Enhanced Multi-Synchro-Squeezing Transform for Fault Diagnosis in Induction Machine Based on Third-Order Energy Operator of Stator Current Signature

Abstract

In traditional motor current signature analysis (MCSA) approach, the spectral leakage of the fundamental supply frequency can obscure the characteristic fault component under low-load or no-load conditions. Furthermore, most time-frequency (TF) methods often have low resolution and are not qualified to produce a narrow band in the output. In this paper, we employ multi-synchro-squeezing transform (MSST) to show its effectiveness in fault detection of induction machines, for the first time. The key innovation of this work is merging MSST (due to its high time-frequency resolution) with the third-order energy operator (TOEO) (due to its high accuracy in fault detection). Specifically, TOEO is used to overcome the leakage effects of the supply frequency, through a demodulation approach for asymmetric fault detection along with MSST technique. The proposed method was evaluated for induction machine fault detection in both steady-state and transient conditions. Both analytical and experimental results confirm that the proposed method can excellently reveal the fault characteristic frequency in steady-state and transient mode, instead of the sideband frequencies.

Keywords

• Fault diagnosis
• condition monitoring
• induction machine
• fast Fourier transform
• current signature