Failure identification of induction motors by sensing unbalanced stator currents

A technique suitable for real-time failure monitoring of cage induction motors is presented. The proposed method utilizes online sensing of the unbalance in the stator current signal to assess the integrity of the motor. This approach is extremely effective where the system neutral is high resistance grounded. The sensor is inexpensive, noninvasive, reliable, and easy to install. Instead of analyzing the spectral components of the signal, the absolute magnitudes are used for identifying the incipient failure; therefore, the need for computing and data processing is partly eliminated. This method is tested for wye connected motors. The results from laboratory investigations are presented. The usefulness of this approach is demonstrated with the aid of simulated fault conditions in the experimental motor. The efficacy of the proposed method in industrial power systems is discussed