基于优化变分模态分解和包络峭度的轴承故障诊断

Bearing Fault Diagnosis Based on Optimized VMD and Envelope Kurtosis

针对变分模态分解(VMD)的分解层数K和惩罚因子α难以选择问题,提出了用减法平均优化器(SABO)对参数寻优的方法。首先,采用SABO对K和α进行寻优,输出最优参数组合并代入到VMD中,将原始振动信号分解得到K个模态分量;然后,用最大包络峭度为指标提取K个模态分量中峭度最大的分量作为最优分量,并计算其相关时域和熵理论特征参数构造特征向量样本集;最后,将特征向量样本集输入到经网格搜索和五折交叉验证调参的支持向量机(SVM)中进行故障诊断。为了验证该方法的有效性,利用凯斯西储大学轴承数据集进行实验,实验结果表明:该方法分类效果更好,准确率达到99.44%;基于江南大学3种不同工况的轴承数据实验,最终故障诊断准确率都达到了95%以上。

In view of the difficulty in selecting the decomposition layer K and penalty factorαof variational mode decomposition(VMD),a subtraction-average-based optimizer(SABO)is proposed to optimize the parameters.Firstly,the SABO is used to optimize K andα,output the optimal parameter combination,and substitute it into VMD to decompose the original vibration signal into K modal components.Then,the maximum envelope kurtosis is used as the index to extract the component with the largest kurtosis among the K modal components as the optimal component,and the eigenvector sample set is constructed by calculating the relevant time-domain and entropy theory characteristic parameters of the optimal component.Finally,the eigenvector sample set is input into the support vector machine(SVM)with mesh search and 5-fold cross-validation for fault diagnosis.To verify the effectiveness of this method,experiments were conducted using the bearing dataset from Case Western Reserve University.The experimental results show that the classification effect of the method is better,and the accuracy rate is 99.44%.Based on the bearing data set experiments of three different working conditions in Jiangnan University,the final fault diagnosis accuracy rate reaches more than 95%.