基于小波与AR模型融合的滚动轴承故障诊断方法

Fault Diagnosis of Rolling Bearing Based on Wavelet Transformation and AR Model Fusion

提出了一种基于小波与AR模型的SVM的故障诊断方法,对故障轴承(型号6205-2RS JEM SKF)振动信号进行分析。该方法首先对滚动轴承振动信号进行小波变换,通过变换提取出每层的小波系数;然后对每层小波系数建立AR模型,最后将自回归模型的参数作为特征向量输入SVM分类器。通过SVM分类器来识别滚动轴承的故障类型,实验结果验证了该方法的有效性。

A fault diagnosis method based on SVM for wavelet and AR model is proposed,and the vibration signal of fault bearing（Model 6205-2RS JEM SKF） is analyzed in this paper.Firstly,the wavelet transform of the rolling bearing vibration signal is extracted,and the wavelet coefficients of each layer are extracted by transform.Then,the AR model is established for each wavelet coefficient.Finally,the autoregressive model parameters are input to the SVM classifier as the eigenvector.The SVM classifier is used to identify the fault type of the rolling bearing.