A Method Using Numerical Simulation and Support Vector Machine to Detect Faults in Bearings

The bearings are important parts in transmission machinery, and its running state directly affects the entire machine, including reliability and stability. Therefore, the efficient diagnosis method to detect faults and distinguish fault types at an early stage in bearings is necessary. This paper presents a fault diagnosis method for bearings using finite element method (FEM), wavelet packet transform (WPT) and support vector machine (SVM). In this method, the accurate FE models of the faulty bearing (faults in inner race and rolling body) are constructed and the vibration signals of the faulty bearing are obtained. Decompose the vibration signal into eight signal components using WPT and ten time-domain feature parameters of all the signal components are calculated to generate the training samples to train the SVM, the measured vibration signal and its components decomposed by WPT serve as a test sample to the trained SVM. The condition of bearings can be determined finally. Practical examples of diagnosis for a bearing are investigated to verify the effectiveness of the methods proposed in this paper. The fault types are finally determined and the classification accuracy rates of inner race faults and rolling body faults are 79%, 81%, respectively.