As the search for oil moves into deeper waters and harsher environments production is moving onto the seabed.
Offshore oil and gas production has seen a significant trend in recent years: the elimination of offshore surface facilities, with the entire production system for some of the more advanced fields located on the seabed and connected back to a terminal on the shore.
The Ormen Lange is a good example of this approach, and is significant because it will supply 20 per cent of the UK's gas consumption in coming years. Remote control plays a vital role, since the subsea hardware is out of reach for manual intervention, but closed loop control is playing an increasing role as the projects become more sophisticated.
The Ormen Lange field came onstream in September 2007, with peak production of 70 million cubic metres of gas and 50,000 barrels of condensate per day. Natural reservoir pressure should be sufficient to drive the well fluids 120km to the shore until the middle of the next decade, but offshore gas compression will be required thereafter. The field lies off mid-Norway, in water depths of 850-1,100m. The first stage of development involves two subsea 'manifolds' with eight large-bore wells drilled from each manifold.
Each of these is connected to the terminal where the fluids are processed. In addition to the main 30in product pipelines, there are smaller 6in pipes to carry mono ethylene glycol (MEG). Large quantities of MEG are injected at each wellhead because the sub-zero temperatures in deepwater tend to promote hydrate formation (a snow-like substance that can block pipes).
The MEG is pumped out from the shore and returns with the produced gas to the terminal, where it is recovered and recycled. This is a costly process and a system has been provided to regulate the injection rate to prevent excess usage. In this system, a Roxar wet gas meter and Sentech water-fraction meter are installed downstream on each well to measure flow rate, water fraction and other characteristics. The adoption of two meters provides redundancy.
The meters have been incorporated in a retrievable module, along with an actuated valve to control MEG injection. Data is conveyed to the shore via optical fibre cables within the overall umbilical, with control signals generated in the control room. The ultimate aim is to automate this process, with closed loop control generated by computers using simulation software.
Separate umbilicals run to each of the eight-well manifolds, with a link between the two manifolds to provide a 'triangulated' link for greater security and reliability. Each umbilical incorporates four high-voltage power cables, 64 fibre-optic lines and three 0.75in-diameter hydraulic lines, along with other services, within a 125mm-diameter sheath.
The production control system is an optical/electronic multiplexed system; the primary communication protocol is TCP/IP. Pressurised hydraulic fluid is transmitted through the three pipes within the umbilical, and stored in accumulators. A retrievable subsea control module is provided on each well to transmit data to the shore and incorporate electro-hydraulic components to actuate valves in response to signals from the control room.
In the future, the industry would like to replace hydraulic actuators with electrical devices to increase reliability and permit longer step-out distances. It would also reduce maintenance, since hydraulic fluid can become contaminated. Such a system would also permit condition monitoring. Furthermore, it would eliminate the need for the hydraulic lines within the umbilical, and reduce hydraulic fluid consumption, thereby reducing costs. There would also be environmental benefits, as the risk of hydraulic fluids leaks would be reduced.
The first step towards this objective took place in 2001-2002 when 16 all-electric choke valves were installed on a subsea manifold lying in the StatoilHydro-operated Statfjord field. The chokes were supplied by FMC Technologies and are used to regulate the production flow from individual wells. The technology proved reliable, and this led to more widespread use of electrical actuators.
As a result, in May 2005, FMC was awarded a further order by StatoilHydro for conversion of a manifold in the Norne Field for all-electric operation. The manifold, which lay in 380m of water 200km off the Norwegian coast, incorporated eight electric gate valves and one pig valve. Speaking at the time of the order, Peter D Kinnear, executive vice president FMC Technologies, said: "The Norne system will be the first conversion of a production manifold to all-electric operation."
Success with the all-electric manifold set the scene for the use of electrical actuators for 'Xmas trees'. The Xmas tree is installed at the head of each well, and incorporates several independent valves to shut-off flow from the reservoir. Multiple valves ensure shut-off integrity, and provide back-up if one valve should fail.
In addition to the Xmas tree a 'sub-surface safety' valve will be provided within the well bore, several hundred metres below the seabed. This valve provides the last line of defence in an emergency. Failure of this system could have catastrophic consequences, with the release of hydrocarbons into the ocean.
Development of all-electric subsea tree systems has been underway for some years, and several companies are able to supply these systems today, including Cameron and FMC Technologies. But development of these safety-critical systems presented an enormous challenge, not least because they were subject to international standards and rigorous scrutiny by regulators. Inherent fail-safe design is a crucial factor.
The technology was first tested in 2004 on the BP Magnus field in the North Sea. A CameronDC tree was installed on a dummy well in 600ft of water, close to the base of the Magnus platform. The tree was controlled by a six-mile long 1in diameter co-axial cable reel, with the valves automatically cycled on a daily basis for about four to six months. The CameronDC system is inherently fail-safe with mechanical spring closure on power failure.
More recently, the French oil company Total undertook the first field development with an all-electric subsea tree system, using two CameronDC trees, on its K5F satellite field in the Dutch sector. The K5F field lies approximately 115km north-west of Den Helder on the Dutch Coast, and came into production during September 2008. The field will produce up to 90 million standard cubic feet per day and is connected by a 10km-long pipeline to the K6N platform. There is potential to install two further Xmas trees with the existing manifold.
When announcing the start of production, Total said: "This step change in subsea technology will bring increased system reliability and enhanced environmental performance. Furthermore, it will add to Total's capability of bringing new production from deepwater fields, including frontier areas of the North Sea."
With growing confidence in subsea production, the industry has spent enormous sums on R&D to expand the range of processes that can be performed on the seabed. Much of this work was aimed at developing systems to separate oil, gas and water on the seabed, so that the individual fluid fractions could be pumped or compressed, as appropriate. Advanced automation is the key to the success of these systems.
The first subsea separation system was actually implemented as long ago as 1989, when a prototype was installed in the Argyll Field in the North Sea, but this trial was halted when seawater leaked into the control pod. There was then a lull of a few years before the Troll Pilot was installed in 340m of water, in the Norwegian sector, by StatoilHydro in May 2000. In this subsea system, the oil, gas and water fractions are separated with water pumped back into a rock formation under the seabed.
One challenge was to measure the water/oil and gas/oil interface in Troll Pilot's gravity separation vessel. This challenge was overcome with the development of novel inductive and nucleonic level sensors. The non-intrusive instruments were installed in pockets, so they could be retrieved to the surface for maintenance without breaching the integrity of the pressure vessel. These measurements were used in a closed-loop control system, with water/oil interface level controlled by regulating the speed of the injection pump.
Subsea sensors and controllers were linked by Fieldbus, with data communication to the host platform done via fibre-optic lines within the umbilical. The umbilical also carried high-voltage power for the water injection pump. Unfortunately, the high power connector (on the subsea unit) suffered a fault soon after start-up, and it was necessary to undertake a $15m project to repair it. Operation has been reliable since completion of this work in August 2001.
The Troll Pilot was followed by Tordis, which was described as the first full-scale field development with subsea separation, boosting and injection. The subsea unit was installed in August 2007. The basic principle of subsea unit was to separate the oil, gas and water, then to inject the water back under the seabed. The oil and gas were remixed and boosted to the host Gullfaks platform by a pump. The benefit is that a smaller volume of fluid needed to be transported back to the host platform, so more oil could be produced.
Two Roxar multiphase meters were incorporated in the Tordis subsea unit to measure the composition of the wellflow and to enable separator settings to be adjusted. The development of subsea multiphase meters was one of the R&D success stories of the 1990s. They enable the quantity of oil, gas and water to be measured within the well stream of a subsea well.
While the base plan for Ormen Lange called for a gas compression platform to be installed by 2015 to cater for falling reservoir pressure, there is now increasing confidence that an alternative subsea compression scheme can be adopted. This technology has yet to be field proven, so the new technology will need to be verified. Submerged full-scale tests are thus to take place on a pilot over a period of two years from 2009 to qualify the equipment. The trial will be carried out in a basin with realistic well stream conditions. If successful, the four train subsea station will be installed, close to the subsea wells, in 860m of water, during 2015.
"The partners have concluded that the technology is sufficiently mature for us to carry out a full-scale test. It is rare that a development project has this much funding available for qualifying new technology. But then again, this is a solution which will mean a great step forward for the oil industry," said Tom Røtjer, project director for Ormen Lange, when the trial was announced.
"If we succeed with this project, we will be able to replace a 25,000t platform with a subsea installation weighing around 3,500t.
"The development costs for future offshore compression for the Ormen Lange field will potentially be reduced by 50 per cent, which means a saving of many billion Norwegian krone (kr). In addition, we will be able to halve operating costs over a 20-year period."
The partners have agreed on a system that involves separation, surge controls and liquid boosting. After separation, the gas will be compressed by a centrifugal compressor and condensate will be pumped into the export gas pipeline. The modularised equipment will be designed for easy retrieval from the manifold base.
To reduce maintenance, the entire subsea system will be 'oil-less' by the use of active magnetic bearings, and high speed electric drive to eliminate the need for a gearbox.
The long step-out electrical power supply system will present a major challenge. The overall power requirement is expected to be around 52MW with transmission voltage from the shore of 145kV, which is reduced and converted on arrival at the subsea station to suit the variable speed drives and controls. Aker Kvaerner Subsea AS was awarded kr850m for the full-size 12MW subsea compression station pilot. Vetco Aibel AS was awarded a kr100m contract for the long step-out power supply pilot. Both the pilots will be subject to the submerged test.
Subsea compression is also being considered for the Ãsgard field. StatoilHydro awarded FMC Technologies a $35m contract to supply a control system for this project in December 2007. The award is part of a major programme run by StatoilHydro to develop and qualify all critical components for an alternative subsea gas compression system.
FMC Technologies says that its scope of supply includes the qualification of critical control system components, including a subsea active magnetic bearing control system for the compressor (to be performed in cooperation with Waukesha Bearings), and a subsea anti-surge actuator and valve. There are also anti-surge control features added to Ãsgard control systems, and development and implementation of new sensor technologies that are vital to the subsea gas compression application. A control system test is also part of the scope of work.
"We have a strong history of subsea technological successes," says John Gremp, executive vice president of FMC Technologies, and the development of these components is another very important milestone in addressing future subsea compression needs in the industry. We are excited to assist StatoilHydro in this next step of subsea technology development."