Motor Current Signature Analysis has been successfully used in induction machines for fault diagnosis. The method however does not always achieve good results when the load torque is not constant. For proper results temporal analysis on determined frequency harmonics is needed. This paper proposes an automatic system for fault analysis based on wavelet functions, which allows on-line fault detection on time domain achieving good experimental results on stationary and non stationary state.

Motor Current Signature Analysis has been successfully used in induction machines for fault diagnosis. The method, however, does not always achieve good results when the load torque is not constant. For proper results, temporal analysis on certain frequency harmonics is needed. This paper proposes an automatic system for fault analysis based on wavelet functions, which allows on-line fault detection in the time domain, achieving good experimental results both on stationary and non-stationary states.