Application of F-HGAPSO Algorithm in Reservoir Flood Control Optimal Operation

Flood control operation of reservoir is a high-dimensional and nonlinear problem with numerous constraints, and intelligent algorithms are widely used to solve the flood control optimization problem. The traditional intelligent algorithms often face issues such as slow convergence and a tendency to find local optima when solving flood control optimization problems. This paper proposes a method that couples a hybrid genetic algorithm-particle swarm optimization with fuzzy adaptive inertia weights (F-HGAPSO) to solve the model established for reservoir flood control operation. The initial population is generated from the reservoir discharge taking as the decision variable, initially optimized using the genetic algorithm, and then further optimized to obtain the results using the particle swarm optimization (PSO) with a Mamdani fuzzy system for adaptive inertia weight. Finally, as a case study, the proposed method is applied to solve the model established for Mushan reservoir flood control operation, with a penalty function handling the constraints. To demonstrate the validity of the method, the results obtained by F-HGAPSO are compared and analyzed with those obtained by PSO and the elite genetic algorithm (EGA). The results show that F-HGAPSO outperforms PSO and EGA in solving the flood control model, achieving peak flow reductions of 20.28%-30.27% with rapid convergence in 13–22 iterations, effectively avoiding local optima issues, and that highlights the F—HGAPSO has strong optimization ability and efficiency. The proposed method can be extended to the flood control operation of other reservoirs with similar conditions.