Read ContentsFeb.indd text version

TECHNOLOGY FOCUS

Offshore Facilities

After many years of development work, as well as many unsuccessful attempts, the first liquefied-natural-gas (LNG) floating production, storage, and offloading (FPSO) initiative was launched in 1976 by Linde, Technigaz, et al. under the name Consortium 76--offshore LNG now is a reality. The first LNG receiving terminal installed on a gravity-based structure offshore Rovigo, Italy, will be commissioned soon. The first floating storage and regasification unit (FSRU) permanently moored in open sea offshore Livorno, Italy, is under construction, and several LNG carriers equipped with regasification facilities are able to deliver gas at several offshore ports in North America. Offshore liquefaction also seems close to materialization, with several participants (including oil companies; engineering, procurement, and construction contractors; ship owners; and others) developing LNG FPSO solutions for standalone gas fields or for associated-gas developments. Two main ways to approach these facilities seem to be pursued by the various participants. The "big way" is followed mainly by major oil companies: It aims at developing large-capacity liquefaction units (3×106 tonnes/a and larger) by use of processes generally derived from latest-generation onshore liquefaction cycles. A second trend, which could be called the "small way," targets smaller capacities, by use of less-efficient processes often based on nitrogen cycles, but having better suitability for use on board a floater. While the latter probably has a wider range of potential application cases and would ease topside layout and safety concerns for a lower overall capital expenditure, the former will provide higher liquefaction efficiency and will target economies of scale. In turn, it might be applicable to only a limited number of relatively large gasfield developments. Today, both approaches have equal chances to open the way for offshore floating liquefaction, but, certainly, both will need to rely on experienced companies that can gather and leverage experience in offshore regasification gained from the first FSRUs, combined with experience in large oil FPSOs and in onshore liquefaction. All the industry needs now is a good gas field!

Offshore Facilities additional reading available at OnePetro: www.onepetro.org OTC 19429 · "A Coupled FE-SPH Approach for Simulation of Structural Response to Extreme Wave and Green Water Loading" by J.C. Campbell, Cranfield University, et al. OTC 19239 · "Innovative Pipe System for Offshore LNG Transfer" by Ing. C. Frohne, Nexans Deutschland Industries, et al. OTC 19315 · "Deepwater Moorings With High-Stiffness Polyester and PENFiber Ropes" by P. Davies, IFREMER Brest, et al.

JPT

Claude Valenchon, SPE, is Manager, Offshore Technology Development, for Saipem in Paris. In 1981, after 7 years with CG Doris, he joined Bouygues Offshore, which became Saipem S.A. in 2002. Currently, Valenchon is in charge of offshore technology developments, aiming at providing solutions, products, or concepts for tenders and design competitions, with current focus in the area of deepwater-field developments, Arctic, and offshore LNG production. He serves on the JPT Editorial Committee and holds an engineering degree from the École Nationale des Ponts et Chaussées in Paris.

74

JPT · FEBRUARY 2009

OFFSHORE FACILITIES

Design of Steel Catenary Risers for Deepwater Offshore Brazil

Petrobras has been investigating the steel-catenary-riser (SCR) alternative since the beginning of the 1990s. Since then, fatigue verification has been an important issue, demanding good representation of the loading conditions that occur during the lifetime of the riser. The concern with fatigue has motivated Petrobras to research several areas, such as metocean data acquisition, hull design for motion optimization, special touch-down-point (TDP) joints, accurate models for vortexinduced-vibration (VIV) analysis, and the corrosion-fatigue effect.

Introduction Installation of the P-18 SCR was a pioneer project of a free-hanging SCR connected to a semisubmersible, and it proved the technical feasibility of the concept. Although this riser was installed as a prototype, it is still working in the gas transfer from platform P-18 to platform P-26. It has been monitored since 1999, and the results are being compared with the design data and with simulations performed with in-house computer programs and other commercial packages that include the complete design methodology. Other SCRs were studied, such as the 12-in. oil-export riser for the P-19

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper OTC 19249, "Influence of Fatigue Issues on the Design of SCRs for Deepwater Offshore Brazil," by A.L.F.L. Torres, M.M. Mourelle, S.F. Senra, E.C. Gonzalez, and J.M.T. da Gama Lima, Petrobras S.A., originally prepared for the 2008 Offshore Technology Conference, Houston, 5­8 May. The paper has not been peer reviewed. Copyright 2008 Offshore Technology Conference. Reproduced by permission.

semisubmersible in 770 m of water and the 10-in. oil- and gas-export lines for the P-36 semisubmersible at a water depth of 1360 m. The free-hanging SCR configuration is considered as an available technology for semisubmersible applications, and there is interest in the application of SCRs connected to floating production, storage, and offloading units (FPSOs) because of the trend to use these units for exploration and production in deep water. This has caused a need to study this concept carefully, given the high offsets and heave motions imposed by the vessel at the top of the riser. Fatigue verification is an important issue that requires accurate evaluation of the loading conditions that occur during the riser lifetime, and it also requires a precise knowledge of construction aspects that could decrease or change riser-materials resistance. Wave-Induced Fatigue Over the past few decades, Petrobras has acquired Campos basin wave, current, and wind data, resulting in a metocean database containing more than 7,000 records. Within these data, the occurrence of multimodal/multidirectional sea states was identified. To use this database in riser design, the in-house software tools for structural fatigue analysis were upgraded to consider bimodal/bidirectional sea states. Because fatigue verification is an important issue in steel-riser design, a good representation of loading conditions that occur during the riser lifetime is needed and use of the entire database is recommended. However, the riser design schedule can be affected if a random time-domain analysis is used. To minimize this, one solution adopted was to develop a statistical procedure to reduce the database to a reasonable number of representative loading cases

to be used in fatigue-damage verification. This method resulted in the adoption of approximately 150 fatigueloading cases. The combined wave, current, and wind data are preserved in the loading conditions that were chosen to represent all the usual metocean situations in Campos basin. Hull Design To Reduce Motion. In the Campos basin, the wave fatigue environment has been shown to be the limiting factor for SCR feasibility. The alternative of optimizing platform motions has been one of the ways to increase the possibility of SCR applications. Among dynamic motions, heave has been identified as the most damaging. For the P-52 design, in 1800 m of water in the Roncador field, a limit for a maximum heave at an extreme point in the hull has been established, and these data determined the choice between existent hull models. When starting a new hull design for the P-55 unit, more-complete criteria was used, with a set of operational waves chosen from the traditionally most damaging ones. The P-52 motions were taken as a reference. There was interest in knowing what level of minimized motions could be obtained considering a deep-draft-hull concept. A study was conducted with a large number of hull geometries, and the most adequate ones, in terms of constructability and other naval aspects, were chosen. The amplitudes of heave motions for these hull models were verified to be from 15 to 20% smaller than those obtained for the P-52 hull. On the basis of this, a set of maximum motions under medium- and high-fatigue wave conditions has been established. The monohull concept also has been designed to achieve minimized heave motions. The same criterion has been applied as a guide to the models stud-

The full-length paper is available for purchase at OnePetro: www.onepetro.org.

JPT · FEBRUARY 2009 75

100% 90% 80% 70%

Infield and Export Risers--D-Class Weld at Outer Diameter

Damage, %

60% 50% 40% 30% 20% 10% 0% Production A VIV long term Heave-induced VIV (100-year wave) * Production B Water Injection VIV short term 1st- and 2nd-order damage Gas Export Oil Export A Heave-induced VIV (1-year wave) * * Relatively low damage values Oil Export B

Fig. 1--Fatigue-damage contribution at the TDP.

ied for the monobore design, with and without storage capacity. Coupled Models. The use of coupled analysis tools in the design of SCRs becomes even more important because a large number of risers and mooring lines are connected to the platform and the system is in deep water like the P-18, P-52, and P-55 systems. Petrobras has developed in-house software programs to analyze the coupling between the nonlinear hydrodynamic behavior of the hull and the structural and hydrodynamic behavior of the lines. Hybrid methods that combine the use of programs based on coupled and uncoupled formulations have been considered the possible "roadmap" toward a fully coupled analysis and design methodology. The coupled analysis carried out with the hybrid model is attractive in contrast to the excessive computer cost of the fully coupled method because a large number of analyses for calculation of the fatigue behavior on risers are necessary with it. In recent studies of SCR design for the P-55, the numerical model of the system was generated in three different coupled programs, and the results were compared with the empirical data from the model tests in terms of platform motions and line tensions. Calibrations

could be adjusted to obtain a more reliable numerical model of the entire system. Some investigations have been conducted related to riser fatigue-damage response, comparing the use of coupled and uncoupled methods. Another issue identified in the model test of the P-55 was vortex-induced motions (VIMs) in this deep-draft semisubmersible platform. The VIMs can result in additional oscillations in riser and mooring-line tension, as well as additional fatigue loading of the riser TDP. These effects are under investigation through towing tests and computational-fluid-dynamics (CFD) calculations to verify if it is necessary to use some mitigation device on the hull. Frequency-Domain Approach. A nonlinear random time-domain analysis has been adopted in fatigue-analysis verification because model nonlinearities are modeled properly and the environmentalloadings random behavior is considered. The disadvantage is the high computer time required. Because fatigue-damage calculation depends on stress variations during the lifetime of the structure, the set of loads used in the analysis should be sufficiently complete to represent all possible situations. Because Petrobras uses its own measured environmental

database that contains a large number of data points, the use of time-domain analysis may affect the design schedule. Another solution, besides the statistical treatment procedure to reduce the database, was the development of a frequency-domain methodology, based on linearization techniques, that was implemented in in-house software to be used as an alternative tool for the initial phase of riser design. In general, results indicated agreement between frequency- and time-domain approaches in identification of criticaljoint and critical-loading cases. In terms of fatigue-damage calculation, frequency-domain analysis when compared to time-domain analysis furnished better results for the lazy-wave SCR configuration. For the free-hanging configuration, larger differences were found and the frequency-domain approach tends to be more conservative. Research to evaluate the soil-structure interaction, aiming at representing the TDP variation that is significant in the free-hanging configuration is ongoing. VIV Fatigue Campos Basin Currents. Campos basin current profiles for deep water are composed of two layers coming from different sources. These layers have different

78

JPT · FEBRUARY 2009

directions. The first layer covers the depths from sea level to approximately 300 to 400 m, going predominantly to the south and southwest directions. Below this level, another layer becomes dominant going to the north and northeast direction. From the first VIV calculations, as a function of the 2D characteristic of the software, the current velocities along the depth were projected to the riser inplane and out-of-plane directions. This was supposed to capture the characteristic of directionality of the Campos basin current profiles. Short-Term Response. The idea of considering a short-term response came from the necessity of predicting the riser response when facing a 100-year current event during its operational life. No extreme stresses were expected, but it was necessary to know the magnitude of the induced fatigue damage. The approach used assumes that the damage from the worst-possible extreme event will be resisted by the riser. The extreme events, however, are not very well characterized in terms of duration and the way the phenomenon evolves. In recent applications, the short-term damage represented a significant percentage of the total damage, as the case for the P-55 design for an 1800-m waterdepth application for the Roncador field, shown in Fig. 1. Alternative CFD Model. As an alternative to VIV traditional-model use, the initiative was to incorporate a CFD procedure into the in-house riser-analysis package. The discrete-vortex method has been implemented and is being tested and compared to the traditional-model results for some real applications. The method brings the possibility of using the current profiles with their directional characteristics, and results obtained so far indicate some less conservative results. Materials The first SCR applications developed by Petrobras were related to import and export lines. When the use of SCRs started to be planned for production lines, as in the P-52 project, the problem of how to face the highly corrosive environment turned out to be a major one. The presence of carbon dioxide (CO2) and/or hydrogen sulfide (H2S) in the produced stream created doubt about the applicability of the S-N curves used.

P-18 SCR The P-18 SCR is the only operating SCR currently in the Campos basin. At the time it was installed in 1998, the riser was the first SCR to be installed on a semisubmersible unit. A complete monitoring system was installed, and the measurement campaign lasted for approximately 2.5 years. Many issues were investigated, and today the generated database still is being used in studies. The confirmation of the expected riser behavior, characterization of platform motions, and identification of critical current profiles for VIV response have been some of the results obtained. The measured data at strain gauges at the riser top, associated with the measured flex-joint-angle variations, are being used in the reassessment of flex-joint fatigue life. The data, besides including real values, include VIVinduced axial vibrations that were not included in the design phase of the flex joint, and today it is the main reason for revisional work regarding the top connection system. The riser was installed without any suppressor device for VIV. The updated method regarding VIV and wave fatigue is being applied to define what will be considered as the riser updated-design fatigue life. The TDP is the main focus because it is the region that suffers damage from platform-induced wave motions and also from VIV. The reassessment plan includes the generation of an updated engineering critical assessment and inspection of welds by two methods, one performed externally and another by umbilical pig. The accuracy of the field inspection may not correspond to ideal values; neither may have the same level of accuracy obtained during riser construction, but both will give important reference values that will support the riser-integrity evaluation. A permanent monitoring system has been designed for the top section of the P-18 SCR that, once field tested, will be considered as a model for other SCRs to be installed in the future. General Comments When planning for a platform with a large number of SCRs, the interference between the adjacent risers can become an issue. Besides working on the distance between supports and in the difference between the azimuths of the

risers, frequently it is necessary to consider a difference in top angles between neighboring risers. As a function of this, the necessity of using top-angle values of 15 to 17° for some risers, in the vicinity of other risers with 20° of top angle, became the usual practice during design. For reduced-heave-motion units, such as the deep-draft unit P-55, the adoption of 17 or even 15° did not cause problems. Conclusion From the last results obtained with Petrobras design methodology and updated data, it is impossible not to consider the option of using VIV suppressors for an SCR. The question is the relative length to be used, but when analyzing total length variation of strakes, their VIV efficiency, and their location along the SCR, the results obtained through traditional modeling do not present monotonic results, thus making it difficult to make decisions about the length of the strakes. The successive evolution in the set of design currents applied for longterm and short-term response calculations has caused the design to be more robust and realistic, but up to now, always increasing the effect of VIV on the overall riser design. The incorporation of CFD procedures, which can keep the Campos basin current profiles directionality characteristic, is a promising way to obtain less conservative results. Up to now, the design for production risers subjected to corrosive fluids has led to clad sections, corresponding to less than 20% of total riser length. This trend represents specific conditions of the Roncador field. More-severe situations may occur that can force the use of longer clad sections. Application of clad pipes in the critical regions requires a better understanding of the weld behavior, geometrical imperfections, and nondestructive-test results. Monitoring new risers to be installed is a key point that can support the evolution of design methodology for VIV and also with respect to waveinduced platform motions. The monitoring system can be planned to be composed of some equipment on the top section that will operate during the entire riser lifetime, and another set of equipment that will be used for a limited duration for evaluation of the JPT design methodology.

JPT · FEBRUARY 2009

79

OFFSHORE FACILITIES

Management of Offshore-Ice Operations

The use of icebreakers in support of offshore-ice operations, and specifically their efficiency in support of vessel-shaped floating platforms in ice, is discussed. New-technology icebreakers equipped with azimuth thrusters achieve high levels of operability with various levels of ice management in a wide range of effective ice thickness.

Introduction The oil industry has increased its interest in ice-covered waters. Operations, especially in water deeper than 100 m, use various vessels for drilling or production, with icebreakers supporting their station keeping. The use of icebreakers enables stationary operations in ice to continue with increasing degrees of difficulty. The use of vessels in such offshore-ice operations is substantially different from and more demanding than traditional ice-transit operations or port operations in icecovered ports and terminals. Offshore-ice projects, which operate in moving pack ice, must ensure that station-keeping operations continue with a high degree of confidence that the station-keeping limits of the floating platform will not be exceeded. Rigorous risk control also is needed to ensure that, if needed, all operations can be stopped and the platform can be removed safely from the location. A high confidence required in the capaThis article, written by Senior Technology Editor Dennis Denney, contains highlights of paper OTC 19275, "Ice Management for Ice Offshore Operations," by A.J. Keinonen, AKAC Inc., prepared for the 2008 Offshore Technology Conference, Houston, 5­8 May. The paper has not been peer reviewed. Copyright 2008 Offshore Technology Conference. Reproduced by permission.

Fig. 1--Pacific Endeavor clearing ice with the wake of its propulsion in low ice pressure.

bility to stay on location before such an operation is justifiable. The overall scope of ice management is more accurately "risk management for offshoreice operations." The following are an integral part of the ice management. · Ice and environment intelligence · Ice and environment forecasting · Defining operability and safe ice operational envelope for the stationary vessel · Operational-risk evaluation and assessment · Alerting of operations Only after the above are completed can the actual physical ice management (i.e., breaking and clearing ice) take place. Safe ice-management logic applies to all floating platforms and to any platforms intended to be removable from their stationary operational locations. This paper focuses on the most demanding version of the ice management, operations of ship-shaped platforms. Vessel-shaped platforms are unique in that they offer a highly favorable ice-interaction condition when ice

moves in the direction of the vessel. However, the challenge is that if ice comes from the side, the ice loads are so much higher that the benefit of the low loads in one direction is totally lost and operation might not be feasible. Azimuth-Icebreaker Technology The most important development in physical ice-management technology since traditionally propelled icebreakers is azimuth thrusters. The wake of azimuth thrusters can be more powerful in terms of breaking ice than the hull of an icebreaker. Also, the azimuth-thruster wake can clear the ice in a highly efficient manner--e.g., from ports, terminals, and the paths of escorted/supported vessels or various offshore-ice platforms (both bottom-founded and floating). It is possible to place an icebreaker accurately, even in moving ice, at a stationary location or to move it in a highly controlled fashion to locations where ice management is required at any time. Figs. 1 and 2, respectively, show the efficiency of azimuth thrusters

The full-length paper is available for purchase at OnePetro: www.onepetro.org.

80 JPT · FEBRUARY 2009

Fig. 2--Pacific Endeavor clearing ice with the wake of its propulsion in 9/10 original ice cover.

in clearing ice in low ice pressure and in clearing ice in the presence of more than 9/10 (very-close-pack) ice. Design Integration A highly competent ice-management system can keep a platform on location by use of a highly aggressive and active approach for time periods corresponding to exploration drilling in ice. However, in this case, vulnerabilities to critical ice interactions most likely will be acceptable or even optimal to the platform. Designing for long-term low-probability events typically does not make sense for exploration platforms. Accepting even a drastically increased risk of downtime might be the best option. In the case of production, the criteria change and it is essential that the whole system be designed to remove significant vulnerabilities risking frequent downtime. The biggest risk may be that real-life operations might introduce unacceptable ice interaction, which could lead to frequent unplanned downtime. Station Keeping in Ice There are several types of ship-shaped platform experiences in ice. · A moored drillship with limited station-keeping capability and no possibility to vane results in high ice loads when pushing ice to the side as ice comes from the bow and when pushing ice down, depending on the bow angles. However, with ice coming from the side, the ice accumulates and pushes the keel part of the ice accumulation down. Severe ice can get into the mooring lines when ice is either pushed down or significantly accumulated at the platform. This type of platform relies on ice management to

break most of the severe ice even though it may be able to handle some lighter ice by itself. Returning to location in ice requires significant ice-management and anchor-handling support. · A dynamically positioned construction vessel can have low station-keeping capability and very high inherent icebreaking resistance because of the vertical stem of the vessel and near-verticalsided wedge-shaped bow. It has a very good ice-clearing performance to the sides, and keeps ice from going under the vessel. It has no protection against ice getting into its moonpools, but is surprisingly efficient at keeping ice away from them, despite its shallow draft. It has excellent ice-vaning capability. In ice-pressure situations, ice is pushed down and into the moonpools. This vessel relies totally on ice management to break the ice into small enough pieces to enable station keeping. In ice pressure, the vessel is able to stay on location only if the pressure and resulting ice compacting can be cleared to ensure that no ice enters the moonpools. Returning to location is easy because it is self-propelled. However, getting out of the severe ice that has stopped operations requires significant icebreaker-escort support, if it is not possible to let the vessel simply drift with the general ice drift. · A manually positioned icebreaker with relatively low station-keeping capability has a dynamic-positioning system that has proved to be unworkable in severe ice. It has relatively low inherent ice loads from the bow and has limited vaning capability because of the long parallel sides of the vessel when using only its main thrusters. The mediumicebreaking bow partially pushes ice

down and to the sides. A custom protection skirt was built into the moonpool of the Vidar Viking to prevent ice from getting into it. The Vidar Viking had to rely heavily on ice management and a large operational radius available of 50 m, to be able to vane into ice drift. Also, it had to rely on ice management to keep ice floes small enough to enable staying on location without exceeding allowable radius of offset. Returning to location in ice was relatively easy. · A floating storage and offloading (FSO) vessel connected to a singleanchor-leg-mooring (SALM) buoy has highly limited ice tolerance and medium ice-load capability with vaning capability but no ice-drift-reversal capability and has inherently very high ice loads on the basis of the shape of the FSO. There was limited protection against ice interfering with the oil hose exiting the SALM. Deep access of the oil hose to the FSO provided good protection from ice. The system relies totally on ice management in the presence of any ice to ensure small ice-floe size, giving low ice loads, and ensure prevention of ice interfering with the oil hose from the SALM. Returning to location and all aspects of operation in the presence of ice require continuous and significant ice-management support, including ice clearing. · A four-point-moored icebreaker with highly limited vaning capability has floating stern lines, low stationkeeping capability, and inherently medium-to-high ice loads. There is no protection against ice getting into the moonpool, and ice gets into it easily. The operation had to accept ice-drift reversals and relied heavily on ice management. Returning to location required competent ice-management support including ice clearing. The main lesson from these experiences was that it is a high priority to design a system specifically for each geographic region and its custom application. This process must take into account the following key factors. · Ice load in the unidirectional ice drift · Ability to respond to rapid changes in the direction of ice drift · Ability to respond fast enough to ice-drift loops, quick turns, and drift reversals · Confidence in full ice clearing, not allowing ice to get into the risers, moorings, and other structures

82

JPT · FEBRUARY 2009

· Returning to location in the presence of ice · Training operators Ice Clearing and Dismantling The main innovations of the use of azimuth thrusters on icebreakers in terms of ice-management work are as follows. · Ice can be broken by the wake of thrusters, which can be even more efficient than breaking ice with the hull of the vessel. · Ice can be cleared in a highly effective manner by use of the wake of azimuth thrusters. · The icebreaker can remain stationary in moving ice while managing ice or can move in any direction while doing so. · The wake of the thrusters can dismantle large first-year ridges by blowing away their keels, causing collapse from lack of buoyancy. The ability of the thrusters to break ice was tested on several occasions, and a model was developed. The azimuth-thruster icebreaker can prepare a wide channel in unbroken level ice and a significantly wider channel in

an area where ice has been prebroken by a primary-ice-management vessel. Because azimuth thrusters can clear ice efficiently with the wake of their propulsion, they offer powerful opportunities for managing and clearing ice from any location. For offshore-ice applications, this ability is a major improvement in terms of managing the exact ice that needs to be managed. Orienting two azimuth thrusters to act opposite to one another will keep the vessel stationary even in moving ice. Orienting them slightly forward or aft, will move the vessel slowly longitudinally; and by use of well-known techniques of operating a twin-azimuththruster-equipped vessel, it also can be moved sideways. Integration of Ice Management Into Design A powerful ice-management system can keep a ship-shaped platform, or a conical one, on location under surprisingly high ice forces, even with the use of traditional icebreakers. Adding the major improvements available to

the ice-management systems, it could be claimed that almost any platform could be kept on location in moving ice, provided a sufficiently powerful ice-management system is used. Even though this is not advisable, it shows that there should not be a specific need to develop very high station-keeping capability nor should the platform need to be designed to handle a major quantity of ice independently. It may be more advantageous to have the platform not break ice very efficiently; possibly, clearing the ice to the sides is more important in certain situations. There clearly is no single optimal solution to a general ship-shaped platform operating in ice or to an icemanagement system to support such a platform. However, the choices available for a designer are from the whole operational philosophy and approach to handling risk and include the supporting icebreaker and azimuth technology. So far, each project has been a pioneering one, with much work remaining until some level of routine development JPT and solutions is established.

OFFSHORE FACILITIES

Lessons Learned From 12 Years of Operations of a Prestressed High-Performance-Concrete Floating Production Unit

Lessons learned from use of a floatingproduction unit (FPU), made of prestressed high-performance concrete, in operation for Total E&P Congo are presented. The unit has been in use for 12 years on the N'Kossa oil field in 170-m water depth. The focus is on structuralmodeling techniques, aging processes, and development of an inspection program. The paper is not intended to make a recommendation between steel and concrete, which would entail many other considerations.

Introduction The FPU NKP is 220 m long, 46 m wide, and 16 m high, with a displacement of 107 000 tonnes. It contains 27 000 m3 of concrete, 2350 tonnes of prestressed steel, and 5000 tonnes of passive steel. During its 12 years of operation, the FPU has undergone one technical stop for process maintenance, as scheduled in the design; otherwise, it has been in uninterrupted service. The unit, shown in Fig. 1, was built in southern France in 1994­95 and installed 1 year later in the N'Kossa oil field in 170-m water depth, 60 km off the Congo coast.

This article, written by Senior Technology Editor Dennis Denney, contains highlights of paper IPTC 12546, "Lessons Learned From 12 Years of Operations of a Huge Floating-Production Unit Made of Prestressed High-Performance Concrete," by Bertrand Lanquetin, Heidi Dendani, and Pascal Collet, Total, and Jose Esteve, Bureau Veritas, prepared for the 2008 International Petroleum Technology Conference, Kuala Lumpur, 3­5 December. The paper has not been peer reviewed. Copyright 2008 International Petroleum Technology Conference. Reproduced by permission.

Fig. 1--NKP FPU.

The production facilities and living quarters for 160 people are fitted on the 10 000-m2 deck, which for construction purposes was subdivided into six modules: accommodation and central control, utilities, electric-power generation, gas compression for re-injection, crude oil, and gas. Design production is 16 000 tonne/d of oil and 1300 tonne/d of liquefied petroleum gas. The unit is held in place, 70 m away from the NKF2 platform, with a spread-moored configuration by means of 12 mooring lines. Asset-Integrity Management A special method was developed to analyze and monitor the condition of the units. The aim of the floating-units integrity-management process is to ensure management and continuous follow-up of floating units from safety, environmental, operational, maintenance, and quality-management viewpoints. It includes recommendations on

inspection, maintenance, and repairs. This process calls for the following. · Structural and anchoring modeling and analysis · Qualitative [risk-based inspection (RBI)] · Yearly reviews of the inspection, repair, and maintenance (IRM) plan · Data management and storage (including reports) · Assistance for emergency response · Framework for analysis The program is divided into four complementary, interacting modules, as shown on Fig. 2. · Structural nonlinear finite-elementanalysis (FEA) model and dynamicmooring model · IRM plan and schedule incorporating class requirements (e.g., renewal of certificates and repairs), and incorporating the risk-based inspection · Database (e.g., plans, results of models, inspection reports, and class

The full-length paper is available for purchase at OnePetro: www.onepetro.org.

84 JPT · FEBRUARY 2009

Periodical (re)assessment

FEA, anchoring models and management tool

RBI analysis and IRM cycle

Data management and reporting

Emergency response

= interaction

Fig. 2--Floating-unit integrity-management modules.

other uses. The most important comes in through Space T9 with a pipe (metallic outside and concrete inside) that penetrates the side shell and runs through two compartments before reaching the piscine, an enclosed basin within Compartment C10. Three pumps (two in service, one spare) then move the water from the piscine to the process plant for cooling purposes. Fig. 3 shows that the hull's selfsupporting structure is made of longitudinal and transversal walls (called bulkheads) that also provide the internal subdivision. They are made of reinforced concrete through which longitudinal, vertical, and transversal tendons (prestressed cables) extend, providing compression in the shell-plane directions. Depending on location, wall thickness varies from 40 to 80 cm. Concrete Prestressing. Concretestructure prestressing consists of applying a compressive load to enhance strength. The main aim is to maintain the concrete in compression under the forecasted external loads. Here, tendons made of several ultrahigh-tensilesteel strands going through metallic conduits set within the concrete section were used. The path of these ducts was defined carefully at the design stage to solve practical constraints (e.g., access holes and equipment foundations) while maintaining the required compressive load. To protect the steel tendons and provide a solid-concrete section, the conduits are filled with injection grout. To ensure appropriate load distribution and shear capacity, passive reinforcement steel is embedded in the concrete. These bars are the same as those typically used in concrete beams in any building, as shown in Fig. 4, a photo taken during construction of the NKP barge. FEA Model A numerical model was built to assess actual and future conditions. The dynamic loads (motions and sea pressures) were assessed with 3D diffractionradiation-analysis software taking into account the latest metocean data from the site. An interface was developed to transfer the hydrodynamic loads directly to the structural model. The model was built with the following constraints. · Reinforced concrete was considered to be a homogeneous material

status), with information shared on a network system · Emergency-response service In alignment with the integrity-management program, the unit was placed within a classification scope. Surveys and maintenance actions required by the Classification Society are introduced and accounted for within the system. Falling within the classification scope of work are the mooring, hull and marine systems, accommodation quarters, and the helideck structures and topside connection to deck. In some cases, risers and subsea equipment also are included. Prestressed-Concrete FPU The unit was constructed with 26 lateral compartments (B and T compart-

ments) and 13 central compartments (C compartments). The lateral compartments can be used as ballast tanks, but only the four tanks in the corners are used on site to maintain trim and pitch. The central compartments are void spaces. Running through the central void spaces as a spine is the "technical gallery" that connects the aft and fore pump rooms and provides access to the internal compartments. The nose on the fore end is a concrete cantilever that supports the flare tower, as far away as possible from the accommodation quarters. The shell is not a fully watertight continuous skin. It is pierced in several locations for water intake for processplant cooling, ballast, fire-extinction means, freshwater production, and

Fig. 3--Compartment plan and starboard profile.

JPT · FEBRUARY 2009

85

Fig. 4--NKP under construction.

with yield-capacity values according to tests performed at construction time. · Prestressed cables were modeled explicitly. The modeled cable tension accounted for the variation of the prestressing load along each cable caused by friction and losses caused by anchorage penetration. · Concrete-creep and -shrinkage properties were considered bias coefficients applied directly on the tendon loads. · Topside loads were introduced as concentrated masses placed at each topside-module center of gravity and connected to its supporting stools at the deck through rigid connections. Operational Experience Inspection and Maintenance Plan. On the basis of drawings review, survey reports, and FEA results, an inspection campaign was set up. It includes the classification requirements and additional tasks to maximize the unit efficiency. The main objectives are · Identify any defect and its deterioration process. · Chemical attack · Corrosion · Crack · Coating deterioration · Accident · Define the severity of damage · Provide recommendations for repair · Provide an image of the condition of the unit to be compared in future campaigns

Means of Survey. Depending on location and inspection time, different means of survey are to be used. · Global-visual inspection (GVI), or overall survey. Intended to report on the overall condition of the hull structure and determine the extent of additional close-up surveys. · Close-visual inspection (CVI), or close-up survey. Details of structural components are inspected at close visual range (i.e., normally within reach of hand). · Nondestructive testing (NDT). A close inspection made by electrical, electrochemical, or other methods to detect hidden damage. · Sample taking. In some cases, the NDT for concrete provides only provides information only of the surface (less than 20mm), and if doubts exist concerning the actual level of chlorides penetration or carbonation depth, samples may need to be taken. Adequate filling of the space left behind is necessary. · In-water survey. Survey is carried out underwater by divers and/or remotely operated vehicle. It usually is for cleaning marine growth. Inspection Program. The inspection program was defined by dividing the asset into different zones. · The submerged zone is everything below the water surface at the service draft. · The splash zone is the area submitted to intermittent wetting by waves.

· The atmospheric zone comprises structure and equipment on and above the upper deck. · The internal zone includes all structure, spaces, and reservoirs beneath the upper deck. Each zone is divided into subareas, each of which envelopes structure and/or equipment with similar inspection scopes. The different surveys were defined to detect typical concrete-degradation processes. · Effect of seawater on cements (i.e., sulfate and chloride) · Lime leaching/carbonation · Alkali/aggregate reaction · Reduction in cement content and strength · Increase of permeability (permitting chloride ingress) · Fatigue As expected with the use of highquality concrete, only a few defects were found. Concrete Damage From Steel Corrosion. Even without mechanical degradation, there are two major situations in which corrosion of reinforcing steel can occur: carbonation and chloride ingress. With either, the removal of the protective passive film leads to the galvanic corrosion. When this occurs, the produced rust requires more space than the original steel, straining the surrounding concrete. Because concrete is relatively weak in tension, cracks develop, exposing the steel to even more chlorides, oxygen, and moisture, and the corrosion process accelerates. Corrosion-protection systems were set in place during construction. Cathodic protection (CP) is a technique to control the corrosion of (reinforcing) steel by making the steel the cathode of an electrochemical cell. CP is the reduction or elimination of corrosion by making the metal a cathode by connecting it to a sacrificial or galvanic anode, or by use of an impressed direct current. Cathodic areas in an electrochemical cell do not corrode. If all the anode sites are forced to function as current-receiving cathodes, then the entire metallic structure would be a cathode and corrosion would be eliminated. Electrical continuity of all passive steels and other structural metallic parts is necessary. The 183 anodes (on internal shell in the four water-ballast compartments)

86

JPT · FEBRUARY 2009

provide a steel-reference electrode potential of approximately -850 mV to protect the reinforcing steel. Exposed steel structures are painted. It is necessary to clean exposed steel and repaint it to avoid loss of steel and prevent concrete damage. FEA Model vs. Real Life. The FEAmodel assessment has verified that the design keeps the overall structure in compression. However, the structural analysis highlighted some localized areas where lack of compression could be found. These localized areas coincide with the findings of an inspection showing superficial defects, mostly the result of the practical difficulty of concrete reinforcement in these areas. Examples are degradation of deck edges and flare-tower support ends, the latter having prestressed cables ending on them. These items are easy to repair and do not need a hightech qualification to restore them to asbuilt condition. Flare-tower-cantilever surface cracks were explained by the

FEA model showing that the design was optimized for the site conditions. Particular surveys have been defined for this member as a result of the numerical calculations. Conclusions Thus far, CP is actively protecting reinforcement steel. The high-quality concrete is providing adequate protection against carbonation and chloride penetration and assuring satisfactory aging. After 10 years, concrete-hull maintenance and repairs consisted basically of restoring concrete-surface cover lost from abrasion and impacts. The design enhanced compartment inspectability. Spaces are open without intermediate members blocking the view GVI is easily conducted with several fixed illumination sources, although additional means of access for CVI of the upper parts are necessary. The main background danger will always be steel corrosion. A better comprehension of the structure-aging process could have

been obtained if samples had been prepared during construction and left on board (at ambient conditions) for later strength- and mechanicalproperties tests. Similarly prestressed samples for fatigue-capacity reassessment should have been prepared before construction, rather than basing the base-design values on literature. Construction-quality control is required to ensure durability. The definition of the floating-unitsintegrity-management program for NKP allowed changing from a passive- and corrective-action frame to a proactive scheme. Main nonaccidental-degradation processes and mostexposed locations have been identified, enabling establishing an inspection program particular to NKP. Creating the FEA model representing as closely as possible the as-built and site conditions, helped in understanding the survey outcome. In case of an accident, the FEA model could be used to evaluate the condition of the unit and help make the right decisions. JPT

RESOURCES

New EXODUS 6.0

Reservoir Simulator For Microsoft WindowsTM

Full featured, user friendly, over 250 installations

¨ Paralleled

- faster runs with more cpus - current clients get 2 cpu version free

¨ ¨

CBM modeling

- with time desorption

Hydraulic fracs template

- automatic LGR with time dependent fracture conductivity

Call: (403) 265-9722 E-mail: [email protected] www.petrostudies.com T.T. & Associates Inc. T

JPT · FEBRUARY 2009

87

Information

ContentsFeb.indd

14 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

233694


You might also be interested in

BETA
ieeev4.dvi
ContentsFeb.indd
2FPS2
Microsoft Word - OTC 2011- Paper 21400 - An Evaluation of Fatigue Performance of Subsea Wellheads _Evans & McGrail_