Copyright 1999, Offshore Technology Conference

This paper was prepared for presentation at the 1999 Offshore Technology Conference held inHouston, Texas, 36 May 1999.

This paper was selected for presentation by the OTC Program Committee following review ofinformation contained in an abstract submitted by the author(s). Contents of the paper, aspresented, have not been reviewed by the Offshore Technology Conference and are subject tocorrection by the author(s). The material, as presented, does not necessarily reflect anyposition of the Offshore Technology Conference or its officers. Electronic reproduction,distribution, or storage of any part of this paper for commercial purposes without the writtenconsent of the Offshore Technology Conference is prohibited. Permission to reproduce in printis restricted to an abstract of not more than 300 words; illustrations may not be copied. Theabstract must contain conspicuous acknowledgment of where and by whom the paper waspresented.

AbstractFree hanging metal risers have become an importantalternative to flexible risers for oil and gas field developments.These risers also have a potential benefit when used in hightemperature and high-pressure applications.

This paper presents a summary of the work performed toestablish Steel Catenary Riser (SCR) concepts for two fields inNorth Sea. They are: Statfjord C, a gravity based concreteplatform located on the Norwegian continental shelf in a waterdepth of approximately 145 m and Heidrun, with a concreteTLP at a water depth of 345 m.

These developed configurations fulfil both the UltimateLimit State (ULS) conditions and fatigue due to first orderwave action and due to vortex induced vibrations. Also, asshown in this paper, the Fatigue Limit State (FLS) governs theglobal configuration of the SCR concept.

In order to achieve a confident design, several designaspects have been studied in detail:

First order wave loading

Vortex Induced Vibration (VIV)

Diffraction effects (from the large volume structure)

Riser/Soil interaction

Fatigue capacity

IntroductionA number of research and development projects are currentlyevaluating the applicability of the SCR concept to floatingproduction systems mainly in deep-water environments (e.g.Karunakaran et al. (1996), Hatton et al. (1998)). However, asshown in this paper, the SCR concept could be an attractivealternative also for tie-in of pipelines to fixed platformstructures, like Statfjord C, see Figure 1.

Even in the absence of top-end motions (as for floatingproduction units), the design challenges for a SCR concept for

this application are significant. Due to the relatively shallowwater and quite severe wave and current environment, the riseris subject to large hydrodynamic loading causing extensivedynamic behavior.

For the Heidrun TLP shown in Figure 2, the designchallenge for the metal risers is due to the riser dynamics fromwave loading and the platform motions. Furthermore, for thisconcept the diffraction effects proved to be a key factor forfatigue response.

In this paper the developed SCR configurations for boththese fields are discussed along with the key issues governingthe design of such riser concepts.

Figure 1 Statfjord C and riser geometry

Figure 2 Heidrun TLP

OTC 10979

Steel Catenary Riser Configurations for North Sea Field DevelopmentsDaniel Karunakaran, MARINTEK, Kjell M. Lund, STATOIL and Nils T. Nordsve, STATOIL

2 KARUNAKARAN, LUND AND NORDSVE OTC 10979

Design CriteriaThe riser design is to comply with the NPD (1990)regulations, which means that:

The developed configurations are to fulfil PLS, ULSand FLS limit state criteria

The extreme stresses are to be checked by the workingstress method

The design lifetime shall be obtained using a factor of0.1 on the calculated fatigue lives. The minimumfatigue life required for the risers is 20 years (projectrequirement).

Environmental conditionsThe riser is designed for the 100-year wave condition in

combination with 10-year current profile as specified bySTATOIL in their design brief.

Statfjord-C100-year design wave:

Wave height 28 m

Corresponding wave period 16.0 sec

The riser is designed for 10-year current velocity alongwith the above mentioned wave condition. The design currentprofile is:

At surface 1.0 m s-1At -50 m 0.8 m s-1At 141.6 m 0.65 m s-1At 144.6 m 0.0 m s-1

The water depth at this location is 144.6 m.

Heidrun TLP100-year design wave:

Wave height 29 m

Corresponding wave period 16.0 sec

The riser is designed for 10-year current velocity alongwith the above mentioned wave condition. The design currentprofile is:

At surface 0.7 m s-1At 34.5 m 0.6 m s-1At 69 m 0.5 m s-1At 138 m 0.4 m s-1At -276 m 0.4 m s-1At 340 m 0.5 m s-1At 345 m 0.0 m s-1

The water depth at this location is 345 m.

Material dataFor these riser configurations, the fatigue limit state is the

most critical one. So the selection of steel quality is dictatedby the fatigue performance of the steel quality.

From initial fatigue analyses, it was found that a fatiguecapacity corresponding to a class E design SN-curve, (NS3472), would be desirable in order to obtain the lifetimerequired.

Given a good girth weld performance and limitedmisalignment, it was believed that this was attainable fornormal carbon steel. However, due to exposure to a corrosivemedium, the fatigue properties are reduced due to the risk forlocalised corrosion. Using carbon steel for this application, nobetter fatigue capacity than a class F2 design SN-curve, shouldbe applied. As this was found to be unacceptable, analternative, corrosion resistant steel quality was selected, i.e.,Super Duplex steel.

Super duplex steel has very good corrosion properties andno localised corrosion is expected, either from the transportedfluid (production fluid, water), or the outside seawatercombined with cathodic protection.

A limited test program was carried out by The WeldingInstitute (TWI) in Cambridge UK, to verify the fatigue andcorrosion properties of super duplex in addition to assess theeffect of plastic cycling (in bending) of the riser duringinstallation.

After completion of the test program, the followingrecommendations were made:

Use of class E design curve with a StressConcentration Factor of 1.2 for the critical sections(inside grinding of the weld is necessary)

Use of class F2 design curve with a StressConcentration Factor of 1.0 for the remaining parts ofthe riser

The yield stresses for the super duplex material at thedesign temperature for the two risers are:

Production riser : 460 N/mm2

Water injection riser : 524 N/mm2

Soil-Riser interactionWhen the riser is subjected to oscillatory motion, there is a

complex interaction between riser movement and the sea-bedat touch down point (TDP), penetrating the riser into the soiland thereby increasing the soil resistance.

A proper description of the pipe-soil interaction istherefore very important for the accuracy in calculation ofriser fatigue damage. Depending upon the stiffness andfriction of the seafloor, out-of plane bending stresses will bemore or less concentrated in the TDP region when the riser issubjected to oscillatory motion.

In riser response analysis tools, the pipe-soil interaction iscommonly modelled by use of friction coefficients (slidingresistance) and linear springs (elastic soil stiffness). However,these parameters must be selected carefully in order toproperly represent the complex pipe-soil interaction.

The pipe-soil interaction model as described in Verley andSotberg (1992) and Verley and Lund (1995), which wasdeveloped for on-bottom stability calculations for pipelines,was used as a basis for selecting representative pipe-soilinteraction parameters for the wave response analyses.

OTC 10979 STEEL CATENARY RISER CONFIGURATIONS FOR NORTH SEA FIELD DEVELOPMENTS 3

During small and moderate wave loading (the seastatescontributing the most to the fatigue damage) the riser TDPresponse (movement) in the lateral direction is very small (inthe order of 0.2 pipe diameters). This will cause the riser to diginto the topsoil layer and create its own trench. This effect willgradually decrease as the riser gets closer to the underlyingstiff clay soil, where very limited penetration is expected. Thewidth of this trench will typically be 2-3 pipe diameters, whichleaves space within the trench for the pipe to move withouthitting the trench edges. During a storm build-up, the trenchwill gradually disappear as a result of larger riser motions inaddition to natural backfill. For the ULS condition, the pipe-soil interaction is found to be of minor importance even ifhigher lateral soil resistance is mobilised.

The following parameters are found to be representativefor this work:

Statfjord-C Sandy soil:

Lateral friction coefficient - 0.8

Soil lateral stiffness - 100 kN/m/m

Heidrun TLP Clay soil

Lateral friction coefficient - 0.9

Soil lateral stiffness - 7 kN/m/mNo problems related to the integrity of coating are

expected as a result of abrasion against the sea bottom.

Hydrodynamic coefficientsCommon practice for this type of marine structures is to

apply a CD in the range of 0.7-0.8. However, using a CD of 1.0would implicitly include a possible increase in externaldiameter from marine growth (it is assumed that cleaning willtake place if the marine growth thickness exceeds 20 mm). Nomarine growth was explicitly included.

The drag coefficient above 40 m is increased to accountfor the increased loading area in a lock-in situation due toVortex Induced Vibration (VIV). It is reported that the CD toaccount for VIV is about 1.2. However, in this study anincreased CD of 1.4 is used to account for both VIV andmarine growth.

To cover all options, It was decided to use the followingcombination of drag-coefficients:

CD= 1.0 in the top part and CD=0.8 at the bottom (Basecase)

CD= 1.4 in the top part and CD=1.0 at the bottom (VIVcase)

CD= 1.4 in the top part and CD=1.4 at the bottom (if VIVsuppression devices are used)

The inertia coefficient CM used in this analysis is 2.0. Dragand inertia forces in the riser axial direction are notconsidered.

Diffraction effectsThe flow around the platform is disturbed by the large

volume shafts and the concrete caisson in Statfjord C platformand by the concrete pontoons in the Heidrun TLP. In order toaccount for this in the fatigue analysis, the exact co-ordinatesof the riser configuration were used as input to a full linear

hydrodynamic analysis, which was performed numericallyusing the program WAMIT and the velocity RAOs along theriser length calculated. These velocity RAOs are applied indynamic response calculations.

Upper-end terminationIn this study the top end of the riser is assumed to be

equipped with a flex joint attached to the riser terminationpoint. In the global analysis model for ULS the top end ismodelled as pinned. However, in the fatigue analysis,rotational springs are attached to the top end, with rotationalspring stiffness of 30 kNm/deg.

Vessel data Heidrun TLPThe static vessel offset in connection with the extreme

response analysis is 21.4 m. In the fatigue analysis no vesseloffsets is considered. Also in the dynamic response analysisthe vessel motions are generated based on the vessel RAOsprovided.

Analysis procedureUltimate limit state analysis

The riser configuration is developed by satisfying the ULSdesign conditions. The basic configurations are obtained byperforming nonlinear dynamic response analysis using thedynamic analysis program RIFLEX, SINTEF (1995). Theresponse analysis is characterised by:

The riser is modelled by FEM principles usingdiscrete beam elements.

The static riser configuration is establishedconsidering the design current

Nonlinear dynamic response analysis is used applyingthe design wave and Stoke's V order wave model. Theriser configuration and tension are calculated at eachtime step by an iterative procedure and the dynamicresponse of the riser system is estimated using theNewmark- method, with constant averageacceleration algorithm.

For the ULS analysis, the diffraction effect is notconsidered, since for the long period waves, it is insignificant

The dynamic analysis was performed for both 0o and 180owave directions (in plane with the riser configuration).

Fatigue analysisIn this work, a non-linear time domain fatigue analysis is

performed. The dynamic response is obtained by performingnon-linear dynamic response analysis using Airy wave modelwithout applying current velocity. For the Heidrun TLP thevessel is kept at mean position for the fatigue analysis.

The step-by-step procedure used in the time domainfatigue analysis is described below:

Divide the wave scatter diagram into various blocks.The blocks used for Statfjord-C platform is shown inTable 1 and for the Heidrun TLP in Table 2.

Perform non-linear time domain analysis for onerepresentative sea state for each of these blocks. This

4 KARUNAKARAN, LUND AND NORDSVE OTC 10979

representative sea state has the highest occurrence ratewith in that block. The non-linear analysis includes thediffraction effect. The diffraction effects are modelledby the following steps:

Read the water particle velocity RAOs for everysection along the riser from a WAMIT result file

From the velocity RAOs the water particleacceleration RAOs are computed

Multiply the velocity RAOs by the wave spectrumfor the particular sea state in frequency domain

Perform inverse FFT to obtain time series ofwater particle kinematics at a given location

Calculate local force by Morison equation withrelative velocities

Perform non-linear time domain response analysis andcalculate stress time series

Estimate the fatigue damage with in each simulationusing rain-flow-counting procedure and weight thatwith the probability of each block. The fatigue damageis estimated at 16 points along the circumference ofthe pipe.

Sum-up the fatigue damage over all the blocks andobtain the fatigue damage for that direction

Perform the same procedure for all 8 direction andsum-up the total fatigue damage is estimated byapplying directional probabilities.

Apply a reduction factor 0.1, as per NPD regulationsfor un-inspectable welds and estimate the fatigue lifeas inverse of damage.

In this analysis a total time series length of 1.5 hour (5400sec) is used for every sea state. The variability of predictedfatigue life associated with the simulation length was studiedby performing three independent simulations of 1.5 hour eachat one sea state and the fatigue life was estimated. Thevariation in fatigue life was much less than 1 % between thesimulations. Hence, it is concluded that this simulation lengthis sufficiently long to estimate fatigue response.

One of the important parameters in the time domainfatigue analysis is the number of blocks used to sub-divide thewave scatter diagrams. A sensitivity study is performed toassess effects of this blocking. This is performed only for theHeidrun TLP, which has both vessel motions and largediffraction effect. Since the diffraction effects are veryimportant for wave periods below 11 sec., it was important toassess the influence of the number of blocks used in timedomain fatigue analysis.

In the base case analysis 17 blocks were used. In thesensitivity study, the blocks are increased to 45, with oneblock for every second in the peak period axis until TP = 11sec. The results indicated that by increasing the blocks to 45,the fatigue life has increased by about 6 % at MWL and byabout 10 % at the lower sections of the riser. This difference issmall in terms of fatigue life compared to other uncertaintiesrelated to the wave kinematics including the diffractioneffects.

Table 1 Blocking of wave scatter diagram for fatigue analysis - Statfjord C

TP (sec)HS(m)

2to3

3to4

4to5

5to6

6to7

7to8

8to9

9to10

10to11

11to12

12to13

13to14

14to15

15to16

16to17

17to18

18to19

19to20

20to21

21to22

22to23

23to24

24to25

0 1 *1 2 *2 - 3 *3 4 *4 5 *5 6 *6 7 *7 88 9

*

9 - 1010 - 11 *11 1212 1313 14 *14 15

(*) Denotes the sea state at which the nonlineardynamic response is simulated

15 - 16

OTC 10979 STEEL CATENARY RISER CONFIGURATIONS FOR NORTH SEA FIELD DEVELOPMENTS 5

Table 2 Blocking of wave scatter diagram for fatigue analysis - Heidrun TLP

TP (sec)HS(m)

2to3

3to4

4to5

5to6

6to7

7to8

8to9

9to10

10to11

11to12

12to13

13to14

14to15

15to16

16to17

17to18

18to19

19to20

20to21

21to22

22to23

23to24

24to25

0 11 2 * * *2 - 3 * * *3 4 * *4 5 * *5 6 * *6 7 *7 8 *8 9 *9 - 10

10 - 11 *11 1212 1313 14 *14 15

(*) Denotes the sea state at which the nonlineardynamic response is simulated

15 - 16

Riser OrientationStatfjord-C

The orientation of the platform and the riser with respect togeographical North direction is shown in Figure 3.

Figure 3 Riser and platform orientation Statfjord-C

Heidrun-TLPThe steel catenary riser is hung from a balcony on the

south side of the platform. The orientation of the platform andthe risers with respect to geographical North direction isshown in Figure 4.

ResultsUltimate limit state

Statfjord-CThe static configuration of the production riser is shown in

Figure 5, including the wall thickness along the riser. The topangle in the static position is 47.5o from vertical. The

maximum top tension is 1380 kN. The top angle has avariation of about 10 o from the mean position. The vonMises stress along the riser is shown in Figure 6. As seen themaximum stress occurs around MWL, and the stresses at TDPis low as expected. For these riser configurations the wallthickness at TDP are 12 mm and 11 mm for production andwater injection risers respectively. The thickness at mid-section is increased to 28 mm for both risers in order toachieve the required tension to fulfil the FLS criteria.

Figure 4 Riser and platform orientation Heidrun TLP

N

ERiser

23 m

N

51 mE

Steel CatenaryRiser

6 KARUNAKARAN, LUND AND NORDSVE OTC 10979

Figure 5 Static configuration with riser wall thickness -

Production Riser Statfjord CFigure 6 Stresses along the riser Statfjord C

Figure 7 Static configuration with riser wall thickness -Production Riser - Heidrun TLP

Figure 8 Stresses along the riser Heidrun TLP

Heidrun TLPThe static configuration of the production riser is shown inFigure 7, including the wall thickness along the riser.

The top angle in the static position is 16.5o from vertical.The top tension is 1040 kN. The top angle has a variation ofabout 7 o from the mean position. The von Mises stressalong the riser is shown in Figure 8. As seen for this riser, themaximum stress occurs at the TDP, mainly due to vesselmovement towards the riser. The thickness at mid-section isincreased to 25 mm for both risers in order to achieve therequired tension to fulfil the FLS criteria.

Fatigue analysisStatfjord-CThe results from the fatigue analysis are summarised in

Table 3. The fatigue lives presented here are after factoring by0.1. As mentioned earlier, two types of S-N curves are applied.For critical sections the E-curve is used with the condition thatthe weld will be ground both inside and the outside. For allother sections the F2 curve is applied. The critical sections attouchdown and at MWL are only about 20 m long.

Table 3 Fatigue life with factor 0.1 Statfjord CFatigue life(years)

At touch down point 27At mid-section 489Production

riser At mean water level 26At touch down point 26

At mid-section 543Water

injectionriser At mean water level 30

The main stresses at the TDP are due to the dynamicvibration of the riser since the second mode of vibration is at10 s, which coincides with the dominant wave action. Thedisplacement envelope of the riser is shown in Figure 9. Asseen the main vibration is in the riser second mode. Hence ifthe second mode of vibration is successfully damped, thefatigue life will improve.

-3 0 0

-2 0 0

-1 0 0

0

-5 0 0 -4 0 0 -3 0 0 -2 0 0 -1 0 0 0

Dept

h [m

]

H o r i zo ntal di s tance fr o m ce ntr e o f pl atfo r m [m ]

20 m m - 200 m 25 m m - 295 m20 m m - 120 m

0

100000

200000

300000

0 100 200 300 400 500 600 700

Stre

ss [k

Pa]

Line length [m]

Maximum Stress Curves - Production RiserPlatform in Near position & waves going out of platform

Cd = 1.0 Cd = 1.4

-150

-100

-50

0

50

0 100 200 300 400 500

Dep

th [m

]

Horizontal distance from top end [m]

12 mm thick - 131.5 m 28 mm thick - 325 m22 mm thick - 90 m

100000

150000

200000

250000

300000

350000

0 100 200 300 400 500

Stre

ss [k

Pa]

Line length [m]

W aves away from platform W aves towards platform

OTC 10979 STEEL CATENARY RISER CONFIGURATIONS FOR NORTH SEA FIELD DEVELOPMENTS 7

Figure 9 Displacement envelop

An increase in the projected area for drag damping at thelower half of the riser will increase the drag damping andthereby improve the fatigue life at TDP. A sensitivity studywas performed by assuming 100 mm thick coating/attachmenton the lower 100 m of riser. The results indicated this wouldincrease the fatigue life at TDP by as much as 65%.

The studies clearly indicate that the fatigue limit state isthe most critical limit state. Furthermore, the soil-riserinteraction is the most important parameter for fatigueresponse at TDP. The fatigue response at various locations canbe improved in the following ways:

The fatigue response at MWL is influenced mainly bythe wave loading and can be improved by increasingtension. which reduces the dynamic stresses. Alsosmaller diameter means reduced loading and therebyimproves fatigue life.

The fatigue life at TDP is influenced by the riserdynamics and soil-riser interaction. Since nothing canbe done with the soil properties, it is important tocontrol the riser dynamics. Similarly, increasing thedrag area in the lower part of riser increases the dragdamping, which also increases the fatigue life.Furthermore, for sand soil condition, providing theabsolute minimum weight of riser at mudline canreduce the transverse friction forces. This alsoimproves the fatigue life at TDP. A combination of allthe three points mentioned above can produce anoptimal riser configuration. A schematic diagramshowing the optimal configuration is found in Figure10.

Heidrun TLPThe results from the fatigue analysis for Heidrun TLP are

summarized in Table 4. The fatigue lives presented here areafter factoring by 0.1. The fatigue analysis is performedapplying diffraction effects and with undisturbed wavekinematics.

Figure 10 Schematic diagram of an optimal configuration

Table 4 Fatigue life with factor 0.1 Heidrun TLPFatigue life(years)

At touch down point 28At mid-section 548Production

riser At mean water level 25At touch down point 21

At mid-section 613Water

injectionriser At mean water level 26

For this riser, the comparison indicates that the totalfatigue life at MWL (for all directions) is reduced by a factorof 0.54, when diffraction is applied. However, at touch downarea there is no effect from diffraction. Also, the diffractioneffect is very dependent on wave direction, as shown inFigures 11 and 12.

Figure 11 Reduction factor for fatigue life when applyingdiffraction at MWL

Fatigue due to VIVThe fatigue responses due to VIV for both risers are

estimated using the computer program SHEAR7(MIT 1995).An equivalent vertical riser with the length equal to thesuspended length of the SCR configuration has been used tomodel the riser. This approximation is good for the out-of-plane modes, but not satisfactory for the in-plane modes.

-150

-100

-50

0

50

0 100 200 300 400 500

Dep

th [m

]

Horizontal distance from top end [m]

90 m - Small thickness100 m - Additional coating178 m - Heavy section120 m - Required thickness from ULS

0.0

1.0

2.0

3.0

0

30

6090

120

150

180

210

240270

300

330

Diffraction vs Undisturbed Kinematics - Fatigue life at MWLFactor for all directions - 0.54

8 KARUNAKARAN, LUND AND NORDSVE OTC 10979

Figure 12 Reduction factor for fatigue life when applyingdiffraction at Touch Down Point

The out-of-plane modes are excited by current velocity inthe direction of the catenary plane. Due to the riser inclinationangle only the current velocity component normal to the riseraxis was considered for the current in this direction, (the cross-flow principle).

The critical location for VIV induced fatigue is at TDP,where the fatigue life was approximately 40 years. But thecritical section for VIV fatigue does not coincide with thecritical section for wave induced fatigue at TDP. The fatiguedue to VIV at MWL is approximately 85 years after applyingthe NPD factor 0.1. This combined with wave induced fatigueis still above 20 years.

The model tests were performed in the towing tank atMARINTEK in Trondheim, Norway. The riser was fixed to avertical frame, which was attached to a wagon. During thetest programme the riser was exposed to:

uniform current only

combinations of waves and current

waves onlyModel testing of the riser has so far verified the drag

coefficient used for modelling the wave and current loadingand has also confirmed the conservatism in assessing VIV andwave response independent.

INSTALLATIONTwo different methods for installation have been evaluatedand are found feasible for the installation of the Steel CatenaryRisers investigated in this work.

The risers can either be installed by use of the reelingmethod or it can be towed from an onshore fabrication facilityto platform location using the Controlled Depth Tow Method(CDTM).

ConclusionsSCR concepts for Statfjord C and for Heidrun TLP aresuccessfully developed, both for production and waterinjection. The following different design aspects has beenstudied in an elaborate manner:

First order wave loading

Vortex Induced Vibration (VIV)

Diffraction effects (from the large volume structure)

Fatigue capacityThe Fatigue Limit State (FLS) is found to be governing for

the global configuration of the SCR concept, in particular thewave-induced fatigue. Utilisation of drag damping byincreasing drag area in the locations of less hydrodynamicexcitation is found to give significant improvement to the riserdynamic response, thereby improves wave induced fatigue.

The diffraction effects have significant influence on thefatigue response at MWL for Heidrun TLP.

The VIV-induced fatigue appears more or less independentand does not add directly with wave induced fatigue. Modeltesting of the riser has so far verified the drag coefficient usedfor modelling the wave and current loading and has alsoconfirmed the conservatism in assessing VIV and waveresponse independently.

Due to the corrosive properties of the transported fluid,super duplex steel was selected in order to achieve therequired fatigue capacity. The validity of assuming a class Edesign curve with a Stress Concentration Factor of 1.2 for thecritical sections is confirmed.

As shown in this paper, the SCR concept could be anattractive alternative also for tie-in of pipelines to fixedplatform in relatively shallow water.

AcknowledgementsThe authors would like to acknowledge Statoil for permissionto publish the results presented in the paper. It is emphasisedthat the conclusions put forth reflects the views of the authorsalone, and not necessarily those of Statoil or MARINTEK.

The contributions from Dr. P.Teigen and Dr. K.H. Halse,both from STATOIL for diffraction analysis and VIV analysis,respectively are highly acknowledged.

ReferencesHatton, S.A., Willis, N. and Bowman, J. 1998: Steel Catenary Risers

for Deepwater Environments Stride, Offshore TechnologyConference, Houston, 1998.

Karunakaran, D., Nordsve, N.T. and Olufsen, A. 1996: "An EfficientMetal Riser Configurations for Ship and Semi Based ProductionSystems", Proc. Sixth Int. Offshore and Polar EngineeringConference, Los Angeles, 1996.

MIT 1995: SHEAR7 Program Theoretical Manual, Department ofOcean Engineering, MIT.

NPD 1990:Regulations relating to pipeline systems in the petroleumactivities, 30.04.1990.

SINTEF, 1995: "RIFLEX - Flexible Riser System Analysis Program -User Manual", MARINTEK and SINTEF Division of structuresand concrete report - STF70 F95218, 1995.

Verley, R.L.P. and Sotberg, T.,1992: A Soil Resistance Model forPipelines Placed on Sandy Soils, ASME J. Offshore Mech.And Arctic Eng., 1992.

Verley, R.L.P. and Lund, K.M. 1995: A Soil Resistance Model forPipelines Placed on Clay Soils, Proc. Int. Offshore Mechanicsand Arctic Engineering Conference ,1995.

0.5

1.0

1.5

2.0

0

30

6090

120

150

180

210

240270

300

330

Diffraction vs Undisturbed Kinematics - Fatigue life at TDPFactor for all directions - 1.03