Application of Empirical Wavelet Transform for Analyzing Inter-turn Fault in FEM Based Closed Loop Speed Controlled Induction Motor

The objective of the work is to analyze the finite element method based inter-turn fault in the stator winding of an induction motor (IM) controlled in a closed loop speed manner by utilizing empirical wavelet transform. Diagnosing inter-turn fault involving less than 5% of turns is a major challenge for electrical machine researchers. The analysis of such faults becomes more complex when the machine is operated from an inverter in a closed loop speed controlled manner due to its operation at variable speed and varying load. In this direction, a 7.5 kW IM is modeled in ANSYS Maxwell with various inter-turn fault severities like 30T (12.5%), 15T (6.25%) & 10T (4.2%) and correlated with ANSYS Simplorer for power supply and loading measures. MATLAB Simulink is utilized to model a closed loop speed controlled inverter and co-simulated with Simplorer to synthesize all the tools in a collective framework for realistic simulation. The airgap flux density analysis shows additional noise harmonics in its spatial FFT spectrum for shorted turns involving even less than 5%. Then empirical wavelet transform is used to analyze the line current data obtained from finite element IM model for various fault severities. The result shows that amplitude of a decomposed component of line current increases in a particular frequency band for 10T short compared to healthy data whereas additional components are induced in the spectrum when fault severity increases to 30T short.