Renmin Zhujiang (Nov 2024)
Optimized Flood Control Scheduling of Goupitan Reservoir Based on Dynamic Programming and Grey Wolf Algorithm
Abstract
Based on the analysis of the importance of different time periods for flood control, this paper proposed a solution approach that segments flood control into key and non-key periods. An efficient solution algorithm for the flood control model that coupled dynamic programming and grey wolf algorithm was proposed through a bidirectional feedback loop mechanism, which used the grey wolf algorithm to provide boundary conditions for the dynamic programming, while the dynamic programming influences the fitness values of the grey wolf algorithm. The extreme and frequent floods of Goupitan Reservoir were solved. Compared to the actual situation, the peak flow of the control section was reduced by 42% and 43%. The time computation of the two scenarios was 19.54 s and 71.27 s, respectively, while that of the dynamic programming is 90~141 s and 300~500 s with the same accuracy. This indicated that the coupled algorithm balanced solution efficiency and accuracy by combining the strengths of both dynamic programming and grey wolf algorithm, providing a new solution for the flood control scheduling model of the reservoir.
