Automatic history matching of discrete fracture model of tight gas reservoir based on EnKF method

Author of the article:LIAN Peiqing, ZHAO Huawei, WU Shuang, SHANG Xiaofei, DUAN Taizhong

Author's Workplace：Sinopec Petroleum Exploration and Development Research Institute, Beijing, 100083, China

Key Words：Tight gas reservoir； Discrete fracture; Ensemble Kalman Filter (EnKF); Automatic history matching

Abstract：Discrete fracture model is a commonly used numerical simulation model to describe fracture water flooding in recent years. Due to many geological and fluid parameters involved, artificially modifying these parameters requires a lot of work, yet is inefficient. The automatic history matching method is an effective method to improve the matching accuracy of discrete fracture models. Taking Xujiahe tight gas reservoir in Sichuan province as an example, this paper establishes an automatic history matching method for discrete fracture model of tight gas reservoir based on Ensemble Kalman Filter (EnKF). The fractures are divided into three levels:large, medium and small scales. Discrete fractures model are used for large fractures, and equivalent methods are used for small and medium fractures to establish a geological model. The EnKF method is used for historical matching of gas and water production data of X2 block, Xujiahe formation. By automatically adjusting the water energy of each sand zone and the porosity and permeability around the production wells, the overall matching rate reaches 86%, which provides an effective reference for numerical simulation of fractured tight gas reservoirs.