A COMPARATIVE INVESTIGATION BETWEEN THE MCSA METHOD AND THE HILBERT TRANSFORM FOR BROKEN ROTOR BAR FAULT DIAGNOSTICS, IN A CLOSED-LOOP THREE-PHASE INDUCTION MOTOR

Although it is one of the most common fault signatures methods for diagnosing motor defects, Motor Current Signals Analysis (MCSA) has some drawbacks that can degrade the performance and accuracy of a motor diagnostics system. In particular, it is very difficult to detect broken rotor bars (BRB) when the motor operates at no load condition. This paper proposes a comparative study of two methods for the diagnosis of BRB defect in closed-loop induction motors. Stator currents in an induction motor have been measured, transferred to a computer, used for MCSA, and advanced signal-processing algorithms. They consist of analyzing the envelope of one-stator phase current. This envelope is obtained, using the Hilbert transformation (HT). Both methods are based on spectral analysis via Fast Fourier Transform (FFT). Both methods have proved their effectiveness to detect BRBs under different load conditions. However, the HT is considered more efficient compared to the MCAS method in the case where the motor operating at no load condition. The simulation results obtained for the two methods applied on a threephase squirrel-cage induction motor controlled are discussed.