The BP King subsea pumps were commissioned in November 2007 in record-breaking water depth of 5430 ft (1700 m) offshore Louisiana in Gulf of Mexico (GoM). Two twin-screw multiphase pumps were installed with a 17-mile (27 km) tieback to the Marlin Tension Leg Platform (TLP). Besides the technical challenges that have been overcome in this record breaking project, installation and commissioning under these conditions are a major achievement. This paper gives a presentation of the system evaluation ahead of launching the project and the various engineering tasks during design of the system. The complete pump system with all necessary topside and subsea components is presented and discussed. Attention is also given to training, pre-deployment work, installation and commissioning of the pumps. This joint BP America Inc. (BP) and Aker Solutions ASA paper also describes the operational experience and methods for continuous monitoring of the various sub-systems at the pump station. Introduction BP King is an oil field located in the Gulf of Mexico (GoM). It consists of three wells, King D5 and D6, producing since 2001 and King West, D3, producing since 2003. The wells are located in water depths ranging from 5000 to 5430 ft (1520 - 1655m). All three wells are 100% owned by BP and are produced back to the Marlin TLP which is also operated by BP. The field is located approx. 100 miles (160 km) offshore the coast of Louisiana in the GoM. The Marlin Tension Leg Platform is located approximately 86 miles SSE of Venice, Louisiana in Viosca Knoll Block 915 at a water depth of 3245 feet. There are a number of fields that are tied back to the TLP and one of the longest of those is the King Field Complex. The wells are connected via rigid jumpers to one of the two flowlines going back to the Marlin TLP approximately 17 miles (27 km) away. As pressure in the reservoir has been declining, BP initiated studies of various methods of keeping up the production and increasing the field life. A structured approach was applied throughout the project period based on BP's project execution stage gate process:Appraise - Determine the feasibility of the project and align with business strategySelect - Select the preferred project option and establish execution plan and fundingDefine - Finalize project scope, cost, schedule and execution plan. Complete contract negotiations with key suppliersExecute - Procure and install equipment in alignment with scope, cost and schedule targets.

Subsea boosting of hydrocarbon flow, either directly from the well or from a subsea separator, will typically result in the need of handling some free gas. Pumping moderate gas volume fractions (GVF) in combination with high pressure increase, is typical for boosting flow from deep water wells and represent a challenge in pump design. Aker Solutions has developed such a pump, called a HybridBooster. This paper presents the design process of such a multistage pump from CFD-simulations to the complete full scaled testing. The hybrid pump design is flexible and suitable for a wide range of flow rates and GVF's due to the many stacking possibilities of impellers with different characteristics. This allows for making a pump with the best possible performance and efficiency. The two-phase testing was done with a mixture of water and air at Aker Solutions facility in Tranby. The pump generates a differential pressure in the range of 150 to 240 bar, depending on the rotational speed and flow rate, at 20% GVF with a suction pressure of 12 bar. The pump design is based on a combination of mixed-flow impellers and radial impellers. Test results show that the mixed-flow impeller technology have great potential for moderate GVF at all flow rates and high GVF capability at best efficiency point and above. The strength of the mixed-flow impeller is the internal remixing of gas and liquid that strongly reduces the risk of gas blocking in the flow channel. The behavior of the pump during testing was stable, granting easy pump control. Introduction The Hybrid Pump Project is a Joint Industry Project supported by Demo 2000 The Research Council of Norway and the following companies: Aker Solutions, Statoil ASA, Total E&P Norway, Nexen Exploration Norway AS, ExxonMobil Upstream Research Company USA. The large variety of oil field characteristics demands a range of tooling for a successful production. One of these tools is the subsea HybridBooster. Unlike a conventional single phase (liquid) pump the hybrid pump is able to pump a liquid/gas mixture with a high pressure increase. The HybridBooster will be a well-suited and cost and production-efficient solution in a number of applications. Typical applications are:Pumping liquid from a gas/liquid separator where a risk for gas carry-under existPumping hydrocarbons from fields producing mainly liquid, but with risk of gas break-through over time To meet the challenging requirements of handling an increased fraction of gas combined with high pressure increase a Hybrid pump development project was established. The key specifications for the subsea Hybrid Pump development are:Pressure increase - target: 200 bar, min. 100 barGas tolerance, base case - target: 20% GVFGas tolerance, advanced case - target: >30% GVFInlet Pressure - > 10 bar, < 50 barFlow - within power limit of 2,5 MW The major design criterion was to design a gas tolerant multi-stage pump that could handle 20% GVF at suction condition, with stable operating conditions and predictable behavior.

For subsea pumps, regular maintenance is required (in the order of every 5 years or so). Being able to plan this intervention to be "just in time" will reduce downtime, and optimize production enhancement. Being able to state with certainty whether a pump e.g. can be allowed to run one extra year (i.e. 6 years instead of 5) gives reduced downtime. Such techniques have been applied to topside systems for a long time, and are now being extended to subsea systems. A data monitoring system with algorithms for monitoring certain Key Performance Indicators for a pumping system has been developed and applied, and is used to determine "time to service". Three subsea pumping systems are used as real examples, with results from system testing and operation. Subsea data networks with "self-healing" capabilities are briefly described.

For subsea pumps, regular maintenance is required (in the order of every 5 years or so). Being able to plan this intervention to be "just in time" will reduce downtime, and optimize production enhancement. Being able to state with certainty whether a pump e.g. can be allowed to run one extra year (i.e. 6 years instead of 5) gives reduced downtime. Such techniques have been applied to topside systems for a long time, and are now being extended to subsea systems. A data monitoring system with algorithms for monitoring certain Key Performance Indicators for a pumping system has been developed and applied, and is used to determine "time to service". Three subsea pumping systems are used as real examples, with results from system testing and operation. Subsea data networks with "self-healing" capabilities are briefly described.

No abstract