ARM based induction motor fault detection using wavelet and support vector machine

This paper proposes a Motor Current Signature Analysis technique to diagnose the stator inter turn fault in induction motor using wavelet transform and Support Vector Machine as tools. This technique is based on the analysis of stator current under both healthy and faulty conditions. Lifting algorithm based wavelet decomposition is implemented in ARM processor based system using which the approximation and detail coefficients of the three phase induction motor stator current signal are obtained. Standard deviation of these wavelet coefficients is identified as the exact feature for stator inter turn fault and fed to SVM which is then used to classify the fault. The whole system for condition monitoring is realized in ARM processor based board to reduce the implementation cost.