t2 Such 'conversion' into an imperfection of form would clearly be influenced by the out-ofstraightness ratio yore/L~. the remainder being due to the prop imperfection per se in the form 6EIuom/L2.6. should they become available. J6 Therein. whilst the onset of yield stress or finite rotations (v '. 12 and 14 with all-round improvements in buckling resistance being provided as anticipated.

unavailable to contact undulation models. widely adopted in contact surface modelling. . 10. Testing upon 6-m lengths of 3/8-in o. the above noted experimental/ theoretical correlation does serve to encourage confidence in the prop model's capabilities. Whilst the interesting asymmetric implications (note below) have been discussed elsewhere. zero peel point curvature. solution data for the post-buckling L > Li phase correspond with those produced by an equivalent model (for common prop height Vom)designed to deal with the infilled prop imperfection case in which buckling initiates with a blister developing upon the overbend crown 7 (recall Fig. 1) by virtue of the cusp upheaval (recall Figs 7. To-date. results in a singular change in direction of wavelength propagation (L) as amplitude commences its monotonic path. Accurate upheaval specification is provided by a simple make-or-break electrical contact. occurs at the crown throughout. Figure 15 illustrates the central region of the pipe during postupheaval buckling with an amplitude of approximately twice the prop height.Pu.106qL2/ EI(L = Lu) = -O. behaviour is effectively common for the two cases. l(c)). It increases from the imperfection value given by eqn (8) to -O. The implication is that whilst infilling of the voids reduces resistance to upheaval by preventing pre-upheaval flexural energy release. these latter values are available from eqn (15) with P -. Spurred by the admission of pipeline buckling failures in the North Sea. Qualitatively. the prop takes the form of a PTFE-coated steel blade of 30 mm height visible below the pipe. the isolated prop model action/response characteristics differ from those associated with contact undulation models (recall Fig. Maximum curvature. important to the buckling mechanism.Subsea pipeline buckling 355 length friction forces during pre-upheaval flexure. Whilst it is not claimed as comprehensive proof. with respect to isolated prop studies involving fixed anchor points and employing imperfection amplitudes of 20 mm and 30 mm. The latter questions the transversality condition. 9 the cusp is associated with the fact that the prebuckling flexure phase. 4-6 full thermo-mechanical system testing is presently being undertaken by several authorities. Theoretical buckle lengths at upheaval are within 2% of the corresponding experimental values.0588qL{/Elat upheaval. Qualitative observations include occurrences of asymmetry 9 and minor buckling of the supposed slip length in the proximity of the peel points. pipe suffering as-delivered imperfections is presently being conducted in-house. the theory presented here provides upheaval temperatures within 4% of the observed experimental cases (average of six tests).d. by the post-buckling state L = L~. Intriguingly. 12 and 14).

residual stressing. this is a complex matter: for example. CONCLUSIONS By not requiring reference to a fictitious stress-free-when-straight datum. preupheaval. Vinh Tran Fig. cyclic operation demands this stress-free-when-initiallydeformed scenario must be considered given its relatively conservative implications. . as-laid stressing data: this is a common feature of all elastic subsea pipeline buckling models available in the literature. The proposed model is capable of dealing with the various enhanced configurations presently being employed in the North Sea and. whilst residual laying tension should improve buckling resistance perhaps beyond idealised values. The proposed model thereby suggests interpreting the prop as generating an imperfection of form on the basis of a worst case scenario: whilst it is not suggested that the stress-relieving mechanism discussed would remove all as-laid. computer mounted. the fact that some degree of relief is highly probable under in-service. field observations have shown buckling failures. is readily suitable for design application. the isolated prop model described here is considered to present a consistent elastic interpretation of the corresponding prototype behaviour subject only to the provision of accurate residual. 15.356 Neil Taylor. However. Isolated prop experimentation.

M. pp. Nielsen.. In Proceedings of the 18th Annual OTC. In Proceedings of the 22nd Annual OTC. 3. A dedicated finiteelement model for analyzing upheaval buckling response of submarine pipelines. & Gan. Vol. & Hobbs. Hobbs. New design criteria for upheaval creep of buried subsea pipelines. Thermal buckling of offshore pipelines. S. 7.. 4( 1) (January 1984) 45-50. Houston. 14. Submarine pipeline buckling-imperfection studies. Texas. ASME. P. B. B. & Liang. P. Upheaval buckling of offshore pipelines: overview and introduction. May 1990. F.. 17. N. pp.. pp. Ju. 243-9. & Ellis. C. C. 4. W. S. May 1990. Richards.Subsea pipeline buckling REFERENCES 357 1. 4. 76-81. N. Journal of Engineering for Industry. & Guijt. 1(2) (January 1981 ) 2-10. B.a case story. P. C. In Proceedings of the 22nd Annual OTC. G. Texas. 5. Ballet. Houston. Texas. Transaction of the American Society of Mechanical Engineers. Thin-Walled Structures. 2. M. May 1989. J. 1989. In Proceedings of the 22nd Annual OTC. 4(4) (1986) 295-323.. & Pedersen. P. In Proceedings of the Offshore Mech. 11. Configuration of submarine pipelines during laying operations. Hutchinson. J. Lyngberg.. Upheaval creep of buried heated pipeline with initial imperfections... Journal of Constructional Steel Research. V. 16. Taylor. Houston.. Vol.. In Proceedings of the 22rid Annual OTC. & Kyriakides. 2. 1 (1988) 11-22. P. Timoshenko. A. Journal of Constructional Steel Research. May 1990. T. paper OMAE-89-812. R.Journal of OMAE. R. Taylor. G. R. Regarding the buckling of pipelines subject to axial loading. & Sluyterman. N. Theory of Elastic Stability (2nd edn). & Gan.. Taylor.. pp.. Houston. J. 540-50. A. Capri... A. Ellinas. J. Boer.. A. 13. In Proceedings of the 22ndAnnual OTC. Guijt. Tran. A. C. Pedersen. et al. C. New York. N. Pipeline buckling caused by axial loads. E. A. L. In Proceedings of 4th International Conference on Computational Methods and Experimental Measurements. J.. Italy. Taylor. 581-600. N. Springer-Verlag. 4. 6. 110 (November 1988) 355-64. 529-38. 12. C. J. & Gan. & Gere. Arctic Engineering Conference. Palmer. Van Helvoirt. Palmer. 269-82. Thin-Walled Structures (in press). Vol. J. J. The Hague. & Jensen.Journal of Constructional Steel Research. Hobbs. 15. N. Buckling considerations in the design of the gravel cover for a high temperature oil line. T. 1961. 573-8. 9. Asymmetric effects of prop imperfections on the upheaval buckling of pipelines. & Richardson. '96 (1974) 111218. 7 (1987) 55-74. S. 10. Nielsen.. F. May 1990. et al. 8. J.. T. E. Texas.. J. pp. E. D. May 1986. Design of submarine pipelines against upheaval buckling. Upheaval buckling failures of insulated buried pipelines -. Refined modelling for the vertical buckling of submarine pipelines.. Texas. D. McGraw-Hill. Klever. R. R. Houston. . Journal of Marine Structures. Houston. Thermal buckling of pipelines close to restraints... Interface modelling for upheaval subsea pipeline buckling. B. May 1990. Texas.

F. Vol. L. 563-72.. J. J. pp.. Vinh Tran 18. N. 59-60.. E. 19. 1982. 4. E. . G.358 Neil Taylor. pp. Texas. McGraw-Hill. Houston. Zrn. May 1990. Soil response for pipeline upheaval buckling analyses: Full-scale laboratory tests and modelling. M. Bowles. & Schotman. P. New York. In Proceedings of the 22nd Annual OTC. Foundation Analysis and Design (3rd edn). Schaminee.

buckling

buckling

- Section 7 - Hobbs Lateral Buckling Method
- UHB1
- Pipeline Buckling Caused by Axial Loads
- Uhb 2
- FEM Analysis for Pipeline Expansion
- Buckling
- Otc 20368
- 1990_Palmer_Upheaval_Buckling
- Damping Pipeline
- Uhb 1
- Otc 20385
- Otc 20317
- susea
- OTC-10713-MS-P
- Otc 23070
- Numerical Simulation of Submarine Pipelines In
- Buried Pipeline Analysis
- Upheaval Buckling Assessment by Dr N I Thusyanthan
- Design of HP HT Pipelines Against Lateral Buckling
- Aspect-96-155 Lateral Buckling Palmer
- Uplift Resistance of Buried Pipelines and DNV-RP-F110 Guidelines
- OFFPIPE2Excel_usingMATLAB_E
- Pipeline_Lecture_4.ppt
- subsea
- Free Spanning Analysis of Offshore Pipelines
- analysis and assessment of unsupported subsea pipeline spans
- pipeline design calculations.pdf
- Pipeline Buckling Caused by Axial Loads
- Thesis Subsea Umbilicals 2013
- Buckle Fomation
- Buckling Concept

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd