论文部分内容阅读
针对人群搜索算法在进化后期大量个体聚集局部最优时,易陷入局部最优,搜索精度低的缺陷,提出一种基于t分布变异的人群搜索算法.算法使用动态自适应方式确定变异步长,引入t分布变异算子以融合柯西变异和高斯变异的优点,促进算法在进化早期具备良好的全局探索能力,在进化后期收获较强的局部开发能力,增加种群的多样性;采用边界缓冲墙策略处理越界问题,避免越界个体聚集在边界值上的缺陷.实验结果表明,算法比基本人群搜索算法具有更高的寻优精度和收敛速度,是一种有效的算法.
Aiming at the defect that the crowd search algorithm aggregates local optimum in the late evolutionary stage, it tends to fall into local optimum and search accuracy is low, and a crowd search algorithm based on t-distribution mutation is proposed.The algorithm uses dynamic adaptive method to determine the variation step, The introduction of t distribution mutation operator to merge Cauchy mutation and Gaussian mutation has the advantage of promoting the algorithm in the early evolution of a good global exploration ability in the late evolution of the harvest local strong development ability to increase the diversity of the population; Strategy to deal with the problem of cross-border, to avoid the boundary of the individual across the boundary value of the defect.The experimental results show that the algorithm has better search accuracy and convergence speed than the basic crowd search algorithm is an effective algorithm.