Well Control Optimization of Offshore Horizontal Steam Flooding Wells Using Artificial Intelligence Algorithm

Abstract

Maximizing the future economic return of the asset is an important issue in petroleum engineering. For heavy oil reservoir developed with steam flooding, its production cost is much higher than that of conventional production methods. Commonly used parameters optimization method such as single factor analysis and orthogonal test cannot guarantee to obtain global optimal economic benefits. It is necessary and urgent to form a better optimization method to achieve higher profit. A new framework is proposed and presented to optimize well control parameters of both steam injection wells and oil production wells by integrating the reservoir simulator into the optimization algorithms. A net present value (NPV) formula for evaluation of horizontal well steam flooding project is proposed and the optimization objective is to maximize the NPV of production over the life. The generally acknowledged Particle swarm optimization (PSO) is used for solution of the optimization problem. This method has been tested for a typical offshore horizontal well steam flooding project. Results indicate that PSO gives good solutions for this problem and the following conclusions can be obtained. The NPV of the optimized project is improved and larger than the NPV of its initial guess. The control frequency has great influence on the optimal NPV, and the optimal NPV increases with the increase of the control frequency. Steam injection and oil production rates need to be controlled and decreased at the latter stage for mitigating ineffective steam cycle between injection and production wells. The new method has been used for well control optimization of the first offshore horizontal well steam flooding pilot and this method would which will provide powerful technical support for the high efficiency development of the heavy oil resource with steam flooding.