A wavelet-based approach to diagnostic and monitoring for ac drives

Industrial applications require, in a more and more impelling way, the consumptions of every load to be monitored, in order to increase the efficiency of the industrial process, ensure that the injection of electrical disturbances into the supply network is kept within the allowed limits, and detect possible deviations from the optimal working conditions that might lead to failures, misoperations or even faults if not removed in due time. In a previous paper the authors showed that the information coming from a measurement system for electric power monitoring can be usefully exploited to realize a sensorless instrument able to monitor an ac motor drive and provide useful information about the drive status in order to prevent that anomalous operating conditions may evolve towards faults. In this paper this analysis is performed with the help of a suitable wavelet transform. Different bases are applied and the results compared.