A Gradient-Accelerated Two-Stage Multi-Objective Optimization Algorithm for WAG Injection

Abstract

Abstract Summary The multi-objective optimization of CO2-water-alternating-gas (CO2-WAG) injection identifies the optimal well control scheme that co-optimizes NPV and CO2 storage. Considering the expensive computational cost of the compositional simulation, a popular approach is to use a surrogate model to replace the costly numerical simulation in the optimization process. The data-driven surrogate model, on the other hand, has poor generalization, and each round of optimization requires the addition of sample points to calibrate the surrogate model, which significantly increases the cost and time required for optimization. In this paper, a two-stage optimization strategy is proposed to address the above issues. The strategy includes two stages. The first stage performs the surrogate-assisted single-objective optimizations of each considered objective. It is designed to find the endpoints of the Pareto front that connects all solutions of the multi-objective optimization, this preparation stage not only provides important search directions for the subsequent multi-objective optimization, but also improves the accuracy of the surrogate model near the Pareto front. The second stage is the surrogate-assisted multi-objective optimization, which aims to find all the solutions along the Pareto front based on the Pareto endpoints obtained from the first stage. In addition, this study successfully combines the gradient of the objective functions with the meta-heuristic algorithms during the multi-objective optimization, which ensures a faster convergence to the global optimum. The proposed multi-objective optimization algorithm shows faster convergence than the conventional optimization methods when applied to the three classical multi-objective optimization problems. Finally, a comparison with the conventional multi-objective optimization is conducted based on one test function and one benchmark reservoir simulation model to verify the efficiency of the proposed method. It is confirmed that the proposed method outperforms the conventional ones for the optimization of CO2-WAG injection.