论文部分内容阅读
针对新颖全局和声搜索(NGHS)算法过早收敛的问题,提出自适应全局和声搜索(AGHS)算法.引入差分向量范数定义和声记忆库多样性,给出新的位置更新策略,排除变异操作.以和声记忆库多样性信息为指导动态产生新和声,提高算法对解空间信息开发的能力,避免算法因过早收敛、易陷入局部最优的不足.AGHS算法操作更简单,需要设置的参数更少,将其与目前文献中较优的几种改进HS算法、PSO算法和GA算法进行性能测试,测试结果表明AGHS算法具有较高的寻优精度和较快的收敛速度.
Aiming at the problem of premature convergence of novel global harmony search (NGHS) algorithm, an adaptive global harmony search (AGHS) algorithm is proposed. The definition of difference vector norm and the diversity of sound memory are introduced, and a new location updating strategy is given, Variational operation.Using the diversity information of harmony memory as a guide to dynamically generate new harmonics and improve the ability of the algorithm to develop the solution space information and avoid the shortcomings of the algorithm getting into the local optimum due to premature convergence.AGHS algorithm is simpler to operate, The parameters to be set are less, and compared with several improved HS algorithms, PSO algorithm and GA algorithm which are better in the current literature for performance testing. The test results show that the AGHS algorithm has higher precision and faster convergence speed.