基于黄金正弦与自适应融合的蜉蝣优化算法

Mayfly optimization algorithm based on gold sine and adaptive merge

针对元启发算法中蜉蝣优化算法(MOA)的求解精度不高、收敛速度慢、稳定性不强等缺点进行研究,提出一种黄金正弦与自适应融合的蜉蝣优化算法。引入自适应惯性权重因子增强算法的搜索和开发能力达到更好的平衡;引入融合Lévy飞行策略和黄金正弦因子进一步改善易陷入局部最优的缺点,增强种群多样性,跳出局部最优。仿真结果表明,改进算法对于测试函数在求解精度、收敛速度和寻优能力上有显著提升。同时,为验证结果的可靠性和有效性,对该算法所得的数据进行统计检验、平均绝对误差分析、求解成功率分析。结果表明改进算法的稳定性、可靠性、鲁棒性都较MOA有所增强。另外,引入具体工程案例进行测试分析,进一步验证了该算法在工程上的适用性。

In order to solve the shortcoming of meta-heuristic mayfly optimization algorithm(MOA),such as low precision,slow convergence velocity and low stability,this paper proposed a new mayfly optimization algorithm,which combined the gold sine and adaptive merge.Firstly,it introduced the adaptive inertia weight factor to enhance the search and development ability of the algorithm to achieve a better balance.Then it introduced the fusion Lévy flight strategy and the gold sine factor to further improve the shortcoming of fall into the local optimum,enhanced the population diversity and jumped out of the local optimum.Simulation results show that the solution precision,convergence velocity and optimization ability of the improved algorithms are significantly improved on the test functions.At the same time,in order to verify the reliability and validity of the results,this paper analyzed the data obtained by the improved algorithm on statistical test,mean absolute error and solution success rate.The algorithm has obvious improvement in the stability,reliability and robustness compared with the MOA.Besides,this paper introduced a specific engineer case to verify the applicability of the algorithm in engineering.