一种改进的蝴蝶优化算法

An Improved Butterfly Optimization Algorithm

本文针对基本的蝴蝶优化算法存在收敛速度慢、精度低和易陷入局部最优等缺陷,提出一种改进的蝴蝶优化算法.首先通过实验分析参数对算法的影响,其次融入差分进化策略和精英策略,通过10个标准测试函数进行测试,结果表明,改进算法在8个测试函数中均找到了理论最优解,其收敛速度、精度和鲁棒性均优于基本的蝙蝠算法(BA)、花朵授粉算法(FPA)、布谷鸟算法(CS)、融合差分进化算法的花朵授粉算法(DEFPA)、蝴蝶算法(BOA)和融合差分进化算法的蝴蝶算法(DEBOA),且寻优性能得到大幅度提升;同时对4个非线性方程的求解也验证了该算法的有效性.

A improved butterfly optimization algorithm was presented to overcome the problems of low-accuracy computation,slow-speed convergence and being easily relapsed into local extremum.Firstly,the influence of parameters on the algorithm is analyzed by experiments.Secondly,the differential evolution strategy and elite strategy are integrated.Ten standard test functions are tested.The results show that the improved algorithm finds the oretical optimal solution in eight test functions,and its convergence speed,accuracy and robustness are better than the basic bat algorithm(BA) and the flower pollination algorithm(FPA).The cuckoo algorithm(CS),flower pollination algorithm incorporating differential evolution algorithm(DEFPA),butterfly algorithm(BOA) and butterfly algorithm incorporating differential evolution algorithm(DEBOA),The optimization performance of the improved algorithm has been greatly improved.At the same time,the experimental results of solving four nonlinear equation group verify the validity of the algorithm.