具有动态方差高斯变异的蝴蝶优化算法

Butterfly Optimization Algorithm with Dynamic Variance Gaussian Mutation

针对基本蝴蝶优化算法迭代速度慢,容易陷入局部最优解的问题,提出一种具有动态方差高斯变异的蝴蝶优化算法.首先引入线性函数的动态切换概率来平衡迭代过程中全局搜索和局部搜索的相对比例,其次利用动态方差高斯变异改进了全局搜索和局部搜索的公式,有利于加快算法的收敛速度,跳出局部最优解,在迭代后期提高寻优的精度.使用了6个基准函数进行仿真实验,实验数据对比发现,提出的算法比其他五种算法具有更好的寻优效果和更强的稳定性.

Aiming at the problem that the basic butterfly optimization algorithm has slow iteration speed and is easy to fall into local optimal solution,a butterfly optimization algorithm with dynamic variance Gaussian variation is proposed.Firstly,the dynamic switching probability of linear function is introduced to balance the relative proportion of global search and local search in the iterative process.Secondly,the formula of global search and local search is improved by using dynamic variance Gaussian variation,which is conducive to accelerating the convergence speed of the algorithm,jumping out of the local optimal solution and improving the optimization accuracy in the later stage of iteration.Six benchmark functions are used for simulation experiments.In the comparison of experimental data,it is found that the proposed algorithm has better optimization effect and stronger stability than the other five algorithms.