Machinery fault feature extraction based on independent component analysis and correlation coefficient

A machinery fault feature extraction method was proposed based on independent component analysis (ICA) and correlation coefficient. The ICA was used for analysis of vibration signals with different fault category. The extracted independent components include the information of the fault. The sum of absolute values of correlation coefficients of the test sample and the extracted indepent components of each category was used as a feature vetor. Then the support vector machine was used as a classification method for fault diagnosis. The proposed fault feature extraction method has been applied to two tasks: gear feault diagnosis and roller bearing fault diagnosis. Experiments demonstrate that the ICA of each fault category and the correlation coefficient can extract useful features for machinery fault diagnosis.