Norway Published online: 27 Jan 2010.

Transcription

1 This article was downloaded by: [ ] On: 24 August 2015, At: 06:35 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: Registered office: 5 Howick Place, London, SW1P 1WG The IES Journal Part A: Civil & Structural Engineering Publication details, including instructions for authors and subscription information: Recent structural design considerations related to floating production systems Bernt J. Leira a a Department of Marine Technology, NTNU, Otto Nilsens vei 10, N-7491, Trondheim, Norway Published online: 27 Jan To cite this article: Bernt J. Leira (2010) Recent structural design considerations related to floating production systems, The IES Journal Part A: Civil & Structural Engineering, 3:1, 50-64, DOI: / To link to this article: PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content ) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at

2 The IES Journal Part A: Civil & Structural Engineering Vol. 3, No. 1, February 2010, TECHNICAL PAPER Recent structural design considerations related to floating production systems Bernt J. Leira* Department of Marine Technology, NTNU, Otto Nilsens vei 10, N-7491, Trondheim, Norway (Received 2 June 2009; final version received 27 July 2009) Recent studies related to structural design of floating production systems are reviewed. The following topics are considered: (i) Hurricane effects and other extreme environmental events, (ii) Design for arctic conditions, (iii) Application of models tests for design purposes and (iv) Design considerations for particular floating production systems such as spars, tension leg platforms, FPSOs and semisubmersible platforms. Keywords: structural design; floating production systems; spars (US Patent ); tension leg platforms; FPSO; semisubmersible 1. Introduction There is a continuous development and extension of concepts for systems for floating production of oil and gas resources. Presently, much attention is paid to environmental load effects due to hurricanes. Load effects created by arctic conditions are also focused upon in relation to future developments. Furthermore, application of model tests for verification and calibration of design tools becomes more and more essential as new concepts are proposed. These topics are addressed in the present article in the light of recent developments. 2. Hurricanes and other extreme environmental events The successive severe hurricanes such as Ivan, Katrina and Rita gave rise to a strong motivation to update the criteria and procedures of designing floating platforms in the Gulf of Mexico (GoM). The harsher environmental criteria based on hindcast of the hurricanes are also implemented by API, especially for the central region of GoM. The impact of these hurricanes was considered by Versowski et al. (2006), Ward and Gebara (2007), Wisch (2006), Berek et al. (2007), Forristal (2007) and Wisch and Ward (2007a,b). Prior to the 2004 hurricane season, offshore practices and standards continued to evolve by incorporating new technology and experience. The disruption of the industry caused by damaged facilities and infrastructure was unprecedented. A critical review was undertaken to identify where standards may need to be adjusted to increase robustness in the future, where gaps may exist and what elements are outside traditional scope. The industry stakeholders responded with several interim guidelines coupled to multi-year programmes of investigation and update of standards. The topsides performance of the floating systems was dominated by the movement of drilling derricks and the damage done by derricks that toppled. Most notable of the floating system damage was the Mars tension leg platform (TLP) derrick incident in hurricane Katrina which not only damaged the drilling packages but also did significant damage to the production system as the derrick package moved. Press reports indicated that Mars repair work exceeded 1,000,000 man hours over a 9-month period shutting in a pre-storm production rate of *175,000 barrels a day. Other significant rig movement incidents included the Medusa and Devils Tower Spars and the Ram Powell TLP during hurricane Ivan. The industry responded with a technical bulletin, API Bulletin 2TD (2006b), prior to the 2006 hurricane season providing a high level guidance for the tiedown of drilling packages and other topsides equipment. Work has been initiated to revise documents associated with topsides equipment design and tiedowns. Reports indicated that the hurricane forces caused mooring lines to break, anchors to drag and other component failures. Following the performance of the Mobile Drilling Unit (MODU) fleet in hurricane Ivan, some vessels were outfitted with global position systems prior to the 2005 hurricane season and were then required for all rigs prior to the 2006 season. A primary objective was to more rapidly locate any rig that broke loose in a storm as well as reduce the aerial overflights to assess any environmental consequences. Fortunately, none of the floating drilling rigs was * bernt.leira@ntnu.no ISSN print/issn online Ó 2010 The Institution of Engineers, Singapore DOI: /

3 The IES Journal Part A: Civil & Structural Engineering 51 destroyed. The jackup fleet however did have some vessels that were destroyed due to the storms. Usually these failures result from leg failures due to waves impacting the hull or foundation failures, and cause the vessel to sink nearby or become grounded in shallow water. Industry started responding to loss of station issues related to the MODUs with the issuance of interim guidelines API RP 95F (2006a) and API RP 95J (2006c) prior to the beginning of the 2006 hurricane season. Work is continuing to further enhance the MODU station keeping. The hurricanes of 2004 and 2005 impacted much of the GoM oil and gas infrastructure including some elements that had not been seriously impacted in previous hurricanes. The API Series 2 documents address key aspects such as offshore fixed and floating structures, floating vessel mooring and riser designs. Though not covered in much detail historically, the API RP 2TD for rig and topsides equipment tiedowns is being updated and the development of additional content to address hardening and/or protection of production system elements is also underway. An important path to reviewing technology is through failure investigations. When extreme environmental conditions occur in which most units respond well or at least satisfactorily, the assessment of exception cases can be taken as a good starting point for such failure investigations. Hence, especially on account of hurricanes Rita and Katrina (during the 2005 hurricane season), which had a notable impact on offshore industry at GoM region, this can be taken as a promising scenario to assess the losses and bring out cases to be more carefully appraised. The case of the TLP Typhoon is shown in Figure 1. This TLP was found upside down after the passage of Rita. Being the only unit located at a significant water depth which followed recent project standards, the case naturally came to the fore. The paper by Timerman et al. (2008) addresses an investigative analysis held out in order to establish and analyse the possible causes that resulted in the capsizing of the referred unit. The investigation focuses on two major paths which were selected as being the most probable and also had feasibility to be carried out one that analyses a possible lack of displacement due to the passage of a hurricane wave and the consequent loss of stability of the TLP and the other, concentrated on a dynamic analysis of the unit under the impact of the hurricane metocean conditions. This second analysis focuses on obtaining the tension on the tethers stating if either they suffered excessive loads causing them to break or lack of it, causing compression and possible buckling. Even though results need to be enhanced by more realistic models, the investigative analysis has appointed actual possible flaws that might have occurred and could have led to the capsizing of the Typhoon TLP. Full-scale measurements were also obtained from other platforms during the hurricanes, such as the Marco Polo TLP, see Mitchell et al. (2006), Van Dijk and Van den Boom (2007), and the Horn Mountain Spar, see Xu et al. (2007) for hurricane Ivan, and Halkyard et al. (2004), Tahar et al. (2005, 2006) for hurricane Isodore. The monitoring system installed on the Marco Polo TLP (see Figure 2) was in operation during the passage of hurricanes Ivan, Katrina and Rita. Although Marco Polo was extremely close to the centre of these severe hurricanes, no significant damage was inflicted to the platform, even though wind speeds in excess of 138 mph and maximum wave heights over 28 m were Figure 1. Artist s impression of Typhoon TLP on the left and the unit upside down after Rita. Figure 2. Marco Polo instrumentation.

4 52 B.J. Leira measured. However, very valuable data were collected on the wave, wind, current, as well as on the response of the TLP during the hurricane conditions. As part of a Joint Industry Project, measurements on board of the Marco Polo TLP have been ongoing for nearly 3 years without interruptions. Although the motions of the platform have some effects on the measured wave elevation, in general these effects are small. For detailed analysis of extreme waves these factors can be taken into account. The three major hurricanes all passed Marco Polo at close distance. The significant wave heights in these three hurricanes as observed on board Marco Polo ranged from typical 10-year return conditions to 100-year return conditions (i.e. H s from 8.5 to 12 m, T p in the range of s). However, the measured extreme wave heights exceeded the expected extreme values. In hurricane Rita, a maximum wave height of 26.9 m was observed (including the correction for platform motions) with an associated crest height of 17.4 m. In some of the extreme waves that were observed on Marco Polo, high-frequency vibrations were observed on the topsides, which indicate impact loads of these extreme waves on the columns, although no structural damage was observed. The Horn Mountain Spar is shown in Figure 3. This Spar experienced the full strength of hurricane Ivan, a virtually 2500-year wave and 600-year wind in the GoM in The measured and hindcast sea state reached a 53-ft significant wave height and 88-knot wind speed (1-h average at 10 m elevation), which is significantly higher than the spar s maximum design criteria of 41.7 ft wave height and 78.2 knot wind speed. Both the environment and spar responses in the hurricane were measured. However, 6 h prior to Figure 3. Horn Mountain Spar. the peak of the storm, the generator which powered the instrumentation stopped, preventing further measurements. In order to estimate the spar s actual responses at the peak of the storm and to evaluate the accuracy of the existing spar motion analysis tools, a blind analysis test was performed using the measured and hindcast environmental conditions as input. The analysis results were then compared with the last available measurements of the spar responses in the hurricane. As the analysis from the design phase was known to be conservative, a fully coupled spar motion analysis model was used in this blind analysis test in order to achieve a higher level of accuracy. The environmental condition was also modelled to a higher level of detail than in the design phase, including wave spreading and dual-peak wave spectrum. Some minor adjustments to the fully coupled model were necessary to achieve the desired accuracy. From the blind test and the subsequent motion analysis for the peak of the storm, the following conclusions can be drawn: (1) The result from the fully coupled analysis model generally agrees very well with the measured spar motion responses early in the storm (when measurements were still available) and (2) The spar s pitch, roll and heave responses were estimated to be within the design limit even though the environment measured at a nearby platform significantly exceeded the maximum design condition at the peak of the storm. There was no drilling rig on the spar at the time of the storm. The calculated Line-6 tension was found to be significantly higher than the design criteria (new chain break strength ¼ 4246 kips), even for the reduced current velocity. The tensions at the wire rope top and anchor were also significantly higher than the design allowable. An underwater inspection of all mooring lines was performed in October 2004 and an underwater survey of all mooring line catenaries was performed in September Neither the inspection nor the survey revealed any damage to Line-6. The findings seem to confirm both the robustness of the current spar design practice and the accuracy of the spar analysis tools. Halkyard et al. (2004) and Tahar et al. (2005) have compared measured spar responses such as motion and mooring line tensions with numerical predictions. In Halkyard et al. (2006), the work was extended based on comparison of the full scale data during hurricane Isidore. Results of time domain and frequency domain simulations were compared with field measurements. Particular attention was placed on the importance of the phase relationship between motion and excitation force. The time domain analysis has better agreement with the field data compared with the frequency domain. Overall, however, the frequency domain

5 The IES Journal Part A: Civil & Structural Engineering 53 method is still promising for a quick and approximate estimation of relevant statistics. With advantages in terms of CPU time, the frequency domain method can be recommended as a tool in pre-front end engineering design or in a phase where an iterative nature of design of an offshore structure takes place. Examples of application of the updated design criteria to new structural concepts are given by Murray et al. (2008) and Yang et al. (2008). In the former paper, two semisubmersible designs are presented that assume a common topside and riser payload. The two designs are sized and analysed for the new GoM metocean criteria. The comparison is based on hull dimensions, including heave plate and structural support construction. Performance focused on riser response, specifically stroke and tension. In both cases, the designs meet the criterion of keeping the riser stroke under 30 ft. However, damaged conditions, such as broken mooring lines and a flooded hull compartment, are listed as topics for further investigation. The latter paper deals with a numerical study of the transient effect of tendon disconnection on global performance of a so-called extended TLP for harsh environmental conditions in the GoM. Twelve tendons support the platform with 12 production top tension risers (TTRs) and one drilling riser attached by hydropneumatic tensioner. A sudden disconnection of one or more tendons causes the unbalance of force and moment of the total system, only to cause the transient motion and tension as well as the mean offset. The transient responses and the mean offsets are compared and discussed in the viewpoint of the safety of the system. The post-katrina API 10-, 100- and 1000-year hurricane conditions are applied (API 2007a d). The survivability of a TLP under severe environmental conditions was investigated by Mansour et al. (2006). Survival conditions related to tendon damages were considered. Tendon damage conditions include tendon slacking condition, unpredicted increase in wave/wind condition after one tendon is removed and sudden tendon breakage. The effect of column spacing on the TLP survivability is also investigated. Numerical results are presented for the tendon tension time histories including any tendon downstroke or unlatching events. On the basis of the analysis results, recommendations are made to improve the survivability of the TLP under extreme environmental conditions. Other extreme environmental conditions than those caused by hurricanes are considered by Guedes Soares et al. (2003), Guedes Soares and Cherneva (2005), Dankert and Rosenthal (2004), Krogstad and Barstow (2004), Holliday et al. (2006) and Lehner and Rosenthal (2006). In the latter paper, results obtained from the European Project MAXWAVE are described. The project dealt with both theoretical aspects of extreme waves as well as new techniques to observe these waves using different remote sensing techniques. The final goal was to improve the understanding of the physical processes responsible for the generation of extreme waves and to identify geophysical conditions in which such waves are most likely to occur. In the papers by Fonseca et al. (2005, 2006a,b, 2007) and Guedes Soares et al. (2006a,b), load effects due to abnormal waves were considered. In these papers, a method to calculate the responses of a Floating Production and Offloading System (FPSO) in deterministic wave traces with abnormal waves has been applied. The motions and global structural loads induced by a large set of such waves that were measured at various locations were computed. Fonseca et al. (2007) focused on the probability distributions of the motions and global structural loads which were induced by the sea states that included such abnormal waves. The paper presents an analysis of the vertical bending moments on a FPSO which were induced by sea states where abnormal waves were measured. The simulations with a time domain seakeeping code were carried out for storm durations of 3 h. It was concluded that the abnormal waves do not induce abnormal vertical bending moments, meaning that while the probability of occurrence of the abnormal waves is very low, the magnitude of the moments induced by these waves are not extreme to the same extent. The 3-h simulations in the storms in which the abnormal waves were detected, result in several wave sequences that induce moments larger than those due to the abnormal wave events. These wave sequences are characterised by groups of 3 or 4 nearly regular waves with a period that maximises the vertical bending moment response. Squalls have been present in the environmental specifications for floating units in West Africa for the last couple of years. It appears that such phenomena tend to be the designing factor for mooring systems of deepwater FPSO s (in spread or turret configuration) and offloading buoys, as highlighted by Legerstee et al. (2006). At the design stage, due to the lack of proper modelling/characterisation, squalls tend to be represented for design purposes by on-site recorded time series of varying wind velocity and associated relative headings applied from any direction. This leads to rapid changes in offsets and loads in the mooring lines induced by the transient response of the vessel to sudden load increase which is generated by such a representation of the squall. Through diverse simplified example calculations, Legerstee et al. (2006) illustrate the influence of the consideration of squalls in the design process. Present shortcomings in the

6 54 B.J. Leira modelling process, either in terms of extreme conditions, or in terms of operating conditions are addressed, knowing that such events are difficult to forecast. 3. Design for arctic conditions Design of floating production systems for arctic conditions is rapidly being focused upon due to the anticipated future developments in the extreme North. A summary of some papers addressing this issue is given here. The response of plates subjected to patch loads with length/height less than the dimensions of the plate field is of importance for the design of all types of hulls strengthened for operation in ice-covered waters and other comparable types of loading. Nyseth and Holstmark (2006) derived an analytical plastic capacity formulation for plates subjected to patch loading of any rectangular geometry configuration. The analytically derived formulations for the load capacity are based on plastic bending and yield line theory. Expressions for single and multiple patch loads are included, where the derivation is based on the assumption that the response for a single patch load is equal to the response from a sequence of identical patch loads located within the same plate field, and spaced a certain minimum distance apart. The expressions may serve as basis for rule requirements for plates subject to ice loads and other types of patch loads. For the considered longitudinally and transversely stiffened plate cases, nonlinear FE calculations show that the proposed plate bending based load capacity formulations generally give rise to permanent deformations in the plate upto 0.40% of the frame spacing. This requires that the analysed plate model also is extended to include adjacent plate fields. Comparison of the analytically derived plate formula with Det Norske Veritas (DNV) Arctic Rules and Finnish Swedish Ice Class Rules (FSICR) indicates that the plate formula of the DNV/FSICR gives similar results for the same load levels. The analytically derived formula, however, was seen to provide a more consistent utilisation of the plate bending based capacity that is valid for a wider range of patch load geometries. Comparison of the International Association of Classification Societies (IACS) Polar Class (PC) Rules formulation indicates that the IACS plate formulation is increasingly non-conservative for small patch lengths, and that the application should be restricted to cover larger patch lengths only. Chou et al. (2007) presented a concept of an innovative floating platform using a conical structure which is intended for operation in arctic regions. It is referred to as MonoCone Arctic Drilling (MCAD) platform. The conical structure is used to reduce iceloading as it facilitates ice to break in flexure while riding the slope of the conical surface. For supporting the weight of the platform, equipment, and ballast, a base structure with sufficient buoyancy is added at the base of the conical structure. The conceptual platform configuration has been analysed for a large payload of more than 25,000 ST operating in *125 ft of water depth. In the winter season, the platform is subjected to more than 12,000 ST of ice load. For warmer season, the platform has been designed to survive a 45- ft significant wave height with 80-knot wind, and a very strong current of 6 knots. A mooring system using 32 lines was designed. For lower ice loads in a milder environment, the number of mooring lines can be reduced considerably. 4. Application of model tests in relation to design verification Spar production systems are subject to vortex-induced motions (VIM) which may impact mooring and riser design. Helical strakes are employed to mitigate such motions. Model tests are typically required to validate the performance of the strakes. Halkyard et al. (2005, 2006), and Atluri et al. (2006) gave results from benchmarking studies that have been conducted over the past few years to compare model tests with computational fluid dynamics (CFD), see Figure 4 for an example of an applied mesh. They also presented best practices for the use of CFD for these classes of problems and issues related to turbulence modelling and meshing of problems at large Reynold s numbers. The aforementioned studies as well as other studies indicate that CFD may be used to successfully predict the occurrence of VIM. In particular, the hot spots for a particular spar configuration are identified and Figure 4. Mesh A.

7 The IES Journal Part A: Civil & Structural Engineering 55 the amplitudes are conservatively estimated. The evidence is that the amplitudes may be more accurately predicted by refining the mesh, particularly in the area of the boundary layer. Furthermore, it may not be necessary to use a large number of nodes if the solver is capable of handling distorted element shapes. Run times are practical for engineering purposes. At the moment CFD seems practical for looking into sensitivities in the early design phases. For example, CFD may be used to consider the relative affect of strake height and pitch and the impact of appurtenances. Similar to SPAR platforms, it was also assumed that monocolumn floaters (Figure 5) can exhibit VIM behaviour. Hence, an experimental investigation was started which focused on VIM responses of small-scale monocolumn floaters in a towing tank. Cueva et al. (2006) considered environmental conditions for the Campos Basin and GoM and presented a set of preliminary results. In accordance with other VIM studies, the strong influence of the heading in VIM was noted, increasing the maximum cross flow amplitude from 0.8 to 1.2D. The need of experiments in a wider range of headings is accordingly illustrated. An assessment is also given concerning VIM of monocolumn floaters and its impact on the mooring line design and riser specification. Application of model test data in global design of TLPs was addressed by Phadke et al. (2006). Rapid advances in computer technology have made it possible to employ sophisticated time-domain techniques as primary tools for the global performance analysis of TLP systems. However, response characteristics such as higher-order tendon response, wave runup and air gap cannot still be accurately predicted using the available numerical tools. Wave basin model tests, therefore, are indispensable to designers for estimating responses that cannot be reliably predicted. At the same time, using model tests alone as an analysis tool is not practical due to large number of design cases typically defined in global performance analysis. It is necessary to verify and calibrate numerical tools using model test data prior to their application in global performance analysis. The paper (i.e. Phadke et al. (2006)) describes a methodology for calibrating and correlating predicted response from time-domain software tools against wave basin model tests. The application of correlation data in conjunction with predicted response to obtain various design quantities of interest was investigated. A simple technique for incorporating model test measured VIM response in time-domain analysis tools was also described. These methodologies have been used in the design of deepwater TLP systems in GoM and elsewhere. Stansberg et al. (2002), Fylling and Stansberg (2005) and Baarholm et al. (2006) addressed model tests for global design verification of floating production systems in depths beyond m, which cannot be made directly at reasonable scales. Truncation of mooring line and riser models, software calibration, as well as extrapolation and transformation to full depth and full scale are required. They discussed the important matters to be taken into account. The choice of proper procedures for the setup and the interpretation, and consistent and welldocumented methods, are essential. A case study with a deepwater semisubmersible was presented. In general, good agreement between results obtained from model tests and numerical calculations based on timedomain coupled analysis of the floater system responses was found. Figure 5. Monocolumn floater. 5. Design considerations for particular floating production systems 5.1. Spars A sequence of truss spars has been successfully installed in deepwater fields since late Compared with other floating systems, truss spars may offer significant advantages in motions, stability and project schedule. One of the unique aspects of a truss spar is that it exhibits both high-frequency and low-frequency motion responses. The high-frequency motions, or wave frequency motions, are peaked around the wave spectral energy, whereas the low-frequency motions correspond to the natural periods of the spar s

8 56 B.J. Leira rigid-body motions. Accurate structural design should include loads due to both wave and low-frequency motions. Resource-demanding time domain analyses were previously employed for design of the truss and a combination of time and frequency domain analyses was applied to design the other structural components. The procedure proved to be time consuming and inefficient, requiring extensive engineering hours. An enhanced hull design procedure is presented by Wang et al. (2002) and Luo et al. (2007), see Figure 6. The procedure is based on developing an integrated structural analysis methodology. Key features of the analysis methodology for both strength and fatigue analysis of the truss spar are discussed. Structural loads determined from the integrated methodology are compared with those from a complete time-domain analysis of the truss spar. The collision mechanisms of spar platforms have not been paid so much attention as that for ships in the past, and this type of collision accident has not been reported. However, this does not guarantee that such events will not occur in the future. Investigation of both the external mechanism and the internal mechanism for a ship colliding with a spar platform was made by Hu et al. (2007). A model test was designed to study the external mechanism. The collision scenario corresponds to a ship striking a spar platform which is moored in 1500 m water depth. Modelling the internal mechanism of the ship colliding with the spar platform is achieved by numerical simulation based on a nonlinear finite element analysis. A truss-spar is considered with a double hull structural design for the part of the hard tank which is located near the water surface. The crashworthiness of the double hull design is verified for the particular striking ship considered by means of the simulation results. The maximal tension forces of the mooring lines are also less than their breaking strength FPSOs A comprehensive list of classification rules, standards and guidelines relevant for design of FPSOs are given by Cocodia (2008). There seems to be two main design issues which are extensively addressed by present research activities. These are: (1) Efficient and accurate methods and procedures related to the ultimate limit state (ULS). This includes ensuring residual strength associated with damage, e.g. due to collision and (2) Fatigue capacity. In addition, the effect of corrosion on the hull strength is also addressed in a number of studies. Full-scale measurements obtained from the Triton FPSO (a turret moored FPSO in the central North Sea, see Figure 7) were reported by Lawford et al. (2008). It was considered that a full characterisation of the individual components of a sea-state is key to enabling the response of an offshore structure to be accurately calculated. The partitioning of a time series of directional wave spectra into wind-sea and swell components with distinct frequency and direction characteristics was addressed. The result of the partitioning and fitting analyses was a time series of wave parameters defining the wave spectrum for each component of the sea state. A 10-year site specific time series of directional wave spectra was partitioned in this way and used in the analysis of the specific FPSO. The representation of the directionality and magnitude of each environmental force acting simultaneously on the vessel, allowed the mooring and hydrodynamic analyses to be performed based on accurate input data. Huang and Moan (2005, 2006) presented new probabilistic models for still-water loads and the combined still-water and wave load effects of FPSOs. A procedure for determining load combination factors, which is suitable for semi-probabilistic and probabilistic design of FPSOs, was established. The most relevant load combination factors in harsh and benign conditions were derived. In general, combination factors depend on the parent distribution, time variation and relative magnitude of individual loads. It was shown that the extreme values of still-water and combined loads can be greatly overestimated if Figure 6. Global FE model for dry tow analysis. Figure 7. Triton FPSO.

9 The IES Journal Part A: Civil & Structural Engineering 57 operational control is not accounted for. On the other hand, the control is unlikely to be perfect. Hence, a partially truncated model is recommended to account for control of the still-water bending moments. Comparison of different methods for ultimate hull girder strength assessment was made by Sun and Wang et al. (2008). FPSOs have their own unique characteristics, including various operational requirements. In addition to that, the expectation of safety and economic aspects of FPSOs require an optimised structure to be designed. This calls for reliable structural assessment methodologies. One of the most important aspects of FPSO structural design and assessment is the hull girder ultimate strength. Numerical calculations of hull girder ultimate strength were presented based on six different FPSO designs by various methods. The results were analysed in terms of their differences, and conclusions were made based upon reliable methodologies for hull girder ultimate strength assessment of FPSOs. The methods applied were the incremental-iterative approach by Sun and Wang (2005), an in-house code HULLST based on Smith s method, and the application of idealised structural unit method (ISUM). All three methods show good agreement in terms of hull girder ultimate strength calculation for the selected FPSOs. In general, the predicted hull girder ultimate strength from HULLST and ISUM are almost identical for most of the cases, whereas Sun and Wang s method gives slightly conservative results in some cases. It is concluded that all three methods can be applied for hull girder ultimate strength calculation of FPSOs. Risk assessment of ship FPSO collision was considered by Pedersen (2002), Wang and Pedersen (2007) and Wang et al. (2003). In the latter paper, the research achievements in relation to ships collision and grounding since the 1990s were summarised. Issues specific to ship FPSO collisions that deserve further development were addressed. The paper also lists the conclusions and recommendations of 2006 V.1 Specialist Committee V.1 on collision and grounding that are relevant to ship FPSO collision risk assessment. The recommendations apply to five different areas. In relation to structural crashworthiness, the committee recommends refinement of analysis approaches. For assessment of probability of collisions, it is recommended that future research should focus on developing risk-based software. The calculated cost of risk reducing measures can then be compared with savings from calculated reductions in expenses. Regarding risk assessment, the committee recommends focusing on integrating predictive calculation tools, including the development of streamlined software/programs. In connection with rule and regulation development, it is found that future rules and regulations need to address design incident/accident scenarios, responses (of ships, offshore installations, bridges, etc.) to an incident/ accident, consequences and acceptance criteria. When it comes to predictive calculation approaches, it is recommended that topics that will further refine these methods include rupture strain and post-accident loads (both still-water and dynamic loads). Reliability assessment of an FPSO considering the effects of both corrosion and collision was performed by Zhang et al. (2006). As FPSO s have long intervals of docking for thorough inspection and maintenance, and are exposed to collision risk at sea, the timevariant reliability of FPSO becomes very important. The corrosion defect was represented as an exponential function of time. The idealised structural unit method was proposed to predict ultimate strength of the hull girder. Reliability of the intact hull during service was calculated to serve as a reference for that of the damaged and corroded hulls. The focus was on damage due to collision, and the damage was modelled according to American Bureau of Shipping (ABS) guidelines. According to these guidelines, the section with highest bending moment should be considered. The reliability of damaged hulls throughout the service life was obtained, which could be applied as a reference for further inspection and maintenance plans. It was found that the intact hull would have sufficient reliability during 25 years in operation. For the damaged hull, the permitted time in operation was reduced to around 8 years for the sagging condition and around 14 years for the hogging condition. On the basis of the obtained reliability estimates, decision related to inspection and maintenance can be made to update the reliability when a target level is defined. Turning next to the topic of methods for fatigue assessment of FPSOs, an extensive number of papers have been published during recent years. Some of them are briefly summarised in the following. Schmidt et al. (2008) evaluated the effect of random wave slamming for a stiffened panel which was located in the fairlead support structure of an FPSO (see Figure 8). A methodology for time-domain slamming prediction was developed. The structural analysis was then performed, considering all the shell plating, main stiffeners as well as other main supporting structures, by using a finite element structural analysis. The fatigue lifespan was estimated in a complete stochastic analysis, considering all possible sea states during the lifetime of the offshore structure as well as each associated probability of occurrence. For the particular case studied, the peak pressure for each discrete slam was assumed to generate one stress cycle and for each stress cycle the damage was estimated through the use of the S N fatigue approach.

10 58 B.J. Leira Figure 8. Fairlead support structure with applied loads. A summary of statistical models related to various parameters associated with calculation of fatigue damage was given by Wa stberg (2007), which was based on the previous report from the ISSC Committee on Fatigue and Fracture. Lotsberg (2006) reported on fatigue data related to fillet welds. The objective of the work is to develop a suitable methodology for the fatigue assessment of fillet welds relevant to FPSO details from the view point of the weld throat. The status on current design recommendations concerning the fatigue capacity of fillet welds was presented by Maddox (2006), based on a literature survey. In order to examine the validity of the recommendations and to supplement the fatigue test database, a test matrix with 33 specimens was developed. This included eight simple fillet welded cruciform joints that were subjected to axial loading and 25 fillet welded tubular specimens that were subjected to axial load and/or torsion for simulation of a combined stress condition in the weld. The test data were presented and also compared with design guidance from other relevant documents. Lotsberg (2008) addressed the calculation of fatigue damage at weld toes based on S N data when the principal stress direction is different from that of the normal direction to the weld toe. Such stress conditions are found at details in different types of plated structures. Some different fatigue criteria for these stress conditions are presented in design standards on fatigue design. Criteria used by the International Institute of Welding (IIW), Eurocode, British Standard and in the DNV standards were assessed against some relevant fatigue test data presented in the literature. An alternative equation for the calculation of an equivalent or effective stress range based on stress normal to the weld toe and shear stress at the weld toe was proposed. The methodology can be based on application of nominal S N curves and it can be used together with a hot spot stress S N curve with stresses read out from finite element analysis. The different design criteria were presented together with recommendations on analysis procedure. Fatigue of fillet welds based on structural stresses (SSs) was addressed by Fricke (2006) and Fricke and Doerk (2006). The papers address practical approaches for typical problems encountered. These approaches have been developed in the recent past in different research projects. They are based on an SS or a local nominal stress in the weld. Their application was demonstrated by several examples taken from ship and offshore structures using relatively coarse finite element meshes for the stress analyses. The SS approach offers a practical alternative to the very simple nominal stress approach and the refined notch stress and crack propagation approaches. It is particularly well-suited for special structural configurations, such as fillet-welded attachment ends and one-sided welds. The computation of the structural weld stresses is possible with rather coarse finite element meshes, which are also typically applied for the computation of the structural hot-spot stresses at weld toes. Maddox (2006) presented a critical review of current design methods, including their background and relevant experimental data, for assessing the fatigue performance of steel fillet welds with respect to failure in the weld throat. The main focus is on fillet or partial penetration welds in cruciform, T or lap joints under transverse loading failing by fatigue crack growth across the weld throat under normal stresses. The review covers the influence of residual stress, applied mean stress, joint fit-up and alignment, weld quality, the use of coated steel, the need for a plate thickness correction and calculation of the optimum weld size. Also considered are cases of fillet welds failing in shear or combined normal and shear stresses. Bergan et al. (2002), Bergan and Lotsberg (2004) and Lotsberg (2005, 2006) presented the new DNV recommended practice for fatigue analysis of offshore ships. This DNV-RP-C206 document is intended for fatigue design of floating production, storage and offloading units. The methodology can also be applied to other types of offshore ships. The experience with fatigue analysis of offshore ships and classification of such structures has demonstrated the need for improved guidance on this subject to designers, ship

11 The IES Journal Part A: Civil & Structural Engineering 59 yards as well as personnel involved in classification. This is in order to get a common understanding of what is required to achieve reliable structures with respect to fatigue. Rodriguez-Sanchez et al. (2005, 2007) addressed the application of controlled weld toe profiles as an option to extend the fatigue life of welded connections when existing tankers are converted in dry docks to serve like offshore ships (FPSOs and FSOs). It will be difficult to implement such fatigue improvement when these vessels are in service, because a converted ship is designed to be inspected, maintained and repaired in situ and not in dry dock. Codes recognise fatigue life extension by means of a controlled weld toe profile. Application of a controlled weld toe profile during conversion in selected areas previously identified by stress analysis of the hull structure can lead to extend the converted vessel fatigue life to comply with an expected field life. Rodriguez-Sanchez et al. (2005, 2007) applied a fracture mechanics approach for the assessment of controlled weld toe profiles for fatigue life extension purposes. A comparison of Stress Concentration Factors (SCFs) for a typical ship hull plate connection with and without weld toe profile control determined by FEA was presented. Then, results obtained from the FEA connection such as through plate stress distribution were used in a fracture mechanics analysis to compare the fatigue crack growth curve in as-welded conditions to that with the controlled weld toe profile. A comparison between spectral fatigue analysis (SFA) using both the surface extrapolation and Battelle structural stress methodologies, (is being implemented in Bureau Variants guidance documents), has been performed by Healy (2004) and Brian and Healy (2007). A side shell connection detail typical of a representative FPSO was considered. Fatigue damage at the toe along a number of weld lines is computed for a variety of surface extrapolation strategies and Battelle method options. The following conclusions were drawn from the study: (1) Implementation of the Battelle structural stress method (BSSM) in SFA for a representative FPSO is straightforward and no more involved or computationally intensive than for surface extrapolation methods. (2) When fatigue damage using the BSSM is computed at the weld end details examined, it is necessary to include the multi-axial in-plane shear contribution to avoid under-prediction. (3) When compared with the fatigue damage produced by the BSSM, the fatigue damage at the weld end details examined tends to be over-predicted by surface extrapolation methods. This is particularly true when extrapolating from h0.5t, 1.5ti. The best agreement at the weld end details comes from the surface stresses extracted at 0.5 t. (4) At the corner of the overlap weld, where the in-plane shear is not significant, there is good agreement between the fatigue damage produced by the BSSM and that produced by extrapolation from h0.4t, ti and extraction at 0.5 t. Application of the SS method developed by Battelle is further assessed by Kim et al. (2007) based on smallor mid-size scale specimens. Fatigue lives derived by class society rules and the SS method were compared. The simplified fatigue analysis methods by ABS and DNV were adopted to check the validity of the SS method. Before applying a complete set of loads based on the class society rules, a static loading case was applied as the class methods have their own correlation factor for wave loading. Simplified fatigue analysis was subsequently performed for more complete loading cases. From the computed results, it was concluded that the SS gives reasonable fatigue lives with respect to those based on hot spot stress or notch stress. Several structural details were subsequently considered, see Figure 9. Comparison was made with results from the simplified fatigue analysis procedures. From the computed fatigue lives, it is hard to conclude that the SS approach matches results from conventional fatigue life calculations well. It was hence recommended that additional structural details should be considered, and also that consideration of actual damage is made. Fatigue assessment was performed for the Agbami FPSO offshore Nigeria by Hwang et al. (2007). The FPSO is positioned with spread moorings at a water depth of about 1500 m for a service life of 20 years. As per the design requirement, the hull of the FPSO shall be designed to meet ABS s SFA notation and the seagoing condition. The seagoing condition is an additional requirement assuming that FPSO is a typical tanker navigating in the North Atlantic. From the results of the analyses it was found that the seagoing condition led consequentially to more critical fatigue damage for most of the hull components than the onsite condition which corresponded to Figure 9. Ship details of target hot spot.

12 60 B.J. Leira more moderate sea states. It was concluded from the analyses that the Agbami FPSO satisfied the fatigue design requirements for both the onsite and the seagoing conditions. Huang and Moan (2007) addressed the effect of mean stress on the fatigue life of welded joints in FPSOs. Such mean stresses are composed of residual stresses and stresses induced by external service loading conditions. These stresses affect the fatigue life of structural details. The initial condition of welding residual stress and its re-distribution by static preload and cyclic load in small scale specimens were evaluated by means of FE analyses and analytical equations. These were compared with test results obtained from measurements based on the sectioning method. Different fatigue analysis procedures to account for the mean stress effect were used to compare the fatigue test data of different specimens representing different typical welded connections in ship-shaped units. A new procedure which explicitly considers the effect of mean stress was also proposed. The effect of shake-down of residual stress is further investigated by Li et al. (2007). Structural members of FPSO hulls often undergo fairly large static loading before they enter service or variable amplitude cyclic loading when they are in service. The combined effect of both the applied stress and high initial residual stress is expected to cause shakedown of the residual stresses. They investigated some typical welded connections in ship-shaped structures by carrying out three-dimensional (3D) elastic-plastic finite element analyses. The effects of residual stress relaxation, initial residual stress and application of the load after variable amplitude cyclic loading were examined. A formula for predicting the residual stress at a hot spot quantitatively was proposed. On the basis of the formula, a new fatigue procedure was proposed. The procedure was subsequently validated against experimental results. There seems to be renewed interest in the application of concrete for hull construction driven in part by steel construction prices, the sizeable order books of shipyards worldwide, the growing dimensions of newly built FPSOs and in particular Liquified Natural Gas (LNG) FPSO concepts. The use of concrete materials for hull construction is not novel. In 1975, the Ardjuna FPSO was installed offshore Indonesia, based on a concrete barge. In addition to materials, new floater shapes are explored. In the Far East a series of cylindrically shaped FPS hulls are being built, with an oil storage capacity of 300,000 bbls. Application is envisaged in Brazil and in the UK sector of the North Sea. Lanquetin (2006) and Lanquetin et al. (2007) addressed the integrity management for the NKOSSA FPU, see Figure 10. This facility is the largest prestressed concrete FPU that has been installed, and has now been in operation for 10 years. A tailor-made methodology in order to ascertain the structural integrity of floating units was developed in The approach is based on continuous monitoring and analysis of the condition of the units. For the NKOSSA FPU, the methodology makes use of a fully nonlinear Finite Element Model of the concrete hull. After 10 years in operation without significant maintenance, the concrete FPU was found to remain in good shape Semisubmersibles and tension leg platforms Estefen and Estefen (2006) and Estefen et al. (2007) focused on the design of a new generation of semisubmersible platforms for oil and gas production offshore that is based on columns with square crosssectional area as shown in Figure 11. The platform Figure 10. Figure 11. platforms. NKOSSA concrete FPU. Hull structure of new generation of floating

13 The IES Journal Part A: Civil & Structural Engineering 61 column is based on an arrangement of stiffened flat panels having their ultimate strength characterised by buckling under in-plane compressive loading. Distortions induced by fabrication have considerable influence on the buckling behaviour and are discussed in order to provide design recommendations. They considered a segment of the column structural arrangement between robust transverse frames in order to analyse the failure behaviour of the stiffened panels. Numerical and experimental simulations were carried out for small scale isolated panels in order to perform a correlation study to adjust the numerical model for further use in more complex numerical simulations. The influence of different geometric imperfection distributions on the buckling behaviour of a column segment is also investigated. The magnitude of the initial geometric imperfections confirmed the influence of this parameter on the axial buckling load. However, the greatest influence on the ultimate strength and on the failure sequence of the plates is due to the initial imperfection mode. An initial distortion mode which coincides with the natural buckling mode of a particular plate generates a lower bound buckling load. On the other hand, some imperfection modes can counteract the buckling failure, hence generating upper bound values for the respective buckling loads. The rounded corners were the last regions to collapse in all analysed models, except for the model with the geometric imperfection mode coincident with the natural buckling mode. For this model the curved plates were the first to collapse, which had a significant influence on the failure sequence for the column. The same structural concept was also considered in relation to collision with supply vessels and the associated residual strength by Estefen and Estefen (2006) and Amante et al. (2008). Normally, damage from such events is in the form of local dents. They evaluated the ultimate buckling residual strength of a typical column after initiation of such a damage. The buckling analysis was validated using a finite element model considering geometric and material nonlinearities. Residual strength for the damaged column is compared with the ultimate strength of an equivalent intact column to estimate the safety margin after collision with a supply vessel. The uncertainty of the frequency and magnitude of the accidental loads associated with the possibility of column buckling demonstrate that analysis of accidental load effects needs to be thoroughly performed during the design stage. The results can be used as an indicator of the severity of offshore collision and provide insight to quantify lower bound safety factors to deal with such accidents, which are rather common. The possibility of optimised arrangements for the column members close to the splash zone could also be investigated further in order to prevent serious effects of collision in relation to the structural integrity. The structural redundancy of semisubmersible drilling vessels was considered by Chakrabarti and Maiti (2007) in order to withstand the loss of a slender bracing member without overall collapse of the structure, similar to requirements for fixed structures. Unlike static pushover type analysis which is applied to relatively dynamically insensitive fixed jacket structures, semisubmersibles require nonlinear dynamic redundancy analysis in the time domain to determine the safety against collapse due to environmental loading. A simple time domain nonlinear analysis procedure was suggested to capture the realistic behaviour of the structure under wave loading. In nonlinear structural analysis, first static loads are applied. Then wave load time history is applied for a few wave cycles in small increments. Results show that nonlinear analysis for one or two cycles usually can predict the safety against collapse. If the structure tolerates these cycles, it passes the redundancy test. If it does not, the structure has a deficiency that needs to be addressed. Rinehart and Buitrago (2006) addressed future TLP designs and the maximum depth in which tendons can be installed while still maintaining adequate and consistent reliability. The new strength guidance in API RP 2T recommends the axial tension hydrostatic collapse interaction equation currently used by API RP 2A load and resistance factor design (LRFD), coupled with a working stress design format with explicit tension and collapse safety factors. The latter factor controls the water depth applicability. To establish a basis for a hydrostatic collapse safety factor, an independent reliability study was performed in order to arrive at a safety factor consistent with the accuracy of the new interaction equation and current fabrication and design tendon practices. Cicilia (2004) and Cicilia et al. (2006) applied a LRFD criterion for the design of TLP tendons in their intact condition. The design criterion considers the ULS of any tendon section along its whole length. The partial safety factors are calibrated through a long-term reliability-based methodology for the storm environmental conditions, like hurricanes and winter storms, in deepwaters of the Campeche Bay, Mexico. When tendons are designed according to the developed LRFD criterion, a less scattered variation of reliability indexes is obtained for different tendon sections. (A target reliability index of 3.75 was applied, and a scatter of 52% was

14 62 B.J. Leira achieved). However, it was emphasised that the derived partial coefficients need to be adjusted taking into account a greater number of TLP models with variations in geometric and material characteristics, hull size, etc. Besides, updated environmental data and their joint probabilistic description need to be considered. Jayalekshmi et al. (2007) considered the hull-tether coupled dynamics that occur for TLPs in very deepwater. This effect cannot always be accurately studied through model tests due to depth limitations of the existing model testing facilities and possible scale effects (in the case of ultra-small scale models). An experimental methodology was presented for physically simulating such coupled behaviour for a single column TLP in a wave basin having dimensions 30 m 6 30 m 6 3 m. A combination of model tests and numerical analyses were carried out. The experimental results were presented for hull motions, tether displacements and dynamic tether tension. Results revealed that significant dynamic amplification could occur as a result of hull-tether coupling, thus highlighting the importance of coupled dynamics for deepwater compliant platforms. 6. Concluding remarks Recent challenges related to design of floating production systems have been reviewed. Environmental load effects due to hurricanes and also load effects created by arctic conditions were focused upon. The role of model tests in relation to verification and calibration of design tools was also discussed. Particular design issues related to the different types of production systems were addressed. It is emphasised that a review of the present type is very likely to be obsolete within a relatively short timeframe. This is due to the rapid developments which are observed in relation to new types of systems and structural concepts. Still, it is hoped that the present review article may serve to give a glimpse of relevant challenges to be considered for design of floating production systems at least in the near future. Acknowledgements ASME is greatly acknowledged for permission to publish the figures in the present article. References Amante, D.A., Trovoado, L., and Estefen, S.F., Residual strength assessment of semi-submersible platform column due to supply vessel collision. OMAE API, 2006a. Recommended Practice 95F, Interim guidance for Gulf of Mexico MODU mooring practice 2006 hurricane season. 1st ed. Washington, DC. API, 2006b. Guidelines for tie-downs on offshore production facilities for hurricane season. 1st ed. Washington, DC. API, 2006c. Recommended practice 95J Gulf of Mexico jackup operations for hurricane season interim recommendations. 1st ed. Washington, DC. API, 2007a. Interim guidance on hurricane conditions in the Gulf of Mexico. API Bulletin 2INT-MET. API, 2007b. Interim guidance on design of offshore structures for hurricane conditions. API Bulletin 2INT-DG. API, 2007c. Interim guidance on assessment of existing structures for hurricane conditions. API Bulletin 2INT- EX. API, 2007d. Recommended Practice 95F Gulf of Mexico MODU mooring practices for the 2007 hurricane season interim recommendation. 2nd ed. Atluri, S., et al., CFD simulation of truss spar vortexinduced motion. OMAE , Hamburg, Germany. Baarholm, R., et al., Model testing of ultra-deepwater floater systems: truncation and software verification methodology. OMAE Hamburg, Germany. Berek, E.P., et al., Development of revised Gulf of Mexico metocean conditions for reference by API Recommended Practices. OTC Paper 18903, Houston. Bergan, P.G., et al., Overview of phase II of the FPSO fatigue capacity JIP. OMAE , Oslo, Norway. Bergan, P.G. and Lotsberg, I., Advances in fatigue assessment of FPSOs. OMAE-FPSO , Houston, USA. Brian, E. and Healy, A., Comparison of the surface extrapolation and BATTELLE structural stress methodologies as applied to a spectral fatigue analysis of a representative FPSO structural detail. OMAE Chakrabarti, P. and Maiti, M.K., A simple time domain structural redundancy analysis procedure for semisubmersibles. OMAE , San Diego, USA. Chou, F., Ghosh, S., and Huang, K., Innovative concept of a drilling and/or production platform for operating in Arctic and deep water. OMAE , San Diego, USA. Cicilia, F.B., Reliability-based design criterion for a TLP tendon system. Thesis (PhD). COPPE-UFRJ, Brazil. Cicilia, F.B., de Lima, E.C.P., and Sagrilo, L.V.S., Reliability-based design criterion for TLP tendons. Proceedings of OMAE , Hamburg, Germany. Cocodia, E., International codes and standards applicable to cost estimating relationships for floating production systems. OMAE , Estoril, Portugal. Cueva, M., et al., Vortex-induced motion: model testing of a monocolumn floater. OMAE , Hamburg, Germany. Dankert, H. and Rosenthal, W., Ocean surface determination from X-band radar-image sequences. Journal of Geophysical Research, 109, C DNV-RP-C203, Fatigue strength analysis of offshore steel structures. DNV. Norway. DNV-RP-C206, Fatigue methodology for offshore ships. Estefen, T.P. and Estefen, S.F., Semisub column ultimate strength under compressive loading. In: World Maritime Technology Conference, 6 10 March. London. Estefen, T.P., Werneck, D.S., and Estefen, S.F., Influence of the geometric imperfection on the buckling behavior of floating platform column under axial load. OMAE , San Diego, USA.

15 The IES Journal Part A: Civil & Structural Engineering 63 Fonseca, N., Guedes Soares, C., and Pascoal, R., Global loads on a FPSO induced by a set of freak waves. OMAE , Halkidiki, Greece. Fonseca, N., Guedes Soares, C., and Pascoal, R., 2006a. Ship Structural Loads Induced by Abnormal Wave Conditions on a Containership. Journal of Marine Science and Technology, 11, Fonseca, N., Guedes Soares, C., and Pascoal, R., 2006b. Global structural loads induced by abnormal waves and design storms on a FPSO. OMAE , Hamburg, Germany. Fonseca, N., Pascoal, R., and Guedes Soares, C., Probability distributions of vertical bending moments on a FPSO in abnormal wave sea states. OMAE , San Diego, USA. Forristal, G.Z., Wave crest heights and deck damage in hurricanes Ivan, Katrina, and Rita. OTC Paper 18620, May, Houston. Fricke, W., Weld root fatigue assessment of filletwelded structures based on structural stresses. OMAE , Hamburg, Germany. Fricke, W. and Doerk, O., Simplified approach to fatigue strength assessment of fillet-welded attachment ends. International Journal of Fatigue, 28, Fylling, I. and Stansberg, C.T., Model testing of deepwater floating production systems: strategy for truncation of mooring and risers. DOT 2005, Vitoria, Brazil. Guedes Soares, C. and Cherneva, Z., Spectrogram analysis of the time frequency characteristics of ocean wind waves. Ocean Engineering, 32 (14 15), Guedes Soares, C., Cherneva, Z., and Antão, E.M., Characteristics of abnormal waves in North Sea storm sea states. Applied Ocean Research, 25 (6), Guedes Soares, C., Fonseca, N., and Pascoal, R., 2006b. Characteristics of abnormal waves and their induced global load effects in marine structures. Transactions on Society of Naval Architects and Marine Engineers, 114. Guedes Soares, C., et al., 2006a. Analysis of wave induced loads on a FPSO due to abnormal waves. Journal of Offshore Mechanics and Arctic Engineering, 128 (3), Halkyard, J., Atluri, S., and Sirnivas, S., Truss spar vortex induced motions: benchmarking of CFD and model tests. OMAE , Hamburg, Germany. Halkyard, J., Liagre, P., and Tahar, A., Full scale data comparison for the Horn Mountain Spar. OMAE , Vancouver, Canada. Halkyard, J., et al., Benchmarking of truss spar vortex induced motions derived from CFD with experiments. OMAE , Halkidiki, Greece. Healy, B., A case study comparison of surface extrapolation and Battelle structural stress fatigue methodologies. OMAE , Vancouver, Canada. Hu, Z.Q., et al., Research on collision mechanisms for a ship colliding with a spar platform. OMAE , San Diego, USA. Huang, W. and Moan, T., Combination of global stillwater and wave load effects for reliability-based design of floating production, storage and offloading (FPSO) vessels. Applied Ocean Research, 27, Huang, W. and Moan, T., Combination Factors Suitable for Semi-probabilistic Design of Floating Production, Storage and Offloading (FPSO) Vessels. In: OMAE , Hamburg, Germany. Huang, X.P. and Moan, T., Improved modelling of the effect of R-ratio on crack growth rate. International Journal of Fatigue, 29, Hwang, O.J., et al., Spectral fatigue assessment of Agbami FPSO Hull for onsite and seagoing as per ABS s SFA Guidance. OMAE Jayalekshmi, R., Idichandy, V.G., and Sundaravadivelu, R., Physical simulation of the hull-tether coupled dynamics of a deepwater TLP. OMAE , San Diego, USA. Kim, K.S., et al., Applicability evaluation of structural stress method on actual project. OMAE , San Diego, USA. Krogstad, H. and Barstow, S., Analysis and applications of second order models for maximum crest height. Journal of Offshore Mechanics and Arctic Engineering, 126, Lanquetin, B., Experience gained from floating units integrity management. OTC Paper 18146, 1 4 May, Houston. Lanquetin, B., Collet, P., and Esteve, J., Structural integrity management for a large pre-stressed concrete floating production unit. OMAE , San Diego, USA. Lawford, R., et al., Directional wave partitioning and its application to the structural analysis of an FPSO. OMAE , Estoril, Portugal. Legerstee, F., et al., Squall: nightmare for designers of deepwater West African mooring systems. OMAE , Hamburg, Germany. Lehner, S. and Rosenthal, W., Investigation of ship and platform accidents due to severe weather events results of the MAXWAVE project. OMAE , Hamburg, Germany. Li, L.B., Moan, T., and Zhang, B., Residual stress shakedown in typical weld joints and its effect on fatigue of FPSOs. OMAE , San Diego, USA. Lotsberg, I., Background for Revision of DNV- RPC203 Fatigue Analysis of Offshore Steel Structures. OMAE , Halkidiki, Greece. Lotsberg, I., Fatigue capacity of fillet welded connection subjected to axial and shear loading. OMAE , Hamburg, Germany. Lotsberg, I., Fatigue design criteria as function of the principal stress direction relative to the weld toe. OMAE , Estoril, Portugal. Lotsberg, I., et al., A new DNV recommended practice for fatigue analysis of offshore ships. OMAE , Hamburg, Germany. Luo, M.Y.H., et al., Integrated structural analysis methodology for truss spars. OMAE , San Diego, USA. Maddox, S.J., Status review on fatigue performance of fillet welds. OMAE , Hamburg, Germany. Mansour, A.M., et al., Tension leg platform survivability analysis. OMAE , Hamburg, Germany. Mitchell, M.S., et al., Real time data for structural integrity and stability assessment: a GoM case study, SPE SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA. Murray, J.J., et al., Two dry tree semisubmersible designs for ultra deep water post-katrina Gulf of Mexico. OMAE , Estoril, Portugal. Nyseth, H. and Holstmark, G., Analytical plastic capacity formulation for plates subject to ice loads and similar types of patch loadings. OMAE , Hamburg, Germany.

16 64 B.J. Leira Pedersen, P.T., Collision risk for fixed offshore structures close to high-density shipping lanes. Journal of Engineering for the Maritime Environmental, 216 (M1), Phadke, A.C., Mansour, A.M., and Huang, E.W., Methodology for application of model test data in TLP global design. OMAE , Hamburg, Germany. Rinehart, A.J. and Buitrago, J., A reliability assessment of collapse safety factors for tendon design. OMAE , Hamburg, Germany. Rodrı guez-sa nchez, J.E., Dover, W.D., and Brennan, F.P., Fracture mechanics analysis of fatigue crack repaired joints. Journal of Offshore Mechanics and Arctic Engineering, 127 (2), Rodrı guez-sa nchez, J.E., et al., Controlled weld toe profiles for fatigue life extension in FSO s and FPSO s. OMAE , San Diego, California. Schmidt, D., Bardanachvili, C.A., and Videiro, P.M., Full stochastic fatigue analysis of stiffened panels subjected to wave slamming. OMAE , Estoril, Portugal. Stansberg, C.T., Ormberg, H., and Øritsland, O., Challenges in deepwater experiments hybrid approach. ASME Journal of Offshore Mechanics and Arctic Engineering, 124, Sun, H.H. and Wang, X., Buckling and ultimate strength assessment of FPSO structures. Transactions on SNAME, 113, Tahar, A., et al., Full scale data comparison for the Horn Mountain Spar mooring line tensions during hurricane Isidore. Proceedings of OMAE 05, Halkidiki, Greece. Tahar, A., Halkyard, J., and Irani, M., Comparison of time and frequency domain analysis with full scale data for the Horn Mountain Spar during hurricane Isidore. OMAE , Hamburg, Germany. Timerman, G.J., et al., Coupled dynamic and static analysis of typhoon TLP accident during extreme environmental conditions. OMAE , Estoril, Portugal. Van Dijk, R. and Van den Boom, H., Full scale monitoring Marco Polo tension leg platform. OMAE San Diego, USA. Versowski, P.l., et al., Hurricane Impact Reviewed Through API. In: OTC Paper 18417, Houston. Wang, G. and Pedersen, P.T., A literature review of risk assessment of ship FPSO collisions. OMAE , San Diego, USA. Wang, G., Jiang, D.J., and Shin, Y., Consideration of collision and contact damage risks in FPSO structural designs. OTC Paper 15316, Houston. Wang, X., et al., Methodologies on hull girder ultimate strength assessment of FPSOs. OMAE Wang, J., Zhang, B., and Berg, S., Truss spar time domain in-place structural strength analysis. OTC Paper Ward, E.G. and Gebara, J.M., Assessment of storm sea fastenings for drilling and workover rigs on floating production systems during Hurricane Ivan. OTC 18324, May, Houston. Wa stberg, S., Fatigue and fracture of ship and offshore structures: summary of the fatigue and fracture committee report to the International Ship and Offshore Structures Congress (ISSC) OMAE , San Diego, USA. Wisch, D.J., Observations of hurricane impact on deepwater facilities. In: OTC Paper 18414, Houston. Wisch, D.J. and Ward, E.G., 2007a. Offshore standards the impact of hurricanes Ivan/Katrina/Rita. OMAE , San Diego, USA. Wisch, D.J. and Ward, E.G., 2007b. Legacy of the terrible triplets and standards of the future. OTC 18979, May, Houston. Xu, Q., Perryman, S., and Prislin, I., Assessment of the Horn Mountain spar in hurricane Ivan. OMAE , San Diego, USA. Yang, C.K., et al., The transient effect of tendon disconnection on the global motion of ETLP. OMAE , Estoril, Portugal. Zhang, D., Tang, W., and Zhang, S., Time-variant reliability assessment of FPSO considering corrosion and collision. OMAE , Hamburg, Germany.