Implementation of a Total-System Production-Optimization Model in a Complex Gas-Lifted Offshore Operation

Abstract

Summary A total-system production-optimization model has been implemented in a complex gas lifted offshore operation, resulting in production gains and operating-cost reductions. Whereas previous optimization models considered only the wells and production-gathering network, the new model is able to consider the combined performance of the total system, including downhole well configurations, the complex production-gathering and lift-gas-distribution pipeline networks, separators, compressors, and pumps. The model is applicable to most gas lifted fields and will be particularly beneficial when applied to those with complex production systems, and those where compressors are a constraint on total-system performance. The output from the optimization model principally comprises recommended values for individual-well gas lift injection rates, separator pressures, compressor discharge pressures, and compressor use. Field results are presented in this paper to demonstrate how implementing the optimizer's recommendations in the field resulted in economic benefits through increased production and reduced operating costs. Also described is how the model allows field operations engineers to reoptimize field control parameters on a more frequent basis and with less manpower than previously. The successful implementation of a complex model with such a broad scope is as dependent on the implementation process as it is on the technology. Therefore, in addition to describing the details of the model itself, this paper will cover the issues that arose during the implementation and how they were resolved. These include the level of manpower and support required, project organization and execution, and the processes required to sustain the benefits after the initial optimization gains have been realized. Introduction Dubai Petroleum Company (DPC) has implemented a production-optimization tool that has yielded production gains and operating-cost reductions. The field-otpimization software is used to model the complex production networks associated with the gas lifted fields, including the downhole well configurations and the surface-facility components such as gas-compressor trains, pipelines, and surface pumps. Key benefits realized from all fields were a 3% total production increase, a 4% reduction in lift-gas requirements, and a 3% reduction in operating costs. Field operations support was critical to the project's success by tracking the operational parameters continuously throughout implementation to validate the recommendations and results. The project was planned to be executed in three phases, including a pilot study to assess the value of a full-field model and to identify and resolve implementation challenges. The full-field model was implemented during 2003 and produced several key learnings about the level of manpower and support required, the importance of accurate well-model tuning, and the value that a detailed compressor model can add to a system highly dependent on compressor efficiency. Challenges associated with the gas lift control systems, which are nearing obsolescence, were also identified and created a need for alternative strategies depending on the length of time that the gas lift rate reallocation would be in effect. The full-field optimization process uses an integrated approach to address operational challenges. A team of engineers and operations personnel now manages events proactively on the basis of a well-defined strategy. The optimization model has allowed gas lift reallocations to be performed on a more frequent basis and with less manpower. On the basis of these reallocations, production increases have been realized and the fields are currently operating at the historically lowest separator pressures. Offline studies have been performed to recommend process-equipment modifications and justify major equipment overhauls. The integrated network model has also been used as a predictive tool to forecast the impact of ambient conditions and scheduled maintenance on production rates. The results are being monitored currently to determine the value of adding a fully automated interface to the system-model software package.