Department of Civil EngineeringISSN 1901-7278 DCE Technical Memorandum No. 025

A Literature Study on the Effects of

Cyclic Lateral Loading of Monopiles

in Cohesionless Soils

Kristian Lange RasmussenMette Hansen

Torben Kirk WolfLars Bo Ibsen

Hanne Ravn Roesen

DCE Technical Memorandum No. 025

A Literature Study on the Effects of Cyclic

Lateral Loading of Monopiles in Cohesionless

Soils

by

Kristian Lange Rasmussen

Mette Hansen

Torben Kirk Wolf

Lars Bo Ibsen

Hanne Ravn Roesen

June 2013

Aalborg University

Aalborg University Department of Civil Engineering

Scientific Publications at the Department of Civil Engineering

Technical Reports are published for timely dissemination of research results and scientific work

carried out at the Department of Civil Engineering (DCE) at Aalborg University. This medium

allows publication of more detailed explanations and results than typically allowed in scientific

journals.

Technical Memoranda are produced to enable the preliminary dissemination of scientific work by

the personnel of the DCE where such release is deemed to be appropriate. Documents of this kind

may be incomplete or temporary versions of papersor part of continuing work. This should be

kept in mind when references are given to publications of this kind.

Contract Reports are produced to report scientific work carried out under contract. Publications of

this kind contain confidential matter and are reserved for the sponsors and the DCE. Therefore,

Contract Reports are generally not available for public circulation.

Lecture Notes contain material produced by the lecturers at the DCE for educational purposes. This

may be scientific notes, lecture books, example problems or manuals for laboratory work, or

computer programs developed at the DCE.

Theses are monograms or collections of papers published to report the scientific work carried out at

the DCE to obtain a degree as either PhD or Doctor of Technology. The thesis is publicly available

after the defence of the degree.

Latest News is published to enable rapid communication of information about scientific work

carried out at the DCE. This includes the status of research projects, developments in the

laboratories, information about collaborative work and recent research results.

Published 2013 by

Aalborg University

Department of Civil Engineering

Sohngaardsholmsvej 57,

DK-9000 Aalborg, Denmark

Printed in Aalborg at Aalborg University

ISSN 1901-7278

DCE Technical Memorandum No. 025

A Literature Study on the Eects of Cyclic Lateral Loadingof Monopiles in Cohesionless Soils

K. L. Rasmussen, M. Hansen, T. K. Wolf, L. B. Ibsen, H. R. Roesen

Department of Civil Engineering, Aalborg University, Denmark

June 13, 2013

Abstract

Today, monopiles are the most typical foundation for oshore wind turbines. During their life-time large diameter, sti piles are subjected to millions of small cyclic loads due to environmentalforces. The long-term cyclic loading can change the granular structure of the soil surroundingthe pile. This may change the stiness of the soil-pile system and create an accumulated ro-tation of the pile. The behaviour of the soil-pile system is very complex and the influence ofsoil parameters, number of load cycles, and size, amplitude and characteristic of the load areexamined, as they all contribute to the rotation an the change in stiness. The scope of thisarticle is to outline current design methods and the state of the art knowledge within the subjectof long-term cyclic, lateral loading of piles.

1 Introduction

Todays focus on renewable energy sources as areplacement for fossil fuels and gasses has madethe oshore wind industry expand rapidly.Large farms with wind turbines still increasingin size are installed in rough environment andare subjected to lateral loads from wind, wavesand current. Monopile foundations are the mostcommon foundation of oshore wind turbines.Currently, these steel cylinders have reacheda diameter of 4 - 6 m and have a slendernessratio, L/D < 10, where L denotes the length ofthe pile and D is the diameter.

A wind turbine will, during its lifetime, besubjected to large loads caused by stormswhich describe the ultimate limit state (ULS).However, also smaller long-term cyclic loadswill aect the serviceability limit state (SLS)and fatigue limit state (FLS). These cyclic loadswill rock the pile and restructure the soil grainssurrounding the pile. This may change thestiness of the combined pile-soil system andinduce accumulated rotation of the tower dueto this change. Change in the stiness of thepile-soil system changes the frequency of thissystem which then can interfere with the exci-tation frequencies. The excitation frequenciesare the frequencies of the rotor and the blades,approximately 0.3 Hz and 1.0 Hz, respectively.The natural frequency of the tower is normally

designed to be in-between to avoid resonance(LeBlanc et al., 2010a). The design criteria isoften very strict due to operating behaviour andoften the accumulated permanent rotation ofthe tower must not exceed 0.5. As the rotationis an important factor in the design criteriait is important to investigate the eect oflong-term cyclic loading on the pile-soil system.In the present standards, i.e. DNV (2010) andAPI (2007) cyclic loading is not given muchattention. These standards use p -y curvesbased on few full-scale experiments for laterallyloaded slender piles and use a simple reductionfactor to reduce the ultimate soil resistance forcyclic loading. The eect of long-term cyclicloading of monopiles placed in cohesionless soilis possible to be a critical design factor andthe eect of change in load characteristic, soilparameters, number of load cycles have notbeen properly examined.

A new potentially critical load case, long-term cyclic loading, is possibly the main designcriteria and the eect of change in the abovementioned factors should be analysed. There-fore, the concept of degradation due to cyclicloading is of interest. Methods for determiningdegradation of p -y curves have been presentedby Long and Vanneste (1994) and Lin and Liao(1999) based on full-scale tests. Other authorshave tried to determine the cyclic load eect

1

by other theories; Testing of soil, small-scaletesting and numerical modelling. Niemuniset al. (2005) have suggested a model to predictaccumulated deformations based on laboratorytests on sand. Triaxial tests in combination withtheoretical and numerical models have beenused by Hinz et al. (2006) and Achmus et al.(2009) to determine the relation between cyclicloading and deflection. Small-scale experimentsare conducted by Peng et al. (2006), Peraltaand Achmus (2010), LeBlanc et al. (2010a)and LeBlanc et al. (2010b) using theories ondegradation and concept of superposition toevaluate cyclic loading eect on displacementand change in soil stiness.

The scope of this article is to outline thecurrent design methods, the state of the artknowledge on the topic and need for furtherinvestigations.

2 Behaviour of Cohesion-less Soil under Long-TermCyclic Loading

Subjected to cyclic loading cohesionless soil canexperience accumulation of excess pore waterpressure. The build-up of this will reduce theeective stresses causing cyclic liquefaction orcyclic mobility, cf. Figure 1.

Figure 1: Definition of cyclic liquefaction and cyclic mo-bility. (Ibsen, 1994)

The build-up of excess pore water pressure is asystem behaviour related to drainage conditionsand therefore more relevant for shallow founda-tions. For cohesionless soils of very loose den-sities a contracting behaviour can be observed.However, the monopile is a deep foundation nor-mally placed in rather dense sands, which makesthe concept of cyclic liquefaction less relevant(Lesny, 2010). During the lifetime of an o-shore wind turbine waves and wind will cause

millions of small cyclic lateral loads. Subjectedto those, cohesionless soil will deform both elas-tic and plastic. Theories on determining the ac-cumulated plastic deformation due to relativelylow long-term cyclic loading takes its origin inshakedown theory. Shakedown theory is origi-nally developed for elastic-perfectly plastic ma-terials. However, the theory is used to some ex-tend in soil mechanics, even though behaviourof soils are more complicated. Shakedown hasdierent deformation outcomes related to thetype of force applied. For the given problem oflong-term cyclic lateral loading elastic and plas-tic behaviour occurs initially. The shakedown isthe development of accumulated plastic strainswhere the plastic strain increments will decreasewith number of cycles and the material will sta-bilise with eventually only elastic deformationoccurring, cf. Figure 2.

Figure 2: Principle of shakedown due to cyclic loading.(Lesny, 2010)

However, when applying the shakedown theoryto soil mechanics in cohesionless soils the the-ory fits only partially. Cohesionless soil keepsdeforming even after long time repetitive load-ing and does not reach perfect elasticity, butwill keep deforming (Goldscheider, 1977). Theconstant development of strains can increase theaccumulated strains of the structure to a pointwhere it becomes unserviceable. Goldscheider(1977) investigated plastic shakedown in sand.After a larger number of cycles the plastic dis-placement increments will have become almostinsignificant. He suggested the allowable totaldisplacement was based on the number of cyclesfor the lifetime of the wind turbine with an ad-ditional small, negligible displacement, Figure 3.

Figure 3: Principle of plastic shakedown. After Peralta(2010)

2

3 Current Design Regula-tions

Reese et al. (1974) and ONeill and Murchison(1983) have formulated the theory on p -y curvesfor sand to describe the relationship betweensoil resistance created in the non-uniform stressfield surrounding the pile and the lateral dis-placement of the pile under lateral load, cf. Fig-ure 4. The bending of the pile is described bythe fourth-order dierential equation for beambending (DNV, 2010)

Ep Ipd4y

dz4+QA

d2y

dz2 p(y) = 0, z [0;L] (1)

where Ep and Ip are the elasticity modulusand the second moment of area of the pile,respectively. QA is the axial load from theturbine tower. The p -y curves are modelledusing the Winkler approach with decoupledsprings along the pile, each supporting a piledivision. For each spring a non-linear p -ycurve is created. These curves are adoptedand used in current methods for designinglaterally loaded piles in the standard codesDNV (2010) and API (2007). The methodsare highly empirical as they are fitted by onlya few full-scale experiments described by Coxet al. (1974). The experiments encompassboth static and cyclic test with up to 100 loadcycles. These experiments are conducted onpiles in sand with a diameter of 0.61 m andwith slenderness ratio about 30. Other testshave been conducted validating the p -y curvesbut all tests are conducted using slender piles.The basis for the p -y curves diers significantlyfrom the piles used as monopiles today as thedierence in slenderness ratio is pronounced andthe amount of load cycles in the tests are limited.

The procedure for creating the p -y curvesfor cyclic lateral load on monopiles in sand byDNV (2010) is

p = Apu tanh

k z

Apuy

(2)

User Manual 3 Program PYGMY

The University of Western AustraliaDepartment of Civil & Resource Engineering

5 Background TheoryThis program analyses laterally loaded piles by the subgrade reaction method, where thepile is idealised as a beam that is restrained from deflection by a series of distributedsprings along its length, Figure 5.1. The basic governing equation for this situation islisted below, including the effect of axial load on bending response.

0kydxydFdx

ydEI 22

4

4=!+ (1)

where:E = Young's modulus of the pileI = Second moment of area of piley = lateral deflection of pilex = distance along the pilek = modulus of subgrade reaction (spring stiffness)F = axial loadThe solution of equation 1 can be achieved using finite difference techniques, or with afinite element formulation of the beam bending equation. The program PYGMY uses afinite element formulation.

p-y springs

pressure, p

displacement, y

Figure 5.1. Idealisation of laterally loaded pile as a beam supported by springs.

If the stiffness of the spring is constant, then solution of equation 1 is relativelystraightforward. However, it is typical that k will vary with the amount of displacementat any point. Thus the springs are non-linear and are commonly called p-y curves, wherep = pressure and y = displacement. In this case an iterative solution procedure isrequired, using the secant spring stiffness.

pup

y

Es

p-y springs

Figure 4: Principle for describing soil behaviour with p-ycurves. (API, 2000)

where the p -y relationship is determined fromthe static ultimate load, pu. k is the initialmodulus of subgrade reaction, z is the distancefrom soil surface and A = 0.9 for cyclic load-ing. The p -y curves are formulated dependingon very few properties of the sand and the pilerespectively. For the sand, the angle of inter-nal friction, the relative density, and the specificweight are considered. The dimensions of thepile are considered in terms of length and diam-eter. However, the general behaviour of the pileis assumed that of slender piles. The monopilestoday have a slenderness ratio < 10 and so, thiswill give the piles a more rigid response whichis not accounted for in the current design guid-ances, i.e. DNV (2010) and API (2007). Thedierence in behaviour of flexible and rigid pileshas great influence on the soil behaviour andthe development of a toe-kick is significant forrigid piles, cf. Figure 5.

Figure 5: Principle for the behaviour of a rigid and aflexible pile.

Accumulation of displacement and change instiness of the soil-pile system are possible overtime due to cyclic loading. The relation betweenthe cyclic loading in coherence with number ofload cycles and the load amplitude is not con-sidered in the design standards.

4 Methodology for Long-Term Cyclic Loading

In order to incorporate the eect of long-termcyclic loading of a pile, the concept of degrada-tion is adopted by means of dierent methods.A degradation index is presented by Idriss et al.(1978) as Equation (3) to describe the change instiness and shape of the hysteresis loop as a

3

function of the number of cycles.

=EsNEs1

= Na (3)

where EsN and Es1 are the secant moduli of Nand 1 cycles, respectively, and a is the gradientof the regression line for a logarithmic scale, cf.Figure 6 and 7.

Figure 6: Degradation of stiness after number of cycles.(Achmus et al., 2009)

Figure 7: Degradation of stiness after number of cycles.K is the subgrade modulus. After (Briaud andLittle, 1988)

This degradation factor has become a generallyadopted concept in determining cyclic load ef-fects, leading to explicit methods for determin-ing the stress-strain relations for cyclic load-ing. The concept was continued by Briaud andLittle (1988) who proposed a power functionfor degrading the soil resistance as a functionof the number of load cycles. With origin inthis formulation and the static p -y curves Longand Vanneste (1994) analysed results from 34full-scale laterally loaded pile tests to investi-gate which model parameters influenced the be-haviour of the pile when repetitively loaded. The34 tests varied in many aspects from each other:pile type and installation method, length anddiameter of the pile, soil density, number of cy-cles, and load characteristic. The slendernessratio spanned from 3 to 84 covering both veryrigid and flexible piles placed in dierent cohe-sionless soils varying from loose to dense com-paction. The piles were loaded dierently; one-and two-way loaded, subjected from 5 to 500

load cycles. A degradation factor, m, was deter-mined, influenced by the cyclic load ratio, FL,The installation method, FI , and the soil den-sity, FD.

m = 0.17FL FI FD (4)

Long and Vanneste (1994) specified expressionsfor calculating soil resistance, p, and displace-ment, y, as a function of load cycles when us-ing static nonlinear p -y curves. The soil resis-tance was decreased while pile deflection was in-creased with increasing number of load cycles, cfEquation (5) and (6).

pN = p1N(1)m (5)

yN = y1Nm (6)

where the subnotation N denoted N cycles and

1 denoted the first cycle. The factor controlledthe relative contribution of soil resistance anddeflection and was applied so change in p -yrelation with depth could be incorporated. Thevalue of the factor varied from 0 to 1. However,changing the factor provided no improvementin results, so a constant value of = 0.6 was ap-plied, making the method independent of depth.

Lin and Liao (1999) also developed an ex-pression for a degradation parameter, t, toaccount for dierent model properties withthe purpose of calculating the accumulationof pile displacements. This was derived fromanalysis of 26 full-scale lateral load tests withpile slenderness ratios from 4 to 84, subjectedto a maximum of 100 load cycles. They derivedthe same factors of influence: Cyclic load ratio,, installation method, , and soil density, . Inaddition, the degradation factor was dependenton pile-soil relative stiness ratio expressed bya depth coecient, L/T.

t = L

T (7)

where the coecient changes with the modelparameters such as soil density, load character-istic and method of installation. To determinethe accumulated displacement the relationshipbetween strain, , and displacement, y, proposedby Kagawa and Kraft (1980) was used

=y

2.5D(8)

where D is the diameter of the pile. Kagawaand Kraft (1980) investigated displacement dueto lateral load and found that a large part of theaccumulated strain happened within a radius of2.5 m diameters of the pile. Lin and Liao (1999)

4

use the strain ratio, Rs, expressed by a logarith-mic function, to determine strains as a functionof load cycles

Rs =N1

= 1 + t ln(N) (9)

where N is the strain accumulation after Ncycles and 1 iss the strain after the first cy-cle. Additionally, Lin and Liao (1999) inves-tigate the combination of variable load ampli-tudes. Here, a principle of strain superpositionsimilar to Miners rule is used (Miner, 1945). Anadapted version is proposed by Stewart (1986)to superpose strains in triaxial tests. This the-ory yields that a specific amount of strain can bedeveloped for various numbers of load cycles atdierent load levels, cf. Figure 8. Thus, for co-hesionless soils it is assumed that at some pointthe maximum strain will have accumulated in-dependently of the size of the cyclic load; thenumber of cycles will dier instead. With ori-gin in Equation (9) the amount of strain for anumber of cycles at a given load level, Na, canbe found and from this, an equivalent number ofcycles for a smaller load level, Nb , is determined.

Nb = exp1

tb

1a1b

(1 + taln(Na)) 1

(10)

where t and 1 are the degradation factor andstrain for the first load cycles for the respectiveload cases. a and b denote two dierent loadlevels. For varying load amplitudes the totalamount of strain can be determined.

N(a+b) = 1b [ 1 + tb ln(Nb + Nb)] (11)

Figure 8: Method used in pile permanent displacementcalculations for mixed loads. (Stewart, 1986)

Lin and Liao (1999) used 20 tests to develop thedegradation factor, t. Measured and calculateddisplacements were then found for six additionaltests with change in load levels and load charac-teristics for each ten cycles. In Figure 9 resultsfrom Lin and Liao (1999) are presented. As com-parison, results from using the method by Long

and Vanneste (1994) are presented for one loadlevel along with the measured result by Briaudand Little (1988). For the first three load levels(up to 30 cycles) the calculated and measureddisplacements are much alike. However, at thefourth load level the calculations overrate thedisplacement.

Figure 9: Permanent lateral displacement for number ofcycles of test pile. After (Lin and Liao, 1999).

Both Long and Vanneste (1994) and Lin andLiao (1999) have presented simple methods forestimating displacements for piles. The disad-vantage of using these explicit methods is theiruse of an empirical foundation: Their methodsare based on experiments conducted on slenderpiles cyclically loaded to a maximum of 500 cy-cles. When using these explicit methods forlarger numbers of cycles, variation in charac-teristic and model dimensions should be investi-gated. Solving implicitly using the finite elementmethod, strains are determined for every loadcycle in the load history. This can accumulatecomputational error when calculating strains forthousands of cycles and the process is time con-suming. The studies already done on the eectof cyclic behaviour show that it is a very com-plex problem. The soil/pile system is aectedby material properties of both soil and pile, ge-ometry of the pile and the multifaceted loading.There is a need for experimental work that canvalidate and improve the theoretical basis so itfits todays problem of cyclic long-term loadingof monopiles.

5 Experimental Studies ofCyclic Loading

The formulations by Long and Vanneste (1994)and Lin and Liao (1999) are based on full-scaletests. The formulas are based on empirical datawith only a low number of load cycles. Furtherexperiments are needed as a basis for determin-ing the eect of long-term lateral loading. Thefull-scale test is the primary and best basis to

5

support the theory but it is very expensive andtime consuming. A small-scale test is thereforeoften used in several experiments to obtaindata from long-term cyclic loading which thenis converted to fit real conditions. Some of thenewer research on cyclic loading is presented inthe following.

When working with cyclic lateral loadingthe load characteristics are defined by the ratiosb and c (LeBlanc et al., 2010a). b describesthe ratio between the maximum cyclic mo-ment, Mmax, and the maximum static momentcapacity, MS . c describes the ratio betweenmaximum and minimum moment, Mmin, of aload cycle.

b =MmaxMS

, c =MminMmax

(12)

Peng et al. (2006) investigated dierent loadingdevices for small-scale testing and invented adevice themselves where the eect of long-termcyclic loading was examined. Most of theirfocus was on the actual testing device but sometest results were presented. They investigatedtwo-way loading of a 44.5 mm wide pile witha slenderness ratio of 9. The pile was placedin dry sand with a relative density of 71.7 %.The applied loading was in the ranges b =0.2 to 0.6 and c = (-1) to (-0.6) creating loadamplitude both in and out of balance. 10000load cycles were conducted for each test andwithin that range Peng et al. (2006) concludedthat the accumulated pile displacement wouldkeep increasing and that displacements werelargest for unbalanced loading.

A development in the concept of degrada-tion was made by Achmus et al. (2009) whoresearched the degradation of stiness incohesionless soils as a consequence of cyclicloading. Based on triaxial tests and FEM,design charts for determining deflection alonga pile as function of the number of cycles weredeveloped. The degradation was expressed bymeans of the ratio of the secant elastic modulus,cf. Figure 6. The secant modulus, Es, is elasticand dependent on the stress conditions alongthe pile.

Es = k at

mat

(13)

where k and are material parameters and atand m are atmospheric pressure and mean prin-cipal stress. The accumulation of strains andthereby the plastic strain ratio is estimated by

the semi-empirical approach presented by Huur-man (1996).

EsNEs1

= cp,1cp,N

= Nb1(X)b2

(14)

where cp is the plastic axial strain. The ratio ofsecant stiness and the ratio of the plastic axialstrain are determined between the N th and thefirst cycle. b1 and b2 are material parametersand X is the cyclic stress ratio which definesthe relation between major principal stresses forcyclic stress state and static failure state. Intriaxial tests the initial stress state is isotropicwith constant confining pressure during cyclicloading. For real in situ conditions the stressesare anisotropic so to overcome these dierencesa characteristic cyclic stress ratio, Xc, is defined.

Xc =X1 X01 X0 (15)

where indices 1 and 0 define states of loading andunloading. The outcome of the study is designcharts recommended by Achmus et al. (2009) forpreliminary design giving the deflection as func-tion of number of load cycles, cf. Figure 10. Thecharts provide deflection curves for up to 10000cycles. However, the study lacks the support offull- or small-scale tests.

Figure 10: Deflection-Number of cycles curve. (Achmuset al., 2009)

Peralta and Achmus (2010) conducted a seriesof 13 small-scale tests on rigid and flexible pileswith 60 mm diameter and slenderness ratio from3.2 to 8.3. The piles were tested in mediumdense to dense sand with relative densities,Dr, from 0.40 to 0.60. They were subjectedto one-way loading of varying load size, b, for10000 load cycles and the displacement of thepile was found.

The accumulated displacements obtainedfrom the experiments were compared with re-sults obtained from the power and logarithmicfunctions by Long and Vanneste (1994) and

6

Lin and Liao (1999) expressed in Equation (6)and (9), respectively. It was found that theresults from flexible piles fitted the logarithmicfunction best while the power function fittedthe results from the rigid piles best.

Small-scale tests were conducted by LeBlancet al. (2010a) who also put a great amountof work in to the scaling of model and realconditions. They made 21 tests on piles in sandwith relative densities, Dr, of 0.04 and 0.38;6 static and 15 cyclic tests. The pile had adiameter of 80 mm and a slenderness ratio of4.5. The tests were conducted with variation inb from 0.2 to 0.53 and c from -1 to 1 describingboth static loading and one- and two-way cyclicloading. The number of load cycles also variedfrom approximately 8000 to 65000. LeBlancet al. (2010a) suggest that the best fit of theaccumulation of rotation is a power function.

(N)

s=N1

s=Tb(b, Rd)Tc(c)N

0.31 (16)

where N is the rotation at N cycles, 1 is therotation after the first load cycle and s is therotation in a static test at a load equivalent tothe one provided by the maximum cyclic load,cf. Figure 11.

Figure 11: Method for determination of stiness and ac-cumulated rotation: (a) cyclic test; (b) statictest. After (LeBlanc et al., 2010a)

Tb and Tc are dimensionless functions dependingon the load characteristics and relative density.For Tb a linear relationship with b is founddepending on Dr, cf. Figure 12. A non-linearrelationship between Tc and c is also foundillustrating that the largest accumulation ofrotation happens when c= -0.6 which is atwo-way loading.A study on the change in stiness of the soil-pilesystem did not provide as clear results as therotation accumulation. It cannot be concludedhow the stiness is aected by the relativedensity. However, similar for all tests is anincrease in stiness with increase in number ofload cycles. This increase is contradictory to

Figure 12: Functions relating (a) Tb and (b) Tc to rela-tive density, Dr, and characteristics of cyclicload in terms of b and c. (LeBlanc et al.,2010a)

current methodology which uses degradationof static p -y curves to account for cyclic loading.

Achmus et al. (2011) presented a FE-modelbased on strain degradation to verify the resultsobtained by LeBlanc et al. (2010a) and foundgood agreement between the simulations andthe test results.

Based on the method by LeBlanc et al.(2010a) and a super positioning concept similarto Miners rule, LeBlanc et al. (2010b) createddesign charts for determining the accumulatedpile rotation due to random two-way loading.The procedure is based on a limited amount ofempirical data from small-scale tests and furtherresearch should by carried out to investigate thecomplicated behaviour of change in parameters.

7

6 Summary

Currently, the design guidance is limited inknowledge on long-term cyclic loading of later-ally loaded piles. They are based on full-scaletesting of slender piles subjected to a lownumber of cycles.

The issue of long-term lateral loading is ofcomplex matters. The physical behaviour ofsand subjected to load cycles is a continuousplastic deformation with decreasing deformationincrements. This eect of long-term lateralloading has been formulated by Long andVanneste (1994) and Lin and Liao (1999) asan exponential and a logarithmic expression,respectively, depending on a degradation fac-tor. Both authors find that the degradationfactor can be determined based on installationmethod, soil density and load ratio. In additionLin and Liao (1999) incorporates a depthcoecient. Still, these theories are based onfull-scale testing of no more than 500 loadcycles. The small-scale tests on laterally loadedpiles focus on a high number of load cycles,i.e. approximately 10000 cycles. Dierent loadscenarios with varying load characteristic andamplitude is tested with the outcome that Penget al. (2006), Peralta and Achmus (2010) andLeBlanc et al. (2010a) agree that the pile willkeep deforming and the exponential expressionby Long and Vanneste (1994) fits rigid pilesbehaviour.

The influence of long-term lateral loadingof oshore wind turbines is a multifacetedproblem. Though many author have studiedthe area it is clear that no general approachhave been accomplished yet and further studiesare needed.

References

Achmus, Kuo, and Abdel-Rahman, 2009.M. Achmus, Yu-Shu Kuo, and KhalidAbdel-Rahman. Behaviour of MonopileFoundations under Cyclic Lateral Load.Computer and Geotechnics, 36(5), 725735,2009.

Achmus, Albiker, and Abdel-Rahman,2011. M. Achmus, J. Albiker, and KhalidAbdel-Rahman. Investigations on theBehaviour of Large Diameter Piles underCyclic Lateral Load. 2011. ISBN:978-0-415-58480-7.

API, 2007. American Petroleum InstituteAPI. Recommended Practice for Planning,Designing and Constructing Fixed OshorePlat-forms-Working Stress Design, RP2A-WSD, 2007.

API, 2000. American Petroleum InstituteAPI. User Manual Program PYGMY. TheUniversity of Western Australia, Departmentof Civil and Resource Engineering, 2000.

Briaud and Little, 1988. J. L. Briaud andR. L. Little. Full Scale Cyclic Lateral LoadTests on six Single Piles in Sand.Miscellaneous Paper, Geotechnical Division,Texas AandM University, College Station,Texas, USA, GL 88-27, 1988.

Cox, Reese, and Grubbs, 1974. W. R. Cox,L. Reese, and B. R. Grubbs. Field Testing ofLaterally Loaded Piles in Sand. Proc. 6thOshore Technological Conference, Houston,Paper No. 2079, 459472, 1974.

DNV, 2010. Det Norske Veritas DNV.Oshore standard DNV-OS-J101: Design ofoshore wind turbine structures. Technicalreport DNV-OS-J101, 2010.

Goldscheider, 1977. M. Goldscheider.Shakedown and Incremental Collapse ofStructures in Dry Sand Bodies, 1977.

Hinz, Lesny, and Richwien, 2006. P. Hinz,K. Lesny, and W. Richwien. Prediction ofMonopile Deformation under High CyclicLateral Load. Institute for Soil Mechanicsand Foundation Engineering, Germany, 2006.

Huurman, 1996. M. Huurman. Developmentof Trac Induced Permanent Strain inConcrete Block Pavements. Heron, 41(1)2952, 1996.

Ibsen, 1994. L. B. Ibsen. The Stable State inCyclic Loading. Soil Dynamics and

8

Earthquake Engineering, Vol. 13, No. 1,6372, 1994.

Idriss, Dobry, and Singh, 1978. I. M.Idriss, R. Dobry, and R. D. Singh. NonlinearBehavior of Soft Clays During CyclicLoading. Journal of GeotechnicalEngineering Division, ASCE, Vol. 104, No.GT12, 14271447, 1978.

Kagawa and Kraft, 1980. T. Kagawa andL. M. Kraft. Lateral load-deflectionrelationships of piles subjected to dynamicloadings. Soils and Foundations, 20(4),1934, 1980.

LeBlanc, Houlsby, and Byrne, 2010a.C. LeBlanc, G. Houlsby, and B. Byrne.Response of Sti Piles to Long-term CyclicLateral Load. Geotechnique 60, No. 2, 7990,2010a.

LeBlanc, Houlsby, and Byrne, 2010b.C. LeBlanc, G. Houlsby, and B. Byrne.Response of Sti Piles to Long-term CyclicLateral Load. Geotechnique 60, No. 9,715721, 2010b.

Lesny, 2010. Kerstin Lesny. Foundations ofOshore Wind Turbines - Tools for Planningand Design. 1. edition, 2010.ISBN:978-3-86797-042-6.

Lin and Liao, 1999. S. S. Lin and J. C. Liao.Permanent Strains of Piles in Sand due toCyclic Lateral Loads. Journal ofGeotechnical and GeoenvironmentalEngineering, 125(No. 9), 789802, 1999.

Long and Vanneste, 1994. J. Long andG. Vanneste. Eects of Cyclic Lateral Loadson Piles in Sand. Journal of Geotechnicaland Geoenvironmental Engineering, 120(No.1), 225244, 1994.

Miner, 1945. M. A. Miner. CumulativeDamage in Fatigue. Journal of AppliedMechanics, 12, A159A164, 1945.

Niemunis, Wichtmann, andTriantafyllidis, 2005. A. Niemunis,T. Wichtmann, and T. Triantafyllidis. AHigh-cycle Accumulation Model for Sand.Computer and Geotechnics, 32(4), 245263,2005.

ONeill and Murchison, 1983. M. W.ONeill and J. M. Murchison. An Evaluationof p-y Relationships in Sands, 1983.

Peng, Clarke, and Rouainia, 2006. J. R.Peng, B. G. Clarke, and M. Rouainia. ADevice to Cyclic Lateral Loaded Model Piles.

Geotechnical Testing Journal, 29(4), 17,2006.

Peralta, 2010. K. P. Peralta. Investigationson the Behaviour of Large Diameter Pilesunder Long-term Lateral Cyclic Loading inCohesionless Soil, 2010.

Peralta and Achmus, 2010. K. P. Peraltaand M. Achmus. An ExperimentalInvestigation of Piles in Sand Subjected toLateral Cyclic Loads, 2010. ISBN978-0-415-59288-8.

Reese, Cox, and Koop, 1974. L.C. Reese,W. R. Cox, and F. D. Koop. Analysis ofLaterally Loaded Piles in Sand. OTCHuston, (paper no. 2080), 1974.

Stewart, 1986. H. E. Stewart. PermanentStrains from Cyclic Variable-AmplitudeLoadings. Journal of GeotechnicalEngineering, Vol. 112(No. 6), 646660, 1986.

9

Recent publications in the DCE Technical Memorandum Series

ISSN 1901-7278 DCE Technical Memorandum No. 025

DCE_memorandum_25_report.pdfIntroductionBehaviour of Cohesionless Soil under Long-Term Cyclic LoadingCurrent Design RegulationsMethodology for Long-Term Cyclic LoadingExperimental Studies of Cyclic LoadingSummary