Well Placement Optimization with an Efficient Hybrid Algorithm to Enhance Oil Recovery
H. Chen, Q. Feng, X. Zhang, S. Wang, W. Zhou and C. Liu
Event name: 80th EAGE Conference and Exhibition 2018
Session: Poster: Well Performance
Publication date: 11 June 2018
Info: Extended abstract, PDF ( 545.4Kb )
Price: € 20
We propose an efficient approach to well placement optimization with meta-optimized hybrid global and local algorithm. First, we take advantage of Cat Swarm Optimization (CSO) algorithm, which has good exploration ability in the global search space, and Mesh Adaptive Directive Search (MADS) algorithm, which can provide efficient local exploration, to establish a hybrid algorithm. Then we use meta-optimization method to determine the proper controlling parameters for this new hybrid algorithm. Our proposed optimization approach is applied in a semi-synthetic model. Besides, we compare the optimization performance of the proposed optimization approach with that of CSO, MADS, hybrid optimization algorithm with specific controlling parameters. Results indicate that hybrid algorithm shows better exploration ability in both global and local space than stand-alone CSO and MADS. Our proposed algorithm outperforms hybrid algorithm. It demonstrates that controlling parameters have significant impact on optimization efficiency of hybrid algorithm. With the meta-optimization method, optimization performance of hybrid optimization algorithm is improved. The proposed approach can serve as an efficient tool to determine optimal well placement to improve oil recovery under sustained low oil price. Besides, it could be useful in other petroleum engineering problems such as well control optimization and fracture parameters optimization.