Cascade Hydropower System Operation Considering Ecological Flow Based on Different Multi-Objective Genetic Algorithms

The main objective of most hydropower systems is to pursue the efficient use of water resources and maximize economic benefits. At the same time the protection of ecological environment should not be neglected. In this study, a coordination model of power generation and ecological flow for the operation of cascade hydropower system was first established using a multi-objective optimization method. Multi-objective genetic algorithms (MOGAs) are widely used to solve such multi-objective optimization problems because of their excellent performance in terms of convergence speed, diversity of solution set space and optimality seeking ability. However, the adaptability of MOGAs to a particular optimized operation problem sometimes varies widely. It is of great significance to investigate the adaptability of different algorithms for a new optimized operation problem and to recommend a more suitable solution algorithm. Three MOGAs namely NSGA-II, NSGA-III and RVEA are selected to solve the proposed optimized operation model. Numerical experiments were conducted to evaluate the performance of the algorithms using real-world data from a cascade hydropower system located in the lower Yalong River. The results show that the Pareto fronts corresponding to NSGA-II and NSGA-III significantly dominate the Pareto fronts corresponding to the RVEA. The Pareto fronts corresponding to the NSGA-III algorithm are slightly better than those of NSGA-II. In terms of the four performance metrics, NSGA-III has certain advantages over NSGA-II and RVEA. NSGA-III is recommended as the solution algorithm for the established coordination model of power generation and ecological flow.

Automated summary

In this study, a coordination model of power generation and ecological flow for cascade hydropower system was established using a multi-objective optimization method. Three MOGAs (NSGA-II, NSGA-III, and RVEA) were selected to solve the model, and the results show that NSGA-III has certain advantages over NSGA-II and RVEA in terms of performance metrics. Therefore, NSGA-III is recommended as the solution algorithm for the established coordination model.