混合策略改进的麻雀优化算法

Mixed Strategy to Improve Sparrow Search Algorithm

针对麻雀优化算法在迭代后期种群多样性减少,收敛速度较慢,高纬度求解精度较低等缺点,本文提出一种混合策略改进的麻雀算法.首先,利用佳点集法初始化麻雀个体位置,提高初始个体的多样性;其次,提出黄金莱维飞行策略和t-分布扰动策略共同改进发现者位置更新方式,解决算法在迭代后期种群多样性减少的问题;最后,提出动态分配侦察者策略,有效平衡侦察者全局探索与局部开采的能力.将本文所提算法与主流算法在14个基准测试函数以及部分CEC2014函数上进行比较实验,使用数值分析、收敛性分析、Wilcoxon秩和检验分析评估本文所提算法的性能.实验结果表明,本文算法具有更好的寻优精度和收敛速度,在高维度问题求解上,具有更好的性能.

Aiming at the shortcomings of the sparrow optimization algorithm in the later stages of the iteration,the population di-versity is reduced,the convergence speed is slow,and solution accuracy in the high-latitude is low,a hybrid strategy was proposed to improve the sparrow optimization algorithm.First,initialize the position of sparrow individuals by introducing the good point set method to enhance the ergodicity of the initial population and improve the quality of the initial solution;secondly,use the golden Lévy flight strategy and the t-distribution disturbance strategy to jointly improve the location update method of the discoverer and improve the algorithm in iterative The shortcomings of the reduction of population diversity in the later stage;finally,the dynamic allocation of scouts is introduced to effectively balance the scouts′ global exploration and local mining capabilities.Use statistical analysis,convergence rate analysis,Wilcoxon rank sum test,classic benchmark function,CEC2014 to evaluate the performance of the improved sparrow optimization algorithm.Experimental results show that the proposed algorithm has better optimization accuracy and convergence speed,and has better performance in solving high-dimensional problems.