Developing optimal policies for reservoir systems using a multi-strategy optimization algorithm

Abstract

It is still a challenge to effectively optimize operation policies for reservoir systems, due to their large-scale and stochastic natures. The development and improvement of the optimization methods for optimizing reservoir operation systems are, therefore, a worthy undertaking. Hence, the objective of this study is to develop an effective hybrid of differential evolution (DE) and particle swarm optimization (PSO) with multi-strategy (MS-DEPSO) to optimize the operating policies for reservoir systems. The proposed MS-DEPSO promotes the local and global search capabilities of the basic DE algorithm to obtain an effective optimal operating policy. Fourteen mathematical functions were applied to verify the performance of the proposed optimization method. Furthermore, a multi-reservoir hydropower system with three various monthly operation periods over 10, 15, and 20 years was used as a real case study to evaluate the efficiency of MS-DEPSO in hydropower energy generation. Finally, the optimal operating rules were obtained based on the reservoir rule curves for a single reservoir with the purpose of agricultural water supply. The results highlighted the competency of the proposed optimization model to reduce the impact of severe drought periods. It is demonstrated in this study that the proposed algorithm has a superior ability to extract the optimal operating rules for reservoir systems.