Efficient utilization of water resources by reservoir operation in arid and semi-arid regions has been a traditional and important approach to mitigate water and energy scarcity. As the complexity of water resources system increases, there are urgent needs to apply effective optimization techniques to derive better operation rules. The meta-heuristic algorithms have been an alternative to the optimal operation of multi-reservoir system (OOMRS) for their parallel search capabilities. This paper presents the application of the cuckoo search (CS) algorithm to OOMRS with the objective to maximize the energy production. The penalty method was used to handle the physical and operational constraints. To make the CS more efficient, the optimized parameters of the CS were selected based on the parameter sensibility analysis. Finally, the performance of the CS was accessed by comparison with the genetic algorithm (GA) and the particle swarm optimization (PSO) under the same objective function evaluations (FEs). A case study of China’s Wujiang multi-reservoir system reveals that the CS can provide better and more reliable optimal results with average energy production of 12.31 billion kW∙h, 10.43 billion kW∙h, and 10.02 billion kW∙h for three different scenarios, which are approximately 0.52, 0.32, and 1.64 % higher than that of the GA, respectively. Meanwhile, the convergence performance of the CS is also satisfying. Therefore, it can be concluded that the CS is quite promising in handling complex reservoir operation optimization problem in terms of its simple structure, excellent search efficiency, and strong robustness.
Read full abstract