水库防洪优化调度模型一般属于高维多峰极值问题，通常采用智能优化算法加以求解。粒子群算法由于其简单易行被广泛应用于水库优化调度中，但是该算法存在局部搜索能力不足、早熟收敛、全局收敛性差等问题。针对这些问题，通过引入Logistic方程和变异算子来提高种群的多样性，采用收敛因子来提高算法的收敛速度，并将改进的粒子群算法应用到东圳水库与木兰溪流域的防洪优化调度中，求得关键处河道的最高水位为6.35 m，最大流量为959.2 m3/s。这一结果与现行规则下的运行结果（最高水位6.93 m，最大流量1 139.5 m3/s）和常规粒子群算法计算结果（最高水位6.51 m，最大流量1 066.3 m3/s）相比，有了很大的改善。

The optimal operation model for flood control of reservoir is a high-dimensional multimodal extremum problem in general, and intelligent optimization algorithm is usually used to solve such problems. Particle swarm optimization (PSO) algorithm is widely used in the optimal operation of reservoir due to its simplicity; however, there are shortcomings in the PSO algorithm such as premature convergence, low efficiency in global convergence, and deficiency in local search capability. Logistic equation and mutation operator are introduced to increase the diversity of population and convergence factor is introduced to improve the convergence rate during the iterative process. The improved PSO algorithm was applied in the optimal operation for flood control of the Dongzhen Reservoir and Mulanxi Watershed. The resulted showed that the maximum water level is 6.35 m and the maximum flow rate is 959.2m3/s at the pivotal watercourse. The solutions were much better than that obtained from the present reservoir control regulations (maximum water level of 6.93 m and maximum flow rate of 1139.5 m3/s) and that determined by the standard PSO algorithm (maximum water level of 6.51 m and maximum flow rate of 1066.3 m3/s).

中央高校业务费资助项目