The contradiction between the reservoir and the target is becoming more and more prominent, and the current reservoir scheduling method is difficult to well support the regional social and economic development. This paper takes the Xiaolangdi Reservoir on the mainstream of the Yellow River as the research object, establishes a dual-objective and three-objective reservoir optimal operation model, uses NNIA, MOPSO and FSS-NNIA to solve the dual-objective reservoir optimal operation model, and validates FSS-NNIA through quantitative analysis for the rationality and effectiveness of the system, FSS-NNIA is used to solve the three-objective reservoir optimal operation model, and to explore the competitive game relationship between the three objectives. The results show that: (1) Compared with NNIA and MOPSO algorithms, FSS-NNIA has a wide range of non-inferior solution distribution, it has advantages in uniformity and stability, and higher-quality Pareto-Front curves can be obtained; (2) The different performance indexes showed the superiority of the FSS-NNIA compared to the other algorithms. For example, the hypervolume (HV) of FSS-NNIA is 0.116 and 0.0016 larger than MOPSO and NNIA; the non-dominated solutions of FSS-NNIA is 71 and 31 more than MOPSO and NNIA. Also, the computational time for the FSS-NNIA was 27 s and 5 s less than MOPSO and NNIA, respectively; (3) Solving the three-objective optimization scheduling model by FSS-NNIA verifies that there is a competitive relationship between the three objectives, and uses the three-dimensional projection method to obtain the transformation efficiency distribution law of any two objectives. The research results provide a new solution idea and method for reservoir multi-objective optimization scheduling, which has important theoretical significance for supplementing and improving the theory and method of the reservoir multi-objective scheduling, and has been important guiding significance and application value for the comprehensive utilization and scheduling operation of Xiaolangdi Reservoir.
Read full abstract