18

Click here to load reader

fpso mystras

Embed Size (px)

DESCRIPTION

fpso mystras

Citation preview

Page 1: fpso mystras

Copyright 2004, Offshore Technology Conference This paper was prepared for presentation at the Offshore Technology Conference held in Houston, Texas, U.S.A., 3–6 May 2004. This paper was selected for presentation by an OTC Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Offshore Technology Conference or officers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented.

Abstract A recently completed FPSO conversion project shows some novel features in FPSO design. An existing trading tanker was selected for conversion into a 48,000-bopd FPSO unit, for a site offshore Nigeria in a water depth of 70 meters. While engineering of the FPSO was already underway it was discovered that the field production characteristics were underestimated. A crude oil throughput of 80,000 bopd was required, which corresponds to a medium size process installation. This happened to be in line with the contractor's “generic FPSO” design and left the designers with the challenge of fitting this medium size process installation on a relatively small sized tanker hull. To accommodate the increased size of the production facilities, sponson extensions were fitted over a considerable length of the vessel's side. By the use of sponsons the vessel's main deck area, ballast tank volume and buoyancy were increased considerably. However installation requirements for the spread mooring system, large central pipe rack, separate cable tray supports and marine deck piping still resulted in a congested deck space, which demanded a thoroughly planned lay-out of the main deck arrangement, pushing it beyond existing limits, resulting in an innovative design. This paper addresses and details the design decisions and their backgrounds, also reviewing alternatives considered. Special attention is given to the life extension calculations of the hull structure and the effect on these of adding sponsons and the cantilever supports. And finally, practical experience during vessel conversion and construction from the conversion yard in Dubai is discussed. Introduction Early December 2000 first preparations were made for the pre-engineering for a small sized FPSO with a production capacity

of 48,000 bopd and a storage capacity of 1,000,000 bbls for Agip Energy & Natural Resources (Nigeria) Limited (AENR). In March 2001 a letter of intent was issued by AENR to FPSO Mystras Producao de Petroleo Limitada (FMPP) - a Joint Venture (JV) of Saipem and SBM - with a tight contractual first oil date of December of that same year. An existing Suezmax tanker “Mystras II” (140,000 dwt) was purchased and laid up waiting for the conversion to commence. Due to the early first oil date required, a separate fast track solution was selected. The existing FPSO Jamestown, also owned by FMPP, was converted and fully refurbished between September and November 2001 in Portugal with minimum modifications to the existing process facilities. Hereby the contractual date of end December 2001 could be achieved and the FPSO Jamestown would remain on site until the FPSO Mystras– with full crude oil production capacity, would become available. Early 2002 first field production data received from Jamestown proved better than initially expected and justified exploring the possibilities of considering a larger unit for the field. It was decided to retain the Mystras hull, but to fit larger topsides on it, which left the vessel designers with a number of challenges and implementation of some novel design features. The contractual first oil date for the new unit was late December 2003. This paper is divided into three parts. The first part describes general project information:

- project design basis, - project organization and execution.

Prior to discussing the design features a brief description of the process facilities and their function is described:

- process description. In the following paragraphs the design process and decisions made are presented:

- vessel description before conversion, - main deck layout and design, - sponson design, - module cantilever support design.

In the last part some design questions are further detailed in the following paragraphs:

- side shell fatigue,

OTC 16198

Design and Conversion of FPSO Mystras T. Terpstra, IHC Gusto Engineering B.V.; G. Schouten, Single Buoy Moorings Inc.; L. Ursini, Saipem Energy International S.p.a.

Page 2: fpso mystras

2 OTC 16198

- longitudinal strength, - motion behavior, - freeboard calculation.

Project Design Basis The FPSO Mystras will initially receive crude oil from Okono & Okpoho oil fields that are located in Nigeria, offshore Port Harcourt in Block OML 119 (previously OPL 91). The OML Block 119 block is located in the offshore Niger delta, 55 km from Bonny Terminal at a water depth of 70 meters, see also table 2 and figure 1 and 2. The main functions of the installed topsides process facilities are:

- oil separation from associated gas and process water, - oil stabilization for storage in the FPSO cargo tanks, - gas dehydration and compression for gas lift

requirements (Okpoho Field) and for injection in a dedicated subsea well (Okono Field),

- produced water de-oiling and degassing before discharge to sea.

The FPSO unit has been designed in compliance with recognized international standards for the process facilities for a design life of 15 years. The vessel has been converted in full compliance with the latest IMO regulations and is classified with the American Bureau of Shipping (ABS) having the following class notation:

- A1 Floating Production Storage and Offloading System (FPSO), FL (15), Site Specific

The FPSO unit is designed to receive production risers from the Okono field sub-sea completed wells and import and export risers from the Okpoho field. Balconies for the risers & umbilicals for the former and the latter field have been installed on the bow and the starboard side of the FPSO respectively. The Okono side balcony is designed for the following risers:

- four (4) x 4inch production risers, - four (4) x 4inch umbilicals, - four (4) future production risers & umbilicals, - one (1) future gas injection riser.

The Okpoho bow balcony is designed for receiving:

- one (1) x 16inch production riser, - one (1) x 8inch gas lift export riser, - one (1) x 20kV power cable, - one (1) x SSIV manifold control umbilical.

A shuttle tanker will offload the FPSO at regular intervals via the offloading riser, subsea line and CALM buoy system.

Project Organization and Execution The main contract was divided within the JV as follows:

- topsides process facilities design & procurement – Saipem,

- vessel refurbishment and conversion design & procurement– SBM,

- spread mooring system design & procurement – SBM,

- topsides fabrication and vessel conversion – FMPP, - FPSO operation in Nigeria – FMPP.

The expertise of the various engineering houses within the organizations of the JV partners was utilized to the full extent to assist in the design of the process facilities (Saipem Energy International (SEI)) and the vessel refurbishment & conversion (IHC Gusto Engineering). As result of the JV the engineering offices and construction / fabrication sites were located all around Europe and the Middle East. During the entire duration of the project a major effort has been put into the various engineering scope of work definitions, but foremost in the initial stages of the project. Even more important was the consensus on the physical interfaces agreed between the engineering parties and construction sites. This approach has led to the design, engineering, construction and installation of the FPSO Mystras within schedule and budget. The actual design and conversion of the enhanced FPSO Mystras commenced in May 2002. The contractual delivery schedule –end 2003- and the aim for cost efficiency required a “fast track” approach based on the experience gained from previous completed FPSO projects. It was decided that the process facilities for this unit had to be based to a large extent on the successful generic FPSO series completed over the last two years. Hence the same topsides normally fitted on a VLCC sized tanker were to be installed on a smaller Suezmax type of tanker. Process Description The FPSO process facilities have been designed for oil separation and stabilization, gas compression and treatment, produced water treatment and various utility systems. The oil processing system is designed for the following peak production rates, which do not necessarily coincide: OKONO & OKPOHO Fields:

- design oil production 80,000 bopd, - design gas processing (incl. returned lift gas)

100 MMscfd, - design water production 55,000 bwpd, - design total liquids production 100,000 blpd.

The process facilities are installed on the upper deck and comprise a total of 16 modules with a maximum individual weight of 1,100 tonnes. The total process facility weighs approximately 5,500 tonnes including flare tower and pipe rack with the following functions:

- Module 01 High Pressure (HP) Separation - Module 02 Low Pressure (LP) Separation - Module 04 Flash Gas Compression - Module 06 Gas Injection Compression - Module 07 Gas Treatment - Module 11 Topsides Utilities - Module 12 Power Generation - Module 13 Local Equipment Room (LER)

Page 3: fpso mystras

OTC 16198 3

- Module 14 Chemical Injection - Module 16 Flare Knock-Out Drums - Module 18 Okono Inlet Manifold - Module 19 Okpoho Inlet Manifold - Module 21 Oil Export and Fiscal Metering - Module 22 Laboratory

The central pipe rack - Module 20 - is installed at centerline of the vessel in between the modules, running from just forward of the accommodation to the bow of the vessel. This pipe rack supports all interconnecting piping – both the marine piping system on the lower and the topsides facilities piping on the higher levels respectively - and provides central means of access to all modules along the length of the vessel. A diesel driven traveling crane with a lorry is mounted on top of the pipe rack. The crane boom is sized to be within the reach of all extremities of the modules including a dedicated lay down area (Module 15) located near the supply boat mooring area and the remaining laydown areas on the main deck. The traveling crane has a minimum and maximum capacity of 5 tonne at 30 meter and 10 tonne at 25 meter respectively, sized such that large pieces of topsides equipment can be transported along the FPSO process deck. Module 01 Module 01 provides the initial stages of the crude separation. The incoming production fluids are heated and allowed to settle first in the HP separator. This provides a primary coarse separation of oil, water and gas phases. Module 02 This module provides the final stages of separation for the production fluids. Crude from module 01 is further separated in the LP separator and then in the electrostatic treater, remaining gas and water are removed from the oil. This produces sales quality oil, which then flows to the cargo tanks for storage until offloading. Water that has been separated from the oil is processed to remove dissolved gas, after which it flows to the slop tanks for settling to remove any residual oil before overboard discharge. Module 04 Separated gas from the HP separator on module 01 is compressed to high pressure to allow it to be injected at the subsea wellhead to facilitate the flow of the incoming production fluids. Some of the gas is used as fuel on the FPSO, both in vessel’s boilers and to fuel the five dual fuel gas turbines, which drive the power generators and the main gas compressor. The gas compression is achieved on module 04 using two three-stage centrifugal gas compressors and associated suction, interstage and discharge coolers and scrubbers. Module 06 High-pressure gas from the second stage of the main gas compressor must be further processed to remove water before it can be used as fuel. This drying is achieved in module 06 where the gas is washed with glycol to absorb the water. Some dry gas is used as the normal source of fuel for all the FPSO

power requirements. The remaining gas is returned to module 04 for the final stage of compression. Module 11 The processes described above need three main utility services –heating, cooling and electricity. The first of these two are provided by module 11, which contains circulating hot fresh water for process heating and cold fresh water for process and compressor cooling. The heating for the hot water system is achieved using steam from the vessel’s boilers, and cooling for the cold water system is achieved by using seawater. Seawater for the topsides is provided by the three (3) submerged sewaterlift pumps installed in caissons at port side integrated in the new sponson structure. FPSO Power Generation Three dual fuel gas turbine driven generators located on Module 12 provide electric power for the FPSO. Under normal conditions two of the three topsides High Voltage (HV) generators are on line. These will supply all topsides and vessel consumers. The topsides and vessel generation & distribution system is linked via a transformer, located in the vessel Engine Room (ER). The High Voltage (HV) and Low Voltage (LV) power distribution for the FPSO consumers is located in a purposely-built local equipment room located on Module 13. The vessel emergency generator supplies power to both the vessel and topsides emergency consumers. Vessel Description before Conversion The original vessel selected for conversion was an existing single hull tanker with a diesel main propulsion plant, built in Japan. The original hull structure is fully made of mild steel, one of the major assets to select this vessel for conversion into an FPSO. The tanker “Mystras II” has the following main particulars: Builder: Mitsui Shipbuilding & Engineering Co. Ltd. Yard number: 1017 Class : Lloyd’s Register of Shipping

: 100A1 Oil Tanker

: LMC & IG SYS Length over all : 271.000m Length between PP : 260.000m Scantling Length : 257.826m Breadth moulded : 44.000 m (excl. sponsons) Breadth moulded : 55.400 m (incl. sponsons) Depth moulded : 22.400 m (main deck) Depth moulded : 22.061 m (sponsons) Draft at Summer Load Line: 17.021 m Keel plate thickness : 26.50 mm Deadweight : 138,930 ton Blockcoefficient : 0.818 (excl. sponsons) Blockcoefficient : 0.690 (incl. sponsons) Design speed : 15 knots (excl. sponsons) Main Deck Layout and Design In table 1 the main differences between a typical VLCC and a Suezmax tanker are illustrated. From this table it is evident that on a VLCC 43% more deck space is available to fit the

Page 4: fpso mystras

4 OTC 16198

same process facilities. The consequences of fitting the larger generic topsides design for a VLCC onto a smaller Suezmax tanker are illustrated in figure 3 and 4 and table 1, resulting in a congested main deck area. In the very beginning of the project it was decided to use a 3D-computer program to model the complete vessel main deck and topsides process facilities. This was not only to have better control of the main deck arrangement but also better coordinate the involvement of various engineering disciplines. The master 3D model was initially developed by the topsides design team to arrange and locate the process modules such that an optimum process design was obtained. This was achieved in combination with the vessel concept design. Hereafter the ownership of the 3D model shifted to the vessel design team. This team used the model to further develop main deck piping layout and detailed routing of cable trays, escape routes and further clash checking of the mooring lines with the main deck equipment. The main structural, equipment and systems that governed the design and sequence of the 3D model built-up are listed and described in more detail here under:

- spread mooring system and riser installation, - center line pipe rack, - topsides module support structures, - marine piping systems, - topsides cable trays.

The FPSO general arrangement is presented in figure 5. Spread Mooring System and Riser Installation The FPSO is spread moored and connected to the seabed via four bundles of three mooring chains each connected to uni-joint chain hawses integrated in the hull structure. There are two balconies accommodating the riser porches and hang-off connection for umbilical are positioned at the bow, Okpoho field, and at starboard, Okono field, forming the connection with inlet module 19 and 18 respectively. For installation of the mooring chains and risers and umbilicals a mooring and riser installation arrangement on the FPSO main deck has been developed using permanent installation winches and sheaves on main deck guiding the wire to the mooring chain hawses and riser balconies. The design adopted on FPSO Mystras is derived and further evolved from a recent FPSO project also moored in West African waters. On this recent FPSO project a single permanent hydraulic driven winch, located at the forecastle deck, served pulling of both risers and mooring lines. This arrangement had the following characteristics:

- one single hydraulic winch, - winch capacity 175 SWL, - wire diameter ∅72mm, - 8 wire guide sheave blocks located along main deck.

A single winch was selected at the time because of cost

reasons. However, during the design phase it appeared that a significant amount of additional sheavs had to be installed, which offset the cost benefit of a single winch. Furthermore pulling in of the wire and changing connection to other mooring lines proved to be difficult and very time consuming during installation at the time, especially handling the heavy wire end socket weighing in excess of 300kg. Secondly the mooring arrangement had been given a lower priority during the design stage at the time resulting in less optimal wire routing clashing at several locations with the main deck piping. Based on experience from previous projects, for FPSO Mystras two separate hydraulic driven installation winches were ordered each served by a separate hydraulic power unit (HPU) and the main deck mooring and riser installation arrangement was given highest priority during the concept and subsequently design stages. The FPSO Mystras installation arrangement has the following characteristics:

- two hydraulic winches, - winch capacity 90 SWL, - wire diameter ∅58mm, - 6 wire guide sheave blocks located along main deck.

The forward winch is located at the forecastle deck at portside at the base of the flare tower and serves the forward mooring chains and the bow balcony. The aft winch is positioned at starboard at frame 59/60, serving the aft mooring lines and the side balcony, see figure 16. Due to restricted deck area, even locating the HPUs became of an problem. To resolve this and also to remain out of hazardous areas, the HPUs and associated control panels are positioned at an elevated level alike the process decks. An additional advantage of this configuration is the improved operator’s view over a part of the wire routing. Various sheaves are located along the FPSO main deck to guide the installation wire in straight lines to mooring boxes and balconies. At centerline fore and aft two sheaves are positioned underneath the pipe rack, from where the installation wires leaves for the fore and aft mooring boxes respectively. The angles relative to the vessel’s longitudinal axis were critical since they correspond with the mooring chain orientation of the field layout. A characteristic difference with the earlier design is the principle of pulling in the mooring chains through the chain hawses and the single-piece-cast steel hawses with a 45 degree-bend. To avoid this on FPSO Mystras straight chain hawses were used in combination with removable bolted vertical rollers mounted to two rails at each mooring box. The rollers assured proper alignment to the horizontal deck sheaves that are located at 640mm above deck at centerline. The bolted configuration allowed positioning of the roller in-line with the different angular orientation of each individual chain, which was eased by pre-drilled holes, see figure 15. The vertical rollers are robustly constructed from plate material assuring

Page 5: fpso mystras

OTC 16198 5

proper passage of the chain, wire end socket and ∅58mm wire and can be relocated using the central traveling crane. To ease exchanging the existing flow lines and risers from FPSO Jamestown the riser arrangement on the starboard balcony on FPSO Msytras has been kept identical. For the forward balcony a new arrangement was defined. Pulling in of the flow lines was achieved via an A-frame structure, see figure 17 and 18. The A-frames accommodate a movable vertical sheave to allow straight pulling trough the riser porches, possible due to flow lines leaving the porches at an angle of 3 degrees. Center Line Pipe Rack The design of the centerline pipe rack used for the generic FPSO series has been copied and adopted for installation onboard the Mystras. Due to the different size in vessel the length of the initial design has been decreased and the frame spacing was changed – from 5,200 mm to 5,000 mm – to ensure a proper alignment with the vessel under deck structures. Due to the installation of the traveling crane on top of the pipe rack, the original rack width had to be maintained. This determined the transverse position of the topside modules on the vessel main deck. As indicated before the central pipe rack supports the marine piping systems located on top of the transverse structural beams approximately 1m above main deck level. The process piping is supported on the higher levels of the pipe rack - 4.0 meters above upper deck level – in order to align with the top of the topsides modules decks. Since the majority of the cargo tank access hatches is located in way of the extremities of the transverse pipe rack supports it was envisaged that hatches had to be relocated due to clashes but this proved to be not the case. Free access to all hatches is possible. However in order to be able to route the various mooring installation and riser pull in wires special slots were cut in the transverse supports to have free running of the wires to the various locations on port side, starboard side, forward and aft. Topside Supports A column support principle has been used for supporting the process modules, a simple and relatively light lattice structure with maximum transparency for passing through deck piping, cable trays and access. Transfer of hull girder deflection to topsides modules is isolated by means of flexible gusset plates, which are full penetration welded to the main deck to enhance fatigue strength. In general three vertical columns in transverse direction are used to support the module. Only the inner bay is used for diagonal braces to cope with roll motions. The outer bay is used for running cable trays and escape routes. FPSO Mystras has a hull frame spacing of only 5,000mm, compared to the 5,200mm-pancake girder spacing of the generic FPSO topsides’ design. This would cause misalignments. Adjusting the spacing in the modules to

5,000mm would cause too much engineering rework and the advantage of copying the previous design would be lost, since it would imply a redesign of the modules and equipment supports. Alternatively it was decided to incline the vertical columns connecting the topsides main girders with the vessel web frames positions. Not only the main girder arrangement was different but also the vessel longitudinal stiffener spacing was different. This also asked for inclining the vertical columns in transverse direction, resulting in an arrangement of double inclined vertical columns. Marine Piping Systems The design of the topsides modules and subsequent support structures was more or less fixed due to the design adopted from the Generic FPSO units. The task of routing large diameter piping on the vessel main deck and interconnecting the various structures and equipment – Inert Gas vent stacks, Fire water deluge boxes & Crude oil offloading - caused a few headaches. Also the fact that there was hardly any deck area or space left on the main deck to relocate any structure, equipment or piping. Many interface meetings were necessary to discuss and agree on the various scenarios and possibilities of changing certain structural or piping items but more than once resulting in new challenges. The routing of the installation steel wires coincided with the marine piping systems mainly forward of the accommodation on starboard side and near the forecastle area, mainly due to the required height of the wires above the upper deck. The wire routing at the various locations was modeled in the 3D model and by re-routing various structures and piping the majority of the clashes were resolved at the design stage. Spool pieces were required to be left out for proper running of the mooring lines during installation of the vessel on site and to avoid last minute surprises that already critical coated systems had to be disconnected. Secondly the spool pieces are such located that during pulling in of the future risers production can continue without disconnecting the critical pipe systems. Cable Tray Routing Routing of the cable trays and their support structures was one of the most difficult tasks to accomplish. The cable trays are positioned beneath the topsides modules and are supported from the main deck via dedicated support structure and are derived from the generic FPSO design series where the principle was first used. The trays are supported from the main deck instead of suspending them from the module pancake structures to allow early cable pulling and maximum outfitting prior to module installation. Construction at the fabrication yards and transport to the installation yard were consequently improved. A disadvantage is the height of the trays and in cases some cable pulling is done after module installation. A disadvantage is that additional staging is then required at main deck level and

Page 6: fpso mystras

6 OTC 16198

this hampered other outfitting work carried out simultaneously. The trays had to be routed such that no tank hatches are obstructed, to allow proper handling during evacuation from the cargo tanks and lifting of equipment into and from the tanks. Routing was done such that they fitted in between the columns and braces of the module supports. Beneath module 04, however, trays were routed around the middle columns since no other space was left. Only beneath module 12 – power generation - and Module 13 – LER -, with the highest concentration of cables - trays are suspended from the module itself, since the number and arrangement of the tubular supports prohibited support from the main deck. Sponson Design The increase in physical size and weight of the process utilities could not fit anymore on the original relative small deck and would also exceed maximum draught in fully laden condition. Consequently the contractual 1,000,000 bbls at a density of 0.82 t/m3 could not be loaded. To overcome this several design options were considered including lengthening the vessel or adding sponsons. Lengthening the vessel would not only imply inserting a new mid body, but also reinforce the existing ship structure over a distance of 0.4L to cope with increased bending moments. This option was considered as too expensive and would disrupt the conversion time frame drastically. A new inserted mid body would require a total of 1,200 tonnes of steel. However cutting the vessel in two in dry-dock, cutting all existing ship systems and disruption all activities on the main deck would incur a large cost and seriously jeopardize delivery date. Adding sponsons evolved as the single viable option with minimum disruption on conversion schedule. Addition of sponsons to the side shell should overcome the following restrictions in the existing vessel:

- increase deadweight, - increase deck space area, - enhance side shell fatigue life, - provide a secondary barrier against supply boat

impact. The first conceptual design comprised sponsons having a width of 5.7 meters extending from frame 61 up to frame 91, a distance of 150 meters, viz. 58% of ship’s length, and would cover virtually the complete cargo tank area. Maximum deck space would be created and the spread mooring integration structures could be directly integrated in the new steel structure, without affecting the internal existing ship structure. However, the sponson length of 150 meters extended beyond the parallel mid body and this would increase complexity and fabrications costs. Secondly the additional amount of buoyancy created was far in excess of what was minimally required. Only extending the sponsons over half the depth of

the vessel solved this, but this would seriously hamper side by side mooring of a supply boat at low draught condition and could cause fenders getting stuck underneath the bottom of the sponsons. Eventually the required amount of buoyancy was met by extending the sponsons over a distance of 90 meters, equal to two cargo tanks, viz. crude tank no. 3 and 4. An inclined plate forms the sponson bottom, with the upper knuckle at 6,300mm A.B. and the connection to the original side shell at 3,700mm A.B. in-line with the lowest original side shell longitudinal. This option did not solve the lack of deck space sufficiently to support the most forward and aft modules. Cantilevering some modules and supporting them by large brackets welded to the side shell solved this. A challenge left for the designers was how to connect the sponson top to the existing main deck structure. Traditionally in tanker design welding to the gunwale, connection of side shell to main deck and one of the most critical components in a tanker, is not desirable and often not allowed by Class. In table 3 three options studied are presented:

- removing existing round gunwale and inserting square section flush with sponson deck,

- top of sponson only extended up to first existing side shell longitudinal at 1,100mm below main deck,

- top of sponson flush with main deck, however connecting plate welded perpendicular to the gunwale.

The first option appeared too expensive, mainly due to the amount of additional new steel required and the significant scaffolding internally required in the cargo tanks 3 and 4. This would extend the dry-dock-period by three weeks and further increase costs. The second option would solve welding in critical areas but would not increase deck area. Fitting a grating deck flush with the main deck could have solved this. However, the additional amount of secondary steel and increased maintenance offset the benefits. The most viable option appeared to be welding perpendicular to the radiused gunwale plating with minimal impact on schedule and costs. In close corporation with Class the structural integrity was validated. The main disadvantage of the third option is the V-groove appearing between sponson and main deck, since it could collect water and prohibit proper draining of the main deck. Also it would obstruct safe access of personnel. Fitting a removable 8mm-thick plate solved the latter problem. To allow for proper drainage of the main deck, removable plugs in every transverse fire coaming on the main deck is fitted. This allows drainage to the aft slop tanks. Filling the V-groove with mastic was disregarded since it would hamper in service inspection and could initiate corrosion. Utilizing higher tensile steel, quality AH36 and DH36, for the sponson longitudinal members allowed for a significant weight reduction of 200 tonnes on a total of 2,000 tonnes. The transverse frames are made of mild steel. To assure sufficient fatigue strength of the new structure special consideration was

Page 7: fpso mystras

OTC 16198 7

given to bracket details and stringer toes in compliance with Tanker Structure Co-operative Forum guidelines [3], resulting in a fatigue life well in excess of 75 years. As a consequence of fitting the sponson tanks, higher shear loads in fully laden condition are introduced into the web frames. The no 3 and 4 wing tank cross-ties-to web frame connections were overstressed and have been replaced by new and modified designed brackets. The modifications and the sponson structure are presented in figure 6. Not foreseen during the basic design phase was the interaction between the sponson lower knuckle and the fenders of the outfitting quay after completion of sponson assembly in dry-dock. The ballast system in the sponson tanks was not completed and topsides not yet installed at that time. Consequently the lower knuckle at 3,700mm A.B. extended beyond the quayside fenders not allowing adequate mooring of the vessel. A creative solution was found in welding three berthing skirts underneath the sponson bottom plate as demonstrated in figure 12. The longitudinal position of the skirts matched the position of the quayside fenders, the FPSO being moored at a predefined position. The temporary berthing skirts served their purpose well. During a severe storm in late April 2003 several ships were damaged and some broke from their mooring line in a port in Dubai. However, no damage was found after inspection at the connection of the skirts and the sponsons. Module Cantilever Support Design The relative short sponson length combined with a 8.8-meter wide pipe rack and pre-designed topsides resulted in four modules cantilevering over the original side shell just fore and aft of the sponsons. Unfortunately this also included the heaviest module 04 weighing 1,100 tonnes. Placing the outer vertical columns diagonally and welding to the main deck was initially considered. However, this resulted in an inefficient load transfer to the hull structure in such a way that internal reinforcements would be required, since the outer column would be less effective in transferring vertical loads. A solution to support the modules was found in welding large brackets to the side shell supporting the outer vertical columns, see figure 13 and 19. The cantilever supports are aligned with the first side shell longitudinal measured from deck level to avoid local welding in the gunwale area and is positioned at 1,100mm below deck at side. Since regular access is not required to the cantilevers, the lower elevation to main deck is not considered as an issue here. Due to the double inclination of the module vertical columnar supports, approximately 10% out-of-plane load is imposed to the cantilevers. This required additional external stiffening to obtain sufficient bending stiffness in the horizontal plane. The outer longitudinal bay of module 04 accommodates pitch

bracings, see figure 13, further imposing horizontal loads on the cantilevers. Three external longitudinal T-shaped stiffeners connecting the cantilever supports have been welded to the side shell to avoid internal reinforcements. The possibility of a supply boat colliding with the cantilever module supports was addressed in Hazid and Hazop studies. From the safety analysis it appeared that the cantilever supports are relatively well protected between sponson ends and the mooring structure. However, an accidental case of one vertical support including cantilever support being damaged was reviewed. The ultimate limit state analysis performed proved that the module support structure has sufficient residual strength and can cope with the loss of one damage support. Design Aspects Side Shell Fatigue A deterministic fatigue assessment of the longitudinal ship structure was conducted using ABS Safehull® Phase A procedure for the FPSO with and without sponsons. A novelty in the fatigue assessment within Safehull® A, is that two separate models have been created:

- tanker without sponsons, FPSO 3C, - FPSO with sponsons, FPSO 5C.

For both models SEALOAD calculations have been carried out. The results with respect to the FPSO3c model show the damage of the vessel accumulated during her life as trading tanker. This could be used to assess the remaining fatigue life of the ship as FPSO without sponsons. Performing SEALOAD calculations with the FPSO5c model gave results as if the ship had sailed all her service years with sponsons. This would produce unrealistic calculation results. Therefore the results for both models were manually corrected using a spreadsheet program based on the following formula:

cFPSOSH

cFPSOSH

cFPSO

ScFPSOSHcFPSOR L

LLLL

3

5

32353

−−− ⋅⎥

⎤⎢⎣

⎡−⋅=α

α ……...(1)

By correcting the data sets this way, the true damage accumulated to the existing structure is calculated, plus the effect of new sponsons in its future life is taken into consideration. While applying the correction to the original results, the following assumptions were made:

02

SL for the new sponsons.

13

5 =−

cFPSOSH

cFPSOSH

LL

for the new sponsons.

Page 8: fpso mystras

8 OTC 16198

The original ship structure is made of mild steel with web frame stiffeners directly welded to the flange of the side shell stiffeners. These structural details are classed as F2-details, showing an enhanced fatigue life compared to the overlap connections frequently used in 25-year old tanker structures. The latter type should is classed as a G-detail, showing poor fatigue strength. A local hull strength reassessment for site-specific environmental conditions based on so-called environmental severity factors of type β was performed. The β-type factors reflect the difference in design loads between site specific and unrestricted service of the vessel. The FPSO Rules do not allow β-factors below 0.5; although that for benign environments like West Africa the average calculated β-factors are 0.35. Adding sponsons changed the original side shell partly into an inner longitudinal bulkhead, the latter not being exposed anymore to so-called intermittent wetting. The fatigue analysis revealed that the majority of the original tanker structure fatigue life already had been consumed by the tanker’s trading history. The differences on future fatigue life enhancement with and without sponsons on the existing side shell were only marginal; the local external wave pressures in benign environment are mild anyway and do not significantly contribute to the cumulative fatigue damage. The only fatigue damage contribution with sponsons stems from the small dynamic hull bending stresses and internal dynamic tank pressures. The differences would have been more pronounced, and in favor of an FPSO with sponsons, in case the FPSO would have been moored in harsher environment. A moderate total of 28 backing bracket reinforcements were required in the region of the former splash zone and were welded at longitudinal SL 16, 17 and 18 at the aft side of every transverse oil tight bulkhead. Longitudinal Strength Re-assessment of hull structural strength for benign environment FPSOs result in reduced scantlings, with a minimum of 85% of those for unrestricted service. Consequently hull girder strength had to be recalculated as well. For FPSO Mystras the hull girder strength has been reassessed for Nigerian conditions, the sponsons largely compensate for loss in section modulus. Although the sponsons do not extend beyond 0.4L amidships (sponson length is 0.33L), they have been included in the longitudinal strength, an assumption verified by the ABS Safehull® Phase B analysis. Due to the asymmetric sponson shape the neutrals axis shifts upward. This is compensated by an increase in moment of inertia; the section modulus at bottom remains unchanged without affecting bottom plate renewal criteria. In fully laden condition the empty sponson tanks reduce the sagging still water bending moment by approximately 15% compared to the trading tanker practice, caused by additional

buoyancy amidships. In light draught condition filling of the sponson ballast tanks reduces the hogging bending moment. The sponson cross section is such designed that in light draught the lower part is emerged from the water, not to further imposing any additional buoyancy forces on the hull. Motion Behavior The FPSO roll motions without modified bilge keels were too severe for the original generic FPSO topsides design and this would result in undesirable down time. The maximum roll angle on site exceeded 12 degrees. Therefore further action had to be undertaken to reduce this. Adding sponsons did not sufficiently reduce roll motion as expected, although an increase in radiated damping from 1.64% without sponsons to 3.11% with sponsons. The relative ineffectiveness in reducing motion behavior, and thus roll damping effect, of the sponsons is explained by the sloped bottom plate, which allow a relative smooth water flow around it. The alternative of redesigning the sponson sloped bottom plate into a horizontal one was soon disregarded in the concept design phase due to the possibility of supply boat fenders getting stuck underneath at portside and to limit any slamming effects. The latter effect was considered less of an issue in benign environment. To further reduce motion behavior it was decided to modify the existing bilge keels. The original bilge keel depth of 500mm has been increased to 650mm. The tanker original bilge keel was constructed of five separate individual sections with a brake at the hull’s main block assembly seam welds, which eased shipyard construction at the time. The final configuration of the bilge keel has a continuous length of 130 meters; the original seam brakes being plated in and thus also contributes to the longitudinal strength of the vessel. The maximum roll angle for the modified bilge keel arrangement does not exceed 9 degrees. Freeboard Calculation Calculation of minimum freeboard for FPSO Mystras proved to be subject to interpretation of the moulded breadth, and thus the corresponding CBF, of the hull at 85% of the depth. Due to the addition of the sponsons the freeboard mark had to be newly defined. Especially the relative short length of the sponsons, only 33% of the vessel’s length, gave room to different interpretation during the design process. Also the sponson shape option whereby the sponson deck was aligned with the first existing side shell stiffener – 1,100mm below main deck – was not clearly identified in the current International Load Line Convention (ILLC) 1966 regulations. However, during the design phase it was generally recognized that the sponsons increase reserve buoyancy and thus improves safety.

Page 9: fpso mystras

OTC 16198 9

According to ILLC 1966 the correction on the tabular freeboard is calculated using the following formula:

TabBF SF*1.36

0.68CC += …………………………………(2)

A larger block coefficient, in excess of 0.68 results in a larger minimum required freeboard. Note that for block coefficient lower than 0.68 no correction is applied. The theoretical background for a higher freeboard in case of a CBF > 0.68 is based on the resultant upward force in a seaway. Freeboard and super structures create reserve buoyancy. They may be identical for two vessels with similar main particulars, although these vessels may have different block coefficients and thus displacements. Due to a CBF higher and mass a smaller upward acceleration will be exerted on the hull. Thus fuller vessels would require more reserve buoyancy to obtain similar seaworthiness. During the sponson design phase the following options were considered:

- original breadth of 44 meters and hull volume including sponsons at 85% of depth,

- new breadth of 55.4 meters after conversion and corresponding volume including sponsons at 85% of depth,

- remain original freeboard and further ignore any effect of the sponsons hereon.

The option based on the existing breadth and including sponson volume at 85% of depth resulted in a freeboard reduction. Taking the new breadth of 55.4 meters into account resulted in a 400mm smaller freeboard compared to the original. However, due to the straight continued deck camber over the sponsons the “new deck line” would be 360mm lower relative to the original deck line. Although the maximum draught could be increased by 40 mm, the scantling draught became the governing factor. In close liaison with Class it was eventually decided to weld the freeboard mark at exactly the same height above base line on the sponson side shell as the original Plimsoll mark was located. Vessel Conversion in Dubai The vessel refurbishment activities started in August 2002 at Dubai Dry-Docks with the initial required hull structure refurbishment to allow the FPSO unit to remain on station for the 15-year design life of the unit. In preparation of the vessel conversion the yard had already started to remove the majority of the main deck piping systems and upper deck structures, the original accommodation interior and the obsolete marine equipment in the engine room. The actual conversion works started beginning 2003. The conversion and refurbishment design was based on the philosophy to retain as much as possible of the original vessel systems, arrangement and equipment – especially the marine systems – and add the necessary systems and equipment for the 15-year design life as an FPSO unit.

Sponson Fabrication Sponson fabrication commenced early 2003 in the yard’s own plate shop and was completed two months later; a monthly steel throughput of 1,000 tonnes. The sponsons were subdivided into seven blocks, each further split in a lower and an upper section. The larger of the two, the lower section, weighs 80 tonnes. Lifting of a lower section is pictured in figure 7. Block assembly commenced late February and was fully completed late March 2003 in the yard’s largest dry-dock. This dock measures 520m by 100m and is equipped with four cranes, two of them having 120 tonnes lifting capacity. The sections were lifted into position along the vessel side shell until temporary connections were made after which welding started. This operation took 8 – 10 hours for the lower sections. The complete sponson block assembly went relatively smoothly without any major problems encountered. Although at the aft end blocks it appeared that the actual hull lines deviated from the original lines plan, resulting in some gaps between sponson bottom and existing shell plating. Fitting some insert plates however, relatively easily solved this. The connection of the sponson to the existing gunwale showed an excellent fit. The sponson fabrication and assembly is illustrated in figure 8 through 11. Upper Deck Marine System Installation Installation of the marine piping system started after the T-beam supports for the central pipe rack had been fitted on the centerline of the vessel. It was decided earlier on in the project to prioritize the routing the centerline piping systems in order to deliver the piping detailed design drawings to the conversion yard for fabrication and installation well before the top sections of the pipe rack were lifted and installed. This would facilitate the installation and hook-up of the piping systems. This approach resulted in only small-bore piping to be fitted after pipe rack installation. The detailed design for the piping branches for the various systems to be connected to the cargo and ballast tanks took more time than expected mainly due to the many engineering disciplines involved in the vessel upper deck structures arrangement and piping layouts as already indicated earlier in the document. One of the more difficult parts in the piping detailed design was the routing of the new piping system to be connected to existing openings in the vessel upper deck such as for the Crude Oil Washing (COW), Inert Gas and Cargo Rundown systems. The original vessel drawings indicated a location different from that of the actual openings in the deck. During the installation of the interconnection spools it became clear that the additional green material indicated on the design drawings served its purpose. FPSO Installation Offshore The offshore installation started early January 2004 and was

Page 10: fpso mystras

10 OTC 16198

completed within a twelve-day period. First oil was successfully produced on January 14, 2004, see figure 24. In figure 21 and 22 the hook-up and connection of the mooring lines and riser is illustrated. Only at the bow balcony some minor clashes were encountered, which could be easily solved. Conclusion This paper presents the design and conversion of FPSO Mystras. The FPSO Mystras is an extraordinary FPSO compared to the more regular designs based on VLCC sized tanker hulls. On FPSO Mystras relative large topsides are fitted based on the contractor's generic FPSO topsides design. These are fitted on a relatively small Suezmax tanker hull. The relatively large generic process modules were used to cope with the increased field production characteristics during the course of the project. A creative solution to cope with the increased production capacity was found in adding sponsons to the original tanker hull structure with the following advantages:

- increase deadweight, - increase deck space area, - enhance side shell fatigue life, - provide a secondary barrier against supply boat

impact. The FPSO was succesfully completed and commisioned early in January 2004. Nomenclature CB = Block Coefficient CBF = Block Coefficient at 85% of depth

L = Hull Length, L, m B = Moulded Breadth of Hull, L, m D = Moulded depth of Hull, L, m SFTab = Tabular freeboard SFReg = Regulated freeboard E = Modulus of Elasticity, m/Lt2, N/mm² I = Moment of Inertia, L4, mm4

S = Web frame spacing, span of stiffener, L, m δZ = Vertical deflection of hull structure, L, mm δX = Longitudinal elongation of hull, L, mm c = Height of deck above hull neutral axis, L, m α2 = Ship ith route environmental severity factor

for existing tankers where historical ship routes wave environment for the service life of the vessel differed from those of the North Atlantic Ocean.

α3 = Site environmental severity factor for existing tanker converted to a FPSO

LSH = Fatigue life calculated in Safehull®

MSW = Still water bending moment, mL2/t2, Nm QSW = Still water shear force, mL/t2, N Acknowledgements The authors wish to thank management of IHC Caland (SBM and Gusto), Saipem (SEI) and AGIP for granting permission to publish this paper. The authors express their special appreciation to their colleagues for their support and advice. References [1] Terpstra T. e.a.: FPSO Design and Conversion; A

Designer’s Approach”, proc. OTC 13210 presented at the 2001 OTC Houston, TX, April 30-May 3.

[2] American Bureau of Shipping; Guide of Building and Classing Floating Production Installations, June 2000.

[3] Tanker Structure Co-operative Forum (TSCF), Guidance Manual for Tanker Structures, 1997

Page 11: fpso mystras

OTC 16198 11

Tables

Table 1 – Main Particulars Item Symbol Unit Typical VLCC

Tanker Typical Suezmax

Tanker LO.A. - m 343.0 271.00 LP.P. - m 330.0 260.00

BMoulded - m 51.80 44.00 DMoulded - m 27.40 22.40 TDesign - m 21.40 17.02

Deadweight - tonnes 277,000 138,930 Displacement ∆ tonnes 299,965 159,100

Block Coefficient

CB - 0.820 0.818

Deck space A m2 13,260 9,240

Table 2 – Field Characteristics Okono and Amenam Field AMENAM Field OKONO Field

Water Depth 62.8 m 71.75 m

Clarke 1880 (MOD) Minna Datum WGS 84 Spheroid

Latitude 04° 00’ 09.90” North 03° 59’ 21.516” North

Longitude 07° 47’ 03.74” East 07° 17’ 28.641” East

NTM Mid Belt (8.5° East) Datum Nigerian Mid-Belt TM Projection

Northings 340 m North - 1,083.2 m North

Eastings 591,00 m East 536,346.0 m East

Table 3 – Sponson to Deck Options Option A Option B Option C

Item

Description Gunwale removed Sponson deck lowered V-groove at gunwale Structural Integration good good moderate Deck Space good reduced good Drainage of water good good moderate Capital expenditure high low low Schedule Impact 3 weeks none none

Page 12: fpso mystras

12 OTC 16198

Figures

Figure 1 – Location OPL 91 field Offshore Nigeria

Figure 2 – Okono and Okpoho Field Layout

Page 13: fpso mystras

OTC 16198 13

Figure 3 – Process Arrangment of FPSO Mystras

Figure 4A – Layout VLCC size FPSO

Figure 4B – Layout Suezmax size FPSO

Page 14: fpso mystras

14 OTC 16198

DN

DN

DN

DN

DN

DN

UP

DN

DN

DN

DN

DN

DN

DN

DN

DN

DN

DN

Figure 5 – FPSO General Arrangement

Page 15: fpso mystras

OTC 16198 15

Figure 6 – Midship Section Drawing

Page 16: fpso mystras

16 OTC 16198

Figure 7 - Installation First Block Figure 8 – First Block Fitted

Figure 9 – Progress, February 23, 2003 Figure 10 – Progress, March 16, 2003

Figure 11 – Sponsons finished, March 29, 2003 Figure 12 – Berthing Skirts

Page 17: fpso mystras

OTC 16198 17

Figure 13 – Module 04 Cantilevers Figure 14 – FWD PS Mooring Box Structure

Figure 15 – Vertical Rollers Figure 16 – Aft Installation Winch

Figure 17 – Bow Balcony Figure 18 – Side Balcony

Page 18: fpso mystras

18 OTC 16198

Figure 19 – Congested Deck Area Figure 20 – Top View Process Facilities

Figure 21 – Riser Hook-up Figure 22 – Mooring Line Hook-up

Figure 23 – FPSO Mystras leaving Dubai Figure 24 – First Oil Production, January 14, 2004