改进微粒群算法在梯级电站长期优化调度中的应用  

Application of Modified Particle Swarm Algorithm in Long-Term Optimal Operation of Cascade Hydropower Stations

作　　者：朱凤霞[1] 熊立华[1] 高仕春[1] 艾学山[1] 

机构地区：[1]武汉大学水资源与水电工程科学国家重点实验室,湖北武汉430072

出　　处：《水文》2007年第5期42-45,77,共5页Journal of China Hydrology

基　　金：国家自然科学基金(50409008);教育部新世纪优秀人才支持计划(NCET-05-0624);霍英东青年教师基金(101077)资助

摘　　要：微粒群算法是一种简洁高效的智能优化算法,但基本算法容易陷入局部最优,并且搜索精度不高。本文在基本算法的基础上引入锦标赛选择机制和自适应惯性权重因子,提出了改进微粒群算法(MPSO)。将MPSO算法应用到黄河上游梯级电站的长期调度中,并与动态规划法和基本算法的调度结果相比较。实例表明了MPSO算法的有效性和可靠性,从而为梯级电站水库(群)长期优化调度提供了一种新的、有效的优化方法。Particle swarm optimization （PSO） algorithm is a simple and high efficient intelligent optimization algorithm. Considering that PSO is easy to trap in local optima and can＇ t find the optimal solution efficiently, the paper proposed a modified particle swarm optimization （MPSO） algorithm, by introducing the tournament selection method and also a self-adaptive inertia weigh into the basic PSO. The MPSO algorithm then was applied in the long-term operation optimization of the cascade hydropower stations on the upper reaches of the Yellow River. The operation results of MPSO were compared with these of the basic PSO as well as the dynamic planning method, which has demonstrated the effectiveness and reliability of this MPSO. Therefore, the MPSO, as a new and effective optimization method, can be applied forthe long-term operation optimization of reservoirs.

关 键 词：微粒群算法 锦标赛选择机制 惯性权重 优化调度 

分 类 号：TV697.12[水利工程—水利水电工程]