基于混合混沌扰动与柯西变异的黏菌优化算法及其应用  

Slime Mold Optimization Algorithm based on Mixed Chaos Disturbance and Cauchy Mutation and Its Application

作　　者：黄棉[1] 王雨虹[2] HUANG Mian;WANG Yuhong(College of Mechanical and Electrical Engineering,Xiamen Huatian International Vocational and Technical College,Xiamen 361101,China;Faculty of Electrical and Control Engineering,Liaoning Technical University,Huludao 125105,China)

机构地区：[1]厦门华天涉外职业技术学院机电工程学院,福建厦门361101 [2]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《成都工业学院学报》2023年第3期62-69,共8页Journal of Chengdu Technological University

基　　金：辽宁省教育厅科技项目(LJ2019QL015)。

摘　　要：针对黏菌优化算法存在自适应能力有限,抗停滞能力弱等不足,提出一种混合混沌扰动与柯西变异的黏菌优化算法。首先,初始化阶段采用折射反向学习策略精英化种群;然后,结合数量自适应调整策略协调算法的全局勘探与局部开发;其次,为降低算法陷入局部极值的缺陷,利用柯西算子对精英个体进行变异,混沌策略对其他个体进行扰动;其次,通过8个基准测试函数和Wilcoxon符号秩检验,结果均表明改进算法具有更好的寻优精度、稳定性;最后改进的黏菌优化算法对核极限学习机进行参数优化,应用于变压器故障故障诊断,实验结果进一步验证了改进策略的有效性。In view of the limitations of slime mold optimization algorithm such as limited adaptive ability and weak anti-stagnation ability,a slime mold optimization algorithm based on chaotic disturbance and Cauchy variation was proposed.Firstly,the refraction reverse learning strategy was used to excellentize the population in the initialization stage.Secondly,the global exploration and local development of the algorithm were coordinated with the quantitative adaptive adjustment strategy.Thirdly,in order to reduce the defect of the algorithm falling into local extremum,cauchy operator was used to mutate elite individuals,and chaos strategy was used to perturb other individuals.Then,the results of 8 benchmark test functions and Wilcoxon sign rank test show that the improved algorithm has better optimization accuracy and stability.Finally,the improved slime mold optimization algorithm was used to optimize the parameters of the kernel extreme learning machine,and was applied to transformer fault diagnosis.The experimental results further verified the effectiveness of the improved strategy.

关 键 词：黏菌优化算法 混沌策略 反向学习 柯西算子 变压器故障诊断 

分 类 号：TD713[矿业工程—矿井通风与安全]