Optimization of Gas Recycling in a Gas-condensate Reservoir Using Genetic Algorithm Based on Proxy Model

Please enable javascript in your browser.

Bagheri, Mohammad Amin | 2014

894 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 46270 (06)
University: Sharif University of Technology
Department: Chemical and Petroleum Engineering
Advisor(s): Goodarznia, Iraj; Masihi, Mohsen
Abstract:
In gas-condensate reservoirs, when pressure reduces to less than dew point pressure, condensate will form out of the gas phase. Gas recycling is one of the most common methods to enhance the production of gas-condensate reservoirs.The purpose of this thesis is to optimize gas recycling process in a gas condensate reservoir in order to produce the accumulated condesate in the reservoir and maximize reservoir economic efficiency.The parameters that need optimization are:
• Injection gas ratio
• Injection gas allocation amongst injection wells
• Bottomhole pressure of production wells
Genetic algorithm was considered as optimization method and Proxy model is used in order to decrease optimization process duration. Proxy models studied in this research are:
• Artificial neural networks
• Quadratic models
Compared to quadratic models, artificial neural networks showed superior performance.After execution of genetic algorithm on artificial neural networks, optimum problem parameters were specified.
It seems that using genetic algorithm in conjugation with proxy model was a proper choice in order to optimize dry gas recycling process in the studied reservoir
Keywords:
Optimization ; Genetic Algorithm ; Artificial Neural Network ; Proxy ; Gas Condensate Reservoirs

No TOC