基于动态自适应t分布变异的人群搜索算法

Seeker Optimization Algorithm Based on Dynamic Adaptive t Distribution Mixed Mutation

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.

The traditional seeker optimization algorithms are prone to fall into local optimum when a large number of individuals gather in the later stage of evolution, which reduces the searching precision. To overcome this disadvantage, a t-based distribution variation seeker optimization algorithm is presented in this paper. The algorithm uses dynamic adaptive method to determine the mutation step, and introduces a t distribution mutation operator with the merits of gauss mutation and cauchy mutation to enhance the global searching ability in the early evolution. The local development ability of the algorithm is improved in the later evolution period, and the population diversity is increased. In order to avoid the defect of the slopping-over individuals gathered in the boundary, a border buffer wall is employed. The experimental results show that the proposed algorithm has higher searching precision and convergence rate than those of the conventional seeker optimization algorithms.