Open Access Open Access  Restricted Access Subscription or Fee Access

Bearing Fault Diagnosis Using ANN and SVM with the Help of Wavelet Transform Based Features

Anshul Chaturvedi, Pratesh Jayaswal, Deepak Kumar Gaud


Many researchers use vibration signals as fault characteristics of any rotating bearing for fault detection and use various optimization techniques available for fault of bearing classification. The method of fault classification uses few neural network (NN) geometry and parameters necessary for the formulation. There is no derived formulation which can be used to select the optimal values for the network parameters. The parameters which are required to be calculated, impacts the statistical calculation directly. This paper investigates the effects and compares the results obtained from the calculation of artificial neural network (ANN) and support vector machine (SVM, a machine learning tool) on rolling element bearing fault diagnosis. Recording of the signals is performed for healthy bearing, bearing with inner race fault, outer race fault and rolling element fault created using an EDM machine. The various features of vibration are calculated with the help of wavelet transform feature extraction methods. Here a comparative experimental study for the effectiveness of ANN and SVM is carried out. The results obtained shows that for this study, SVM is a better classifier than ANN.

Keywords: Artificial neural network, support vector machine, MATLAB, condition based maintenance (CBM), wavelet transform feature extraction

Cite this Article

Anshul Chaturvedi, Pratesh Jayaswal, Deepak Kumar Gaud. Bearing Fault Diagnosis Using ANN and SVM with the help of Wavelet Transform Based Features. Recent Trends in Electronics & Communication Systems. 2017; 4(2): 26–34p.

Full Text: PDF


  • There are currently no refbacks.