Subsea gas feels the pressure

The final chapter of a 30-year saga is unveiling itself in Norway. For here, the dream that was a twinkle in the eye of engineers in the 1980s is fast approaching reality as three existing offshore developments look to extend their economic producing lives by adding gas compression facilities — on the seabed.

The offshore fields — Aasgard and Gullfaks operated by Statoil and Ormen Lange operated by Shell — have been the focus of intensive technology development and testing programmes over the past decade. Each field presents a unique set of challenges and accordingly the technology solutions now being finalised have some fundamental differences. But all three — and the wider industry — stand to benefit significantly if subsea gas compression delivers on its long-held promise.

The pressure in gas reservoirs falls over time, causing gas production to decline and ultimately cease. In later field life operators must accept lower output as pressure falls, and modify processing equipment accordingly, or add gas compression, usually onboard an existing or new platform in the field to boost production.

“Installing subsea gas compression will provide a very beneficial alternative,” says Torstein Vintersto, Statoil’s vice president with responsibility for managing the company’s subsea compression portfolio. “Subsea compression has the potential to be applied at a lower capital cost, particularly if building a new platform is eliminated, and with greater flexibility. In fact, for subsea fields in water depths too great for offshore surface facilities, subsea gas compression is the only viable solution.”

Indeed, if subsea gas compression technology is proven to work reliably, then it will not only be a retrofit option for boosting declining gas pressure and enhancing recovery in existing fields. It could also open up opportunities for many new developments to proceed from the outset without surface facilities, using seabed compression to push gas directly to shore over long distances and doing so without the higher price tag of a platform or floater. In this regard, subsea gas compression is a major step for Statoil in realising its concept of the “subsea factory” where all production operations will be sited on the seabed, enabling developments to proceed in difficult environments such as the Arctic and very deepwater regions.

Advantages such as these, plus the fact that production will be accelerated and overall recovery increased by using subsea gas compression, have been driving operators such as Statoil and Shell, supported by a host of contractors and suppliers, to establish gas compression as a highly effective enabling technology in the industry’s subsea toolkit.



Aasgard world first

Statoil’s Aasgard field in the Norwegian Sea is currently on course to become the location for the world’s first subsea gas compression project to come onstream, with a startup date in the first quarter of 2015.

Aasgard, about 200 kilometres off mid-Norway, ranks among Norway’s largest offshore developments with more than 50 wells drilled through 16 seabed templates. The Aasgard A oil production vessel came into operation in 1999, followed by the Aasgard B semisubmersible gas platform a year later.

Production of gas and condensate from Aasgard and its two subsea satellite developments, Midgard and Mikkel, located in 250 to 325 metres of water about 40 to 50 kilometres from the surface platforms, has declined and towards the end of 2015 the fields’ reservoirs will have insufficient gas pressure to produce steadily. To maintain stable production, gas output pressure from the satellites must be boosted. In addition, a minimum gas flow is also necessary to avoid the accumulation of condensate, water and monoethylene glycol (MEG) liquids in the flowlines. MEG is used for hydrate inhibition and flows from Aasgard B through separate feed lines to be injected at the wellheads. The MEG returns with the wellfluids to Aasgard B where it is separated from the well stream and recycled through a regeneration plant for further use.

By installing subsea compression near to the satellite wellheads, the backpressure on the reservoirs will be decreased, enabling gas to flow at a high and stable rate and enhance overall recovery.

“The benefits for Aasgard of installing compressors on the seabed are significant,” Vintersto points out. “The project will extend the producing life of the fields by about 15 years and will add about 278 million barrels of oil equivalent to our production.”

In the summer of 2011, Statoil and its partners in Aasgard gave the green light for the NOK15 billion subsea compression project, with the ambitious target of the compressors being online four years later.

The subsea compression station, weighing about 4750 tonnes, will be sited near the satellite fields at a step-out distance of about 40 kilometres from the surface facilities. The giant modularised station, measuring 74 by 45 metres and 26 metres high, will house two compressor trains capable of boosting gas pressure by up to 50 bar and together delivering up to 21 million standard cubic metres per day of gas. Power to the station will come via umbilical cables from Aasgard A, with gas and condensate being transported to Aasgard B.

At the heart of each train will be a compact horizontal centrifugal 11.5 megawatt (MW) compressor built by MAN Turbo & Diesel. The compressors have active magnetic bearings, with the motor and compressor housed in the same casing, and the motor being cooled by the process gas from the second stage of the unit. In the initial phase of operation, the two machines will be operated in parallel.

The trains will each have a passive inlet cooler using seawater to lower the gas temperature coming from the fields to 10o to 16oC, after which the gas will pass through a vertical scrubber to remove liquids from the gas stream before it enters the compressors. A condensate pump will return the separated condensate liquids into the export gas line downstream of the compressors.

In passing through the compressors the gas will be heated up, requiring it to be cooled again to 50o to 60oC to protect the coating on the downstream export pipeline to Aasgard B, for which gas discharge temperature must be kept below 70oC. The compressor system also incorporates anti-surge lines, which will open in certain conditions of low gas flow and pressure to route gas back to the compressors to be recirculated through the machines.



Technology under test

Large gas compressors have something of a reputation for being temperamental in operation even on a platform topsides, hence the move to place them on the seabed and control them remotely is a bold one. A lengthy process of technology qualification and testing of a wide range of process, mechanical and electrical components, large and small and capable of long-term operation in deep water, has been necessary and is still ongoing.

“Between 2008 and 2011 we tested an 8MW pilot compressor,” says Ole Jorgen Johansen, Statoil project manager for the Aasgard subsea compression project. “This was run on hydrocarbon gas, condensate and water/MEG at K-Lab, Statoil’s test facility at the Kaarsto terminal where Aasgard gas comes onshore. The pilot unit has now been upgraded to an 11.5MW machine and further robustness and verification testing is scheduled to begin later this year.”

The testing he refers to will begin as soon as a new shallow-water test pit is fully commissioned at K-Lab, a large structure measuring 15 by 20 metres and 13 metres deep. The pilot compressor will be submerged in the pit and tested on an incoming feed of 17 million standard cubic metres per day of gas from Aasgard, in combination with condensate and water/MEG.

For the forthcoming tests, the upgraded pilot compressor is being integrated into a module at the Egersund fabrication yard of Aker Solutions as part of a NOK3.4 billion contract awarded to the contractor for the design and construction of the subsea compression system. Referring back to the “30-year subsea compression saga”, Aker Solutions can lay claim to having made one of the earliest technology pushes toward subsea gas compression in the mid-1980s through its heritage company Kvaerner, which led to the successful testing of the pilot Kvaerner Booster Station in the early 1990s.

The upgraded Aasgard pilot compressor will be equipped with more instrumentation than used in the first trials to provide a wider range of performance data.

After the pilot, next up for testing will be the actual ‘delivery units’ for Aasgard, the two compressors plus a full size spare, which will be kept onshore in normal operations. Each compressor module will weigh 236 tonnes and stand 10 metres high.

“The delivery units will be subjected to a different testing programme,” adds Johansen. “This will be a full functional test, allowing us to achieve important actions such as the fine tuning of the magnetic bearings.

“It is vital that we know the operating parameters for the machines and that they are in perfect working order before they go offshore.”

Gaining that confidence, and not only for the compressors, has been hard earned through a series of technology qualification programmes for the project, which have helped to mature technologies from competing suppliers. Among these, the project’s electrical system has presented no shortage of challenges, be they associated with the all-electric subsea control systems — FMC has supplied more than 60 electrical valve actuators to the project — the power supply cables, or the high voltage connectors and penetrators.

A case in point are the high voltage (HV) wet mateable connectors and penetrators developed by Deutsch for the project. These components are critical to maintaining a reliable supply of power to the compressor station, and must be insulated, waterproof and pressure resistant. There are three HV connectors for each compressor and six for each of the two condensate pumps. Additionally, there are nine HV penetrators for each compressor and pump. The largest HV connector is rated at 12 kilovolts and 1600 amps and is approximately 1.5 metres long and weighs 500 kilogrammes. The largest penetrator is rated for the same duty and is 0.6 metres long and 0.2 metres in diameter and weighs 320 kilogrammes.

“These large components represent technical achievements in their own right,” observes Vintersto. “Electrical components such as these will evolve in the coming years in a programme of continuous improvement, already witnessed over the period of the Aasgard project.”



Installation challenges

Power to the compressor station will be delivered from Aasgard A’s existing power generators through umbilical cables supplied by Nexans. Two combined dynamic umbilicals contain the electrical power cables for the compressors, pumps and control systems, along with a line carrying barrier fluid used in the pump units between motors and pumps, and fibre optic cables for control signals. The dynamic umbilicals will descend from Aasgard A onto the seabed, running about 1500 metres before being ‘spliced’, dividing into four static umbilicals on the seabed that will then travel 42 kilometres to the compressor station, two umbilicals per train.

Aker Solutions, under a separate NOK650 million contract, has carried out modifications on Aasgard A for supplying power. This includes ongoing work to bring the two combined umbilicals into the swivel stack on Aasgard A, and the construction of a new 800 tonne power module containing transformers and variable speed drives, supplied by ABB, for the compressors and pumps. The new module will be installed on the vessel in June this year by Saipem.

In due course, Saipem will also install the compressor station structure on the seabed — requiring very strict tolerances to ensure the machines will be level in operation — and the associated manifold station. The 965 tonne manifold, measuring 34 by 27 by 15 metres, will gather together four new 12 to 18 inch diameter flowlines delivering gas from Midgard and Mikkel en route to the compressor station.

After the gas has been compressed it will return to the manifold before being sent via two routes to Aasgard B. One route will be though a new 20 inch diameter pipeline, while the other will be through the existing 20 inch Midgard pipeline in operation now. The connection into the Midgard line was made in August 2012 in readiness for the eventual tie-in from the manifold.

The 10-day operation to make the connection is worthy of note as this involved the remote retrofit installation of a new 12 inch diameter T-piece into the line by ‘hot tapping’ into the main line while it was flowing gas at a pressure of 91 bar. The technique, which avoided the alternative of a lengthy production shutdown, is believed to be a world first by Statoil as it was carried out on a pipeline that was not prepared in advance (see story on page 28). A new suite of remotely controlled subsea machines, designed by Statoil for this purpose, has been brought onto the offshore stage and deployed for the Aasgard hot tap by Technip’s Skandi Arctic vessel. Technip has the contract for marine operations for the subsea compression project, covering the installation of control and power umbilicals, diverless tie-ins and connections to existing subsea infrastructure.

Technip will also install the modules into the compressor station, 11 of them in all.

“Marine installation and intervention operations are occupying us a great deal at present,” notes Vintersto. “In the early project stages we did not think enough about this. We must be able to disconnect, retrieve and replace components and modules weighing up to 300 tonnes, quickly and efficiently, and do so at any time of year in sea states with 4.5 metre significant wave heights.”

He sees specialist vessels, with large moonpools and tower cranes, evolving to meet these needs on future subsea compression projects, but as the pioneering project, Aasgard’s installation and intervention needs will be met for the next three to four years by Technip’s North Sea Giant construction vessel. Six modules, the larger ones, will be installed ‘over the side’, while the remaining five will pass through the vessel’s moonpool.

Looking ahead, Vintersto believes subsea compression will become the chosen solution for many projects — Gullfaks and Ormen Lange are already in the frame (see below), Statoil’s Snohvit field off northern Norway could be a likely candidate, and other operators are thought to have identified several prospects for the technology around the globe.

“Aasgard, Gullfaks and Ormen Lange are providing us with a lot of learning to feed into the next generation of subsea compression projects,” Vintersto concludes. “We already have three types of subsea compressors available and expect to see more suppliers come into the arena. These are exciting times for subsea compression. In Statoil we are not sitting on the fence.”



Gullfaks wet gas contender

Although Aasgard is scheduled to be the first subsea gas compression project to come onstream, another contender in the subsea compression race is hard on its heels.

In parallel with the Aasgard project, plans are progressing for installing subsea compression in the Gullfaks South subsea satellite fields off Norway in the North Sea, operated by Statoil with Petoro as partner. The partners are investing about NOK3 billion in the project with the goal of adding about 3 billion cubic metres of gas production from the Brent reservoir at Gullfaks South — 22 million barrels of oil equivalent — about 6% more than the current recovery estimate Statoil holds for the field.

“We expect the Gullfaks South subsea compression project to be in operation in the autumn of 2015, soon after Aasgard has started up,” says Vintersto.

Gullfaks South comprises several satellite fields tied back to the Gullfaks A and C platforms. A second phase of gas and condensate production from Gullfaks South began in 2001 through two four-slot templates (L and M) tied back to Gullfaks C, about 15 kilometres away. The plan is to install a two-train subsea compression station near to the templates in 135 metres of water, and boost gas recovery by an additional 10 million standard cubic metres per day.

While the overall concept of boosting gas output pressure to gain more production from the reservoir is the same as that for Aasgard — subsea compression is expected to extend the life of Gullfaks South to 2030 — there are some fundamental differences in approach for Gullfaks South.

Most notable among these is that the two compressors that will be used for Gullfaks are wet gas compressors that can handle a gas stream containing high percentages of incoming liquids — condensate and water. Although the compressors for Aasgard have been shown during testing to be capable of handling a considerable volume of liquids in the gas steam for short periods, the normal operating design is for dry gas, requiring scrubbers upstream of the compressors to take out the liquids.

“Wet gas compressors do not need this additional process stage,” explains Simon Davies, subsea technology project manager with Statoil. “Nor do they require the anti-surge system used to protect dry gas compressors, leading to a substantial reduction is the size and weight of the wet gas compressor station. But this difference may not be the case in future. We are already looking ahead to the next generation of subsea compressor concepts, wet and dry, which may operate within a much simplified process arrangement.”

The wet gas compressors selected for Gullfaks are being built by Framo Engineering in Norway. In May 2009, Statoil entered into an agreement with Framo for a technology qualification programme, aimed at developing a subsea wet gas compressor that could be in operation in the field by 2015. To this end, Framo has developed a compact contra-rotating gas compressor it claims can tolerate 100% liquid flow.

The company began development of contra-rotating compressors in 1987, and in 2000, Statoil and Framo began collaborating on the first vertical model, tested on dry gas at Kaarsto. When the Gullfaks South project was initiated in 2006-7, the two companies began developing the WGC4000 machine, which is now going through technology qualification for Gullfaks.

The WGC4000 is a 5MW multi-stage, contra-rotating, vertically mounted axial compressor, which for Gullfaks will be capable of boosting pressure by 30 bar. The inner and outer rotors spin in opposite directions at speeds between 1200rpm and 4500rpm. The contra-rotating design eliminates the need for diffusers on the outlet gas stream, as required for centrifugal compressors, thereby enabling a large number of axial compressor stages to be fitted into a short length — in this case, there are 21 stages, giving good inter-stage phase mixing. According to Framo, the design is very robust and incorporates many of the same elements the company has proven in its subsea pumps over many years of operation, for example the motors, seals, bearings and control systems.

In 2010-11 the unit was put though endurance testing submerged in a tank for more than 3300 hours in Framo’s purpose-built hydrocarbon test loop at Fusa near Bergen. The test loop was operated at flowrates up to 6000 cubic metres an hour using a gas of identical inlet composition to that at Gullfaks.

Following the success of the trials, Framo was awarded a NOK900 million EPC contract by Statoil last year to design and build the full-scale compression station. Detailed engineering is now under way and will lead to testing of the full-scale machines and all ancillary equipment at Framo’s new test facility at Horsoy near Bergen in 2014. The compressors themselves are compact, standing about eight metres high, which will have a positive impact on the size of the compression station on the seabed. For Gullfaks South the modularised station will weigh about 950 tonnes overall and measure 34 by 20 by 12 metres — much lighter and smaller than the 4500 tonnes Aasgard station. The compressors, each one driven by two contra-rotating 2.5MW electric motors, will weigh about 70 tonnes each, making their retrieval from the seabed station, along with the two associated 60 tonne upstream coolers, relatively straightforward to do for maintenance purposes using a standard intervention vessel.

In addition to the coolers, which are present to ensure well fluids temperature does not exceed the limits of downstream equipment, a well fluids mixer is required to mix the incoming multiphase fluids and provide equal feeds to the two compressors when working in parallel — the compressors can be operated in parallel or in series to deliver a range of flow and pressure outputs.

Power to the station will be supplied via a Nexans seabed power and control umbilical cable running from the Gullfaks C host platform from where the compressor station will be controlled. According to Framo, the relatively low rotational speed of the compressors helps with the design of the variable speed drives on Gullfaks C. The compressor design requires a barrier fluid between motor and compressor, for which a hydraulic power unit is required on the topsides to circulate the fluid to the compressors.

As part of a $70 million contract, Subsea 7 will install the subsea cable, along with the subsea compressor station, protection structure and pipeline tie-ins. Apply Sorco has the contract for topsides modifications on Gullfaks C, valued at NOK375 million. With so many apparent advantages, not least the ability to handle wet gas in a more compact and lighter machine, is the wet gas compressor likely to become the first choice for all subsea compression projects?

“It is a matter of scale,” explains Vintersto. “The Framo compressor is targeted at handling the gas volumes from smaller fields. You would need many compressor units in operation to have the capacity for a large gas field, which would decrease the economic benefits. This is another reason why we currently have three different subsea compressor designs being developed in parallel, as a range of compressor capabilities is needed to match different field applications.”



Ormen Lange looms large

When it comes to scale, the industry’s largest and arguably most ambitious subsea gas compression project remains under consideration for Norway’s Ormen Lange field, operated by Norske Shell.

Ormen Lange, Norway’s second largest gas field, came onstream in September 2007 and reached plateau production two years later, delivering about 70 million cubic metres of gas and 50,000 barrels of condensate daily to the Nyhamna processing plant in mid-Norway. Production from the field comes from 17 wells located about 120 kilometres offshore in water depths of 850 to 1100 metres, the deepest production wells in Europe and believed to be the world’s largest individual gas wells in terms of output. Critically for any proposed subsea gas compression project is the fact that there are no offshore surface facilities at Ormen Lange, with gas and condensate flowing directly to shore through two 30-inch diameter multiphase pipelines. Reservoir pressure is steadily falling and by about 2016 gas compression will be required to boost pressure and maintain the flow, helping production from the field to last until about 2035. Ormen Lange’s gas compression requirements are significantly more challenging than those of Aasgard or Gullfaks, as about 60 million standard cubic metres per day of gas must be raised in pressure from 80 bar to 140 bar.

As its base case for meeting this challenge, Shell has developed a design for a manned “slim” tension leg platform with a topsides weighing about 32,000 tonnes, which could be powered from shore or be assisted by gas turbines onboard, and would have gas compression installed.

As an alternative, about 10 years ago, Shell began to investigate the possibility of the entire compression requirement being met by subsea gas compressors, offering significant savings in capital and operating costs compared to a platform solution. If full subsea gas compression facilities were installed to meet this challenge, Shell foresaw that it would require 58MW of power to be transmitted through a subsea power cable at high voltage from shore, complex electrical control systems on the seabed, subsea gas/liquid/sand separation and condensate pumping, plus four hefty 12.5MW subsea compression trains, in total weighing about 8000 tonnes.

Shell has more recently changed its approach in part and has committed to installing some of the compression capacity onshore at Nyhamna where two 38MW compressors will be installed in 2014-16 as part of the expansion of Nyhamna facilities, in readiness for accepting gas from other new offshore developments, for example Aasta Hansteen. Gas pressure from Ormen Lange will be boosted by these new compressors. However, additional offshore gas compression in the field will still be required, either on a TLP or subsea, about 2020 — the installation of the large compressors at Nyhamna will enable Ormen Lange production to continue unimpeded in the meantime, giving more breathing space for the offshore compression decision.

To investigate the subsea compression option thoroughly, since 2011 a full-scale pilot subsea compression train, designed and built by Aker Solutions under a $160 million contract, has been undergoing a major qualification testing programme in a water-filled pit at Nyhamna. Statoil, as partner in the Ormen Lange development, is carrying out the testing programme on behalf of Shell, putting the pilot through its paces on “live” Ormen Lange gas and condensate.

The pilot consists of seven large full-scale subsea modules, grouped to form the process, control and high voltage power systems needed to support a single compression train, based on a 12.5MW compressor supplied by GE Oil & Gas. In a similar arrangement to the Aasgard subsea compression station, the incoming gas stream is cooled and liquids are taken out of the gas flow in a scrubber upstream of the compressor and returned to the downstream flow via a condensate pump.

However, a major difference from the Aasgard or Gullfaks South subsea compression stations is the need for large electrical power supply components to be incorporated into the compressor station on the seabed, as there is no surface platform on which to locate these for Ormen Lange.



Learning curve

The GE compressor is a compact five metre high vertical centrifugal machine that operates at 11,000 rpm, directly driven by an electric motor.

The innovative design of the compressor has reduced the number of moving parts and supporting systems, for example, the active bearings are magnetic and do not require a lubrication system, and need only low voltage power for their operation. The compressor and its high speed electric drive motor are housed in a single, hermetically sealed enclosure that is pressurised with a barrier system to keep the motor and compressor spaces separate, and also to provide clean operating conditions for the motor and bearings — these are cooled by process gas in a closed loop with an external seawater heat exchanger.

The Ormen Lange pilot modules are submerged in the 42 by 28 by 14 metre deep test pit, cut into the rock at Nyhamna. The test loop, simulating Ormen Lange’s well stream, has capacity to circulate 15,000 standard cubic metres per day of gas and 1800 standard cubic metres per day of condensate, generate operating pressures of 40 to 155 bar and supply 16MW of electrical power to the pilot, as well as creating liquid slugs and injecting fines (sand).

“Since introducing hydrocarbons into the pilot plant at the beginning of 2012, we have notched up over 3000 hours of compressor running to date, plus about 1000 hours on the condensate pump,” says Mathias Owe, Shell’s manager for the Ormen Lange subsea compression project. “We are now nearing the end of the final performance testing part of the qualification programme, which includes numerous different modes of testing.”

To date the pilot project has passed seven stage gates in Shell’s Technology Realisation Process. The eighth and final stage will hopefully approve the technology for deployment in the field. Further approval by the Ormen Lange licensees could be given in the second quarter of this year.

“This will provide the basis for the big decision about whether to go for subsea compression or compression on a TLP,” adds Owe. “But it’s not just about the technology, it’s about the economics too. We should know the final outcome by the end of 2013.”

The testing programme has served its purpose well, providing Shell and Statoil with valuable information that will influence the design of the new subsea compressor station for Ormen Lange, should this be the selected option.

For example, at the start of the project the large subsea electrical power components — notably transformers, switchgear and variable speed drives for the compressor and pump — were considered to be high risk items when operating remotely subsea. They have all functioned well and Shell intends to retain the same power components in a new station, although the arrangement of some items will change.

The internals for the variable speed drives are one case in point, where their original hexagonal shape will be changed to be more ladder-like for easier maintenance, and the circuit breakers another, with these large components being rearranged into a vertical rather than horizontal vessel.

The active magnetic bearings in the compressor will also be modified. The bearings, which are powered in operation, have a dedicated control system module located at the top of the compressor module that controls the rotational movement of the vertically-mounted rotor in several directions simultaneously, picking up vibrations at various points to adjust the bearings automatically and provide smooth running.

“The tests have demonstrated that more electrical power is required at the thrust bearing, which holds the rotor in the vertical position,” explains Owe. “This will be upgraded for the delivery units.”

About 30MW of electrical power would be required for the two-train offshore compressor station. This would be transmitted from Nyhamna at 90 kilovolts in a single power and control umbilical, and stepped down to 32 kilovolts in a subsea transformer at the compressor station before entering the circuit breakers. After the circuit breakers another transformer would step down the voltage further, to 6.6 kilovolts to run the compressors and to 2.5 kilovolts for the pumps.

The station would also incorporate two uninterrupted power supplies (UPS), essentially banks of rechargeable batteries, to act as emergency backup power. If power from shore were to fail, the active magnetic bearings would not function and could cause the compressor to experience a sudden hard landing onto its roller bearings — in these circumstances, the UPS would a

User

Become an Upstream member!

Membership includes a subscription to our weekly newspaper providing in-depth news from the energy industry, plus full-access to this site and its archives. Still not convinced? Try our free trial.

Already a member?

Login