Improved Techniques for Analyzing Broken Rotor Bars at Different Supply Frequencies in Inverter-Fed Induction Machines

This study introduces enhanced multi-dimensional demodulation techniques and their use in the analysis and detection of faulty rotor bars in an induction machine driven by an inverter. The proposed approaches involve squaring each phase’s stator current before applying multi-dimensional demodulation techniques to transform it into two-dimensional in-phase and quadrature components. The modulus signal of these components serves as a failure signature, based on frequency domain analysis. The theoretical validity and empirical confirmation of these methods are demonstrated using offline experimental signals collected from testing three induction machines operating under both low and high load conditions. The machines were operated using an inverter that could handle several frequency ranges, including 25 Hz, 40 Hz, and 50 Hz. Experimental results across various operational conditions underscore the efficacy of the proposed methodologies when compared to conventional multi-dimensional demodulation techniques and Motor Current Signature Analysis (MCSA). The proposed approaches demonstrate significant effectiveness, especially in detecting faults related to broken rotor bars under different supply frequencies and low-load conditions in inverter-fed induction machines.