Research on flood control and drainage scheduling based on a hybrid genetic algorithm

With the acceleration of urbanization and the frequent occurrence of extreme weather, traditional flood control and drainage scheduling methods face challenges such as low computational efficiency and single optimization objectives when dealing with complex urban river network systems. The article proposes a hybrid intelligent algorithm (GA-PSO) that combines the global search capability of genetic algorithm (GA) with the fast convergence characteristics of particle swarm optimization (PSO), and constructs a multi-objective optimization model with drainage benefits and economic costs as the core. By using dynamic weight adjustment strategies and smoothing operators based on first-order low-pass filtering, the frequent opening and closing of gates can be effectively suppressed, improving the engineering applicability of scheduling schemes. The model aims to minimize the over alert water level, flooded area, flooded time, and scheduling cost, combined with constraints such as river water balance and gate opening. The experimental results show that: 1) the convergence speed of the hybrid algorithm is improved by 25% compared to NSGA-III, and the HV value is improved by 8.8%; 2) The filtering operator reduces the daily operation frequency of the gate by 40% and the start stop frequency of the pump station by 35%; 3) The trade-off between drainage effectiveness and economic cost was quantified based on the Pareto frontier solution set. This study provides an efficient decision-making tool for urban river network flood control and scheduling, and has algorithm innovation and engineering practicality.