A mixed integer optimization strategy for oil and gas production planning

Dimitrios Gerogiorgis, Efstratios N. Pistikopoulos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Oil and gas production is the cornerstone of the modern petrochemical industry, and its upstream as well as downstream processing provides many challenges to the process modeling, optimization and control areas. Mixed-integer optimization is a research field with a strong implementation record, having already been used to solve a wide spectrum of crude oil production, transport, distribution, planning and scheduling problems. Production optimization challenges are however perplexed by multiphase flow of oil, gas and water in the
sub-surface circuits: the respective elements (reservoirs, wells) induce complexity in oil and gas transport which can only be handled suboptimally by use of linearized approximations of true pressure-flowrate curves. This paper addresses the problem of oil production maximization from a particular oilfield with several oil wells, all connected to one production platform and operating assisted by gas injection (secondary extraction). The proposed approach explicitly takes into account multiphase flow (based on a previously presented model) and relies on an MINLP model formulation toward calculating: (a) the operation (or shutting-in) of each well, (b) the volumetric flows of gas injection required in order to operate open production wells in gas-lift mode. An improved oil production optimum has been obtained for a case study considering a set of 6 gas-lift wells. This MINLP model can also be used for multiperiod optimization under additional cost and price constraints.
Original languageEnglish
Title of host publicationFoundations of Computer-Aided Process Operations (FOCAPO 2008)
Publication statusPublished - 2008
EventFOCAPO 2008 - Foundations of Computer Aided Process Operations - Boston, MA, United States
Duration: 29 Jun 20082 Jul 2009

Conference

ConferenceFOCAPO 2008 - Foundations of Computer Aided Process Operations
Country/TerritoryUnited States
CityBoston, MA
Period29/06/082/07/09

Fingerprint

Dive into the research topics of 'A mixed integer optimization strategy for oil and gas production planning'. Together they form a unique fingerprint.

Cite this