狼群算法的改进及其在水库优化调度中的应用!  

Improvement of wolf pack search algorithm and its application to optimal operation of reservoirs

作　　者：王建群[1] 焦钰[1] 

机构地区：[1]河海大学水文水资源学院,江苏南京210098

出　　处：《武汉大学学报（工学版）》2017年第2期161-167,173,共8页Engineering Journal of Wuhan University

基　　金：国家重点研发计划课题(编号:2016YFC0400909);国家自然科学基金项目(编号:41371047)

摘　　要：为了在水电站优化调度领域更加有效地应用狼群算法,对基本狼群算法(WPA)进行了改进研究.为了克服WPA过分强调个体的独立性、缺乏群体之间信息的共享及个体对历史经验的认知学习等缺陷,对WPA的召唤奔袭算子和围攻猎物算子进行了改进,提出了改进的狼群算法(NWPA);基于基准测试函数和水库优化调度实例对NWPA算法的综合性能展开了仿真实验并与粒子群算法(PSO)、WPA算法及动态规划(DP)进行了比较;对NWPA算法的关键参数有效取值范围进行了仿真实验分析.结果表明,NWPA算法的寻优能力和收敛速度优于PSO算法、WPA算法;给出关键参数的有效取值范围和建议取值,为NWPA算法应用于水库优化调度提供了参数取值依据.In order to more effectively applying the wolf pack algorithm to optimal operation of reservoirs, the improvement of the wolf pack search algorithm （WPA） is studied. To overcome defects of WPA such as overemphasis of the individual independence, failing to share information among groups and learn indi- vidual historical experience, the call raid operator and the siege prey operator are improved; and a new wolf pack search algorithm （NWPA） is proposed. Simulation experiments based on the benchmark functions and the example of the reservoir optimal operation are carried out to test the comprehensive performance of the NWPA compared with the particle swarm optimization algorithm （PSO） and WPA. The effective range of the critical parameters of NWPA are studied with the simulation experiments. The results show that the searching ability and convergence speed of NWPA has an obvious improvement compared with PSO and WPA. The effective value range and the recommended value of the parameters are given; and the parameter value basis is provided for the application of NWPA to the optimal operation of reservoirs.

关 键 词：狼群算法 粒子群算法 基准函数 水电站优化调度 参数 

分 类 号：TV11[水利工程—水文学及水资源]