中文摘要：

针对梯级水库群优化调度多约束、高维、非线性和难以求解的特点，将鲶鱼效应机制引入到粒子群算法中提出鲶鱼效应粒子群算法。该算法在进化中通过鲶鱼启发器引入鲶鱼粒子，依据鲶鱼效应调整种群的飞行模式，一方面利用鲶鱼粒子的驱赶作用使粒子种群跳出稳定状态激发活力，从而提高种群多样性；另一方面利用鲶鱼的高素质动态调节对进化过程进行有目的指导，进而保持算法的高搜索性能。算例表明，和标准粒子群算法、混沌粒子群算法相比，鲶鱼效应粒子群算法具有更好的全局寻优能力和较快的收敛速度，能有效地应用于梯级水库群优化调度中。

英文摘要：

In the light of the characteristics of the optimal operation of cascade reservoirs, such as multi-restriction, multi-dimension, non-linearity and being difficult to slove, the catfish effect mechanism is introduced into the particle swarm optimization, named catfish effect particle swarm optimization. The arithmetic introduces catfish particles through the catfish generator in the evolution and adjusts the flying pattern of population swarm by using the catfish effect. On the one hand, the driven influence of catfish particles is used to force the swarm out of steady-state and inspire its vitality in order to improve the whole diversity; on the other hand, it takes the advantage of the high quality of dynamic adjustment of catfish to guide the optimization and hence keep remaining its high search function. Calculation results show that compared with the standard particle swarm optimization and the chaotic particle swarm optimization, the catfish effect particle swarm optimization has better global searching capability and faster convergence speed, which can be effectively applied to the optimal operation of cascade reservoirs. This work is supported by National Natural Science Foundation of China （No. 40971300 and No. 50609007）.