基于MPA-VMD的滚动轴承故障诊断方法  

Fault Diagnosis Method of Rolling Bearings Based on MPA-VMD

作　　者：吴科伟 封远鹏 王超 王广斌 何水龙[1] 蒋占四[1,2] WU Kewei;FENG Yuanpeng;WANG Chao;WANG Guangbin;HE Shuilong;JIANG Zhansi(School of Mechanical and Electrical Engineering,Guilin University of Electronic Technology,Guilin 541004,Guangxi,China;School of Mechanical and Electrical Engineering,Lingnan Normal University,Zhanjiang 524048,Guangdong,China)

机构地区：[1]桂林电子科技大学机电工程学院,广西桂林541004 [2]岭南师范学院机电工程学院,广东湛江524048

出　　处：《噪声与振动控制》2023年第2期112-119,共8页Noise and Vibration Control

基　　金：国家自然科学基金资助项目(51565008,51965013);广西科技基地和人才专项资助项目(2019JJB160062);广西区研究生创新资助项目(YCSW2021187);桂林电子科技大学研究生教育创新计划项目(2022YCXS017)。

摘　　要：针对变分模态分解方法(Variation mode decomposition,VMD)在提取滚动轴承振动信号的故障特征频率时受参数设置影响及敏感模态分量的选取问题,构建一种基于海洋捕食者算法(Marine Predator Algorithm,MPA)优化变分模态分解的滚动轴承故障诊断方法。首先,利用以包络熵为适应度函数的海洋捕食者算法对变分模态分解算法的模态个数K和二次惩罚因子α进行自适应选定;其次,使用获得的最佳参数组合对故障振动信号进行变分模态分解,得到多个本征模态分量(Intrinsic Mode Function,IMF);最后,计算各模态分量的平方包络基尼系数(Squared Envelope Gini Index,SEGI),选择系数最大的模态作为最优IMF并进行包络分析,提取相应的故障特征频率。通过公开数据集和实验数据验证表明该方法可解决VMD受参数设置影响的问题,成功诊断轴承故障。且相比于峭度和相关系数指标,平方包络基尼系数指标在筛选最优IMF具备更佳的准确性和鲁棒性。Aiming at the problem that the variational mode decomposition(VMD)method is affected by the parameter settings and the sensitive mode components selection when extracting the characteristic frequencies of the rolling bearing faults,a bearing fault diagnosis method based on VMD optimized by the marine predator algorithm(MPA)is constructed.Firstly,the MPA is used to adaptively select the number of modes K and the secondary penalty factorαof the VMD.Then the fault vibration signal is decomposed to obtain intrinsic mode function(IMF).Finally,the squared envelope Gini index(SEGI)of each IMF is calculated and the best IMF corresponding to the largest SEGI is selected for envelope analysis.And the corresponding fault characteristic frequency is extracted.Verification by published data sets and experimental data shows that this method can solve the problem of VMD affected by parameter settings and successfully diagnose the bearing faults.Furthermore,compared with kurtosis and correlation coefficient indicators,SEGI has better accuracy and robustness in screening the best IMF.

关 键 词：故障诊断 滚动轴承 变分模态分解 海洋捕食者算法 

分 类 号：TH133.3[机械工程—机械制造及自动化] TH165.3