Well pattern optimization based on stosag algorithm

Abstract

The well pattern optimization in the oilfield is challenging and intricate work due to the heterogeneity of the permeability and viscosity. Traditionally, the well pattern optimization is conducted by comparing the results of several plans manually designed by the reservoir engineer, which is difficult to obtain the optimal well pattern. To address these challenges, a framework that integrates a reservoir simulator into the StoSAG algorithm is proposed. The well pattern operators proposed by Onwunalu and Durlofsy are applied to obtain the variations of the well pattern and used as the optimization variables. During the framework, the optimization variables are continuously adjusted by the StoSAG algorithm in order to obtain the optimal one which maximize the objective function value. The framework is applied to a synthetic reservoir. The results show that the StoSAG algorithm can be successfully applied in the well pattern optimization and remarkably improve the development effect. This method can be widely used in new oilfield development plan and offer reference for well pattern deployment.