摘要
为解决经典粒群算法对于初值选取的依赖性,引入一种具有排斥效应的粒子群优化算法,通过引入粒子间的引力-斥力效应,使得算法对于初值的依赖性显著减弱。数值仿真表明,该算法相比于传统算法具有更快的收敛速度。
引文
[1]KENNEDY J,EBERHART R C.Particle swarm optimization[C].IEEE Press,1995:1942-1948.
[2]NA G,ZHOU W,CHANG X.A nove particle swarm optimization algorithm based on particle migration[J].Applied Mathematics&Computation,2012,218(11):6620-6626.
[3]JIN Q B,CHENG Z J,DOU J,et al.A novel closed loop identification method and its application of multivariable system[J].Journal of Process Control,2012,22(1):3525-3528.
[4]李孟山,黄兴元,柳和生,等.基于混沌自适应粒子群人工神经网络的气体在聚合物中的溶解模型[J].化学学报,2013,71(7):1053-1058.
[5]SHEN X,CHI Z,YANG J,et al.Partial swarm optimization with dynamic adaptive weight[C].Challenges in Environmental Science and Computer Engineering,2010:287-290.
[6]张佐光,张晓宏,仲伟虹,等.多相混杂纤维复合材料拉伸行为分析[M].北京:世界图书出版公司,2006:410-416.
[7]胡建秀,曾建潮.二阶微粒群算法[J].计算机研究与发展,2007,44(11):1825-1831.
[8]胡旺,李志蜀.一种更简化而高效的粒子群优化算法[J].软件学报,2007,18(4):861-868.