基于双种群协调进化果蝇算法优化SVM的变速箱故障诊断  

GEARBOX DIAGNOSIS BASED ON SVM OPTIMIZED BY DOUBLE GROUP COEVOLUTION FRUIT FLY OPTIMIZATION ALGORITHM

作　　者：雷彪[1] 惠恩明 关海英[1] 王小军 LEI Biao;HUI EnMing;GUAN HaiYing;WANG XiaoJun(Department of Mechanical and Electrical Engineering,Inner Mongolia Institute of Mechanical and Electrical Technology,Hohhot 010070,China;National Nc System Engineering Research Centre,Huazhong University of Science and Technology,Wuhan 430074,China;Wuhan Huazhong CNC Co.,Ltd.,Wuhan 430000,China)

机构地区：[1]内蒙古机电职业技术学院机电工程系,呼和浩特010070 [2]华中科技大学国家数控系统工程技术研究中心,武汉430074 [3]武汉华中数控股份有限公司,武汉430000

出　　处：《机械强度》2022年第3期753-757,共5页Journal of Mechanical Strength

基　　金：内蒙古自治区教育厅自然科学一般项目(NJZY21354)资助。

摘　　要：为提高果蝇算法(FOA)对支持向量机(SVM)参数优化的效果,对FOA的进化策略进行了改进,提出了双种群协调进化果蝇算法(DGCFOA)。将DGCFOA用于SVM参数优化并进行变速箱的故障诊断,诊断结果表明,DGCFOA算法能够搜寻到更优的SVM参数,相比于FOA,明显提升了变速箱故障诊断的精度。此外,与其他一些方法的对比结果也显示出DGCFOA得到的诊断精度更高,优势较为明显。In order to improve the optimization effect of the fruit fly optimization algorithm(FOA) on support vector machine(SVM) parameters, the evolution strategy of FOA was improved, and the duoble group coevolution fruit fly optimization algorithm(DGCFOA) was proposed in this paper. The DGCFOA was used to optimize the parameters of SVM and then used to gearbox fault diagnosis. Diagnosis results show that DGCFOA algorithm can obtained better SVM parameters when compared with FOA, it significantly improved fault diagnosis accuracy of gearbox. In addition, the diagnosis results also show that DGCFOA has higher diagnostic accuracy and more obvious advantages when compared with some other methods.

关 键 词：果蝇算法 协同进化 支持向量机 变速箱 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化]