07121-DW1901

Goal
The goal of this project is to develop and validate a Subsea Processing Simulator capable of predicting the performance of subsea processing systems over the range of conditions and fluid compositions found in the Gulf of Mexico. The intention is to provide a simulator that can be used as an industry standard to predict the in situ performance of compact separation components and systems over their productive field life.

Performers
General Electric (GE) Global Research, Niskayuna, NY 12309
General Electric (GE) VetcoGray, Sandvika, Norway

Background
As existing oil and gas fields become depleted and energy demand continues to rise, hydrocarbon production is moving toward increasingly challenging environments, including deep and ultra-deep water. For this reason, subsea processing systems that can operate at extreme depths and pressures (up to 3000 m and > 300 bar) are becoming increasingly important. On-site, sub-sea processing and separation of multi-phase flow streams can increase production rates, extend field life, and make otherwise marginal fields economically viable.

Despite assertions by equipment suppliers that compact subsea processing systems are ready to deploy, operating engineers remain less certain of that readiness and have particular concerns about separator performance under harsh conditions. The objective of this project is to develop a physics-based subsea processing simulator that can be used by equipment suppliers and facility engineers to make reliable predictions of sub-sea separator performance over a specified range of operating conditions.

The simulator developed in this project will be based on robust analytical models for compact separation devices operating in a subsea multi-phase flow environment. The project will be carried out by first compiling a library of analytical models of separator components; then developing an integrated processing simulator; and finally by laboratory testing to validate the simulator and its component analytical models.

The project will be carried out by GE Global Research and GE VetcoGray. The teaming arrangement is intended to combine GE VetcoGray’s experience with subsea processing and GE Global Research’s experience with developing and testing numerical simulators. The project team will develop a simulation model with four tiers: 1) component model library; 2) separator; 3) separation system; and 4) statistical performance. The simulator will be validated at the component and simulator levels in an existing GE multiphase flow test loop that will be optimized for this project.

Deliverables for this project will include: 1) a preliminary version of the Subsea Processing Simulator; 2) a report on the laboratory testing plan; and 3) a report on the simulator testing results. The Subsea Processing Simulator will be comprised of the simulator code and all supporting documentation on its theoretical basis and recommended methodologies.

Potential Impacts
This project is intended to result in a simulator that can be used by industry to predict with confidence the performance of subsea processing systems and components prior to their deployment. This will allow for quicker and more widespread deployment of subsea processing systems, in particular in deep water environments. Currently, the technology exists to perform hydrocarbon processing, separating, compressing, and pumping on the sea floor, but the technology is under-utilized because operators lack technical certainty concerning system designs and outcomes.

Confidence in subsea processing will be developed by allowing manufacturers through simulation to optimize their system designs and determine benefits to operators. Simulation tools will also allow operators to optimize their practices with respect to operation of the subsea processing systems as well as other controls that affect oil and gas production, which will minimize production risks.

Subsea processing systems have the potential to bring deep and ultra-deep oil and gas production to market faster while extending the life of existing fields, particularly in harsh environments. In addition, subsea separation and pumping can cut topside facility costs and allow for development of marginal fields using existing infrastructure.

In the near term, more widespread utilization of subsea processing systems is likely to result in increased production in deep and ultra-deep fields in the Gulf of Mexico. Over the long term, acceptance and deployment of subsea processing systems should bring more domestic oil and gas production to market, while extending field life and making marginal fields economically viable. Increased domestic oil and gas production will result in increased tax revenues, royalties, and regional economic benefits.

Accomplishments
The alpha version SubSea Simulator prototype of a Horizontal 3 Phase Separator has been coded in Matlab to understand the unit creation, flow of information across units and solution of a networked set of units. The Matlab code played a vital role in setting the structure & design of the NPSS/C++ based full-up Simulator that has been developed in parallel. Within the Simulator, the system is composed of instanced units from a unit library. The units are connected to create a flow network of nodes and links which communicate the relevant stream information. A solver iterates as necessary to solve the network within tolerance. The entire Simulator is made up of name tag and pointer references to minimize data overload and permit accessing and sharing of data across units. This is essential to keep the complexity of the system in check as the architecture is expanded to model increasingly intricate systems. The model has the handles/hooks to incorporate time dependent physics terms such as inertia and latencies to permit a transient simulation.

A preprocessor and postprocessor have been developed to implement statistical analyses using the Simulator. First, a preprocessor allows the user to create a set of run cases to be simulated. These cases can consist of any combination: A) correlated statistically distributed inputs, B) grid-wise distributed inputs, and C) design / operation / model tuning parameters. All cases are then executed by the Simulator with all results available for post processing. The postprocessor allows analysis of the resulting data sets respect to the various inputs, including the expected performance of the system being modeled given the statistical likelihood of the various cases occurring, and the operability envelope with respect to one or more user specified criteria. The statistical toolbox allows determination of the effects of various upstream conditions and design parameters and facilitates the development and refinement of the various unit models. On the full-up NPSS/C++ version the GRC team successfully developed the capability to perform flashing of thermodynamic properties from a compositional fluid description using a commercial package available for this purpose. They have also demonstrated the ability to embed a simulation within HYSYS and solve the system.

Current Status
Work is nearing completion on most aspects of the Simulator development. The key tasks to be undertaken center on preparing the Simulator for experimental validation and rolling the Matlab models into the NPSS/C++ version. Key points are outlined below.

Design and Develop Integrated Subsea Processing Simulator. The project team is developing a library of analytical, physical, and chemical models that are pertinent for simulating subsea processing. The individual component models are structured to allow more complexity to be added, as needed. The Simulator framework utilizes four object levels: 1) separator component models such as inlets, cyclones, demisters, weirs, volumes, etc.; 2) separators that are constructed of groupings of the components; 3) subsea processing systems that are composed of one or more separators, along with any valves, pipes, fittings, autonomous control systems, and pumps; and 4) solver wrappers that exercise subsea processing system numerical models through various design variations and multi-phase flow cases and perform statistical analyses to characterize system performance, support robust design development, and calculate probabilities of system failure.

Design and Perform Laboratory Testing. Laboratory testing will be performed at GE VetcoGray’s multi-phase flow test facility and will be designed to validate the component analytical models and overall Simulator performance. Tests will be carried out in the laboratory to reproduce both steady state and dynamic flow conditions. A key element will be development of a conceptual scheme for up-scaling the lab-scale test results to a full-scale subsea operation. The GE project team will analyze and report on the laboratory testing results and provide a recommendation on future work needed to improve the analytical models and the Subsea Processing Simulator.

Technology Transfer. The RPSEA quarterly meetings are the key mechanism for technology transfer. Overviews are presented at the Technical Advisory Committee meetings and detailed reviews are presented at the Working Group meetings. In addition, GE will hold training for operators interested in learning to use the Subsea Processing Simulator.

Project Start: December 3, 2008
Project End: June 2, 2010

DOE Contribution: $1,200,000
Performer Contribution: $311,448

Contact Information:
RPSEA – Jim Chitwood (jchitwood@rpsea.org or 713-372-2820)
NETL - Jay Jikich (Sinisha.Jikich@netl.doe.gov or 304-285-4320)
GE – Chris Wolfe (wolfe@research.ge.com or 518-387-5106)

Additional Information:

Final Project Report [PDF-3.64MB] July, 2011

Article in E&P Focus [PDF] June, 2009