An extended model of induction machines with saturation suitable for fault analysis

Abstract

Literature review suggests that modeling of induction machines with saturation has received considerable attention. Most of these models are very application specific. None of them, however, seems to be completely adequate for the purpose of fault diagnosis. This is because for noninvasive detection technique such as motor current signature analysis (MCSA), one has to detect different frequency components in the line current of the machine. Since saturation is present to a certain extent in almost any machine, incorporating it in the model creates a condition to analyze and detect symptomatic harmonics in the line current spectra that may arise due to the interaction of saturation with a specific fault. The present paper attempts to create such a model. The model uses modulation of air gap permeance in order to emulate saturation. Modified winding function approach (MWFA) has been used to model the induction machine since the air gap becomes nonuniform in presence of saturation. To demonstrate the effectiveness of the model, certain high frequency harmonics that can be predicted theoretically, are shown to be present in the line current of the simulated machine. Experimental results also confirm the presence of these harmonics. The author plans to use this model in future for fault analysis.