Multi-objective optimization of reservoir flood dispatch based on MOPSO algorithm

This paper proposes a method using multi-objective particle swarm optimization (MOPSO) algorithm to solve the multi-objective optimal dispatch problem of reservoir flood control, which take minimum value of the highest water level before dam, minimum value of the releasing peak discharge, and water level after flood season very close to flood control level as the objective functions. By using the archiving technique, crowding distance sorting algorithm and mutation technique to improve the algorithm convergence speed and accuracy and enable the Pareto solution set to converge to optimal front promptly and distribute evenly. The algorithm is applied to optimize the dispatch of the Yuecheng reservoir in upper Zhanghe River of the Haihe basin for typical floods occurred in history and the relative relations between dispatching objectives are analyzed. The result indicates that a lot of noninferior dispatch schemes can be generated in a short time, which can provide scientific basis for the decision-maker to make optimal operation and evaluation decision.