基于乌鸦搜索算法优化变分模态分解的滚动轴承故障诊断方法研究  

Research on Rolling Bearing Fault Diagnosis Method Based on Variational Mode Decomposition Optimized by Crow Search Algorithm

作　　者：李萌 刘晏铭 LI Meng;LIU Yanming(College of Machinery and Vehicle Engineering,Changchun University,Changchun 130022,China)

机构地区：[1]长春大学机械与车辆工程学院,长春130000

出　　处：《长春大学学报》2023年第4期32-39,43,共9页Journal of Changchun University

基　　金：吉林省科技厅项目(20230101208JC)。

摘　　要：针对滚动轴承故障信号的自适应提取和分解的问题,提出一种基于乌鸦搜索算法优化变分模态分解的滚动轴承故障诊断方法。将变分模态分解(variational mode decomposition, VMD)方法的关键参数K和α采用新型的乌鸦搜索算法(crow search algorithm, CSA)进行优化,得到最优参数组合;再将最优参数组合输入到变分模态分解算法中,对故障信号进行分解从而得到多个本征模态分量(intrinsic mode function, IMF);以样本熵值为适应度函数挑选最优分量,对最优分量进行包络解调,分析其包络谱判断出轴承的故障类型。结果表明,提出的方法在兼顾全局搜索和局部搜索的同时也能将复杂的轴承故障信号准确地进行分解,提取出最优分量进行分析从而判断出轴承故障类型。In view of the problem of self-adaptive extraction and decomposition of fault signals of rolling bearing,a fault diagnosis method of rolling bearing based on variational modal decomposition optimized by crow search algorithm is proposed.Firstly,the key parameter K andαof variational mode decomposition(VMD)method is optimized with new crow search algorithm(CSA)to obtain the most optimal parameter combination;Secondly,the most optimal parameter combination is input into VMD algorithm,and the fault signals are decomposed to obtain multiple intrinsic mode functions(IMF);Finally,the most optimal component is selected by taking the sample entropy value as the fitness function,and the envelope demodulation of the optimal component is performed to determines the fault type of the bearing through analyzing its envelope spectrum.The research results show that the method proposed in this paper can not only take into account the global search and local search,but also can accurately decompose the complex bearing fault signals,extract the most optimal component and judge the bearing fault type.

关 键 词：轴承 故障诊断 乌鸦搜索算法 优化变分模态分解 包络解调 

分 类 号：TH133.33[机械工程—机械制造及自动化]