Application of the Prony’s method for induction machine stator fault diagnosis

The fundamental positive and negative sequence components of the voltage and current are always characters in induction machine stator fault diagnosis. This paper presents the application of the Pronypsilas method for deriving these components from the measured data. It is known that the Pronypsilas method is sensitive to noise and takes a long computation time by increasing of the length of the data. In order to enhance its performance, pretreatments, including frame transformation, frequency shifting and decimation, are employed before using the Pronypsilas method. Finally, the fundamental sequence components, figured out by using the Pronypsilas method, are put into a neural network to calculate the indicator for diagnosis. The proposed technique is verified by applying to diagnose the stator fault in a three-phase induction motor.