Space vector analysis for the diagnosis of high frequency amplitude and phase modulations in induction motor stator current

International audienceThis paper deals with mechanical fault monitoring in induction machines from stator current measurements. At variable speed, the different faults can be characterized by their time-frequency signatures. However, the time-frequency representations apply to complex signals for a univocal phase and modulus definition. Under usual modulation conditions, this complex signal is the analytic signal obtained through the Hilbert Transform of the real measured signal. Indeed, classical mechanical faults result in modulation frequencies which are lower than the carrier frequency. The analytic signal complex envelope and instantaneous frequency thus provide the amplitude and phase modulations respectively. However, bearing faults, for instance, may produce high frequency modulations. Then, the analytic signal phase and modulus each carry information on both phase and amplitude modulations. This study proposes an alternative complex signal representation in this case. The proposed representation, namely the Concordia Transform, takes advantage of the additional information available in the case of a three phase machine. By using two stator currents, the Concordia Transform builds a complex vector which conveniently carries the information about phase and amplitude modulations. This paper applies the Hilbert and Concordia Transforms to simulated and experimental signals and provides a diagnosis of amplitude or phase modulation in various conditions