Investigation of motor current signature analysis in detecting unbalanced motor windings of an induction motor with sensorless vector control drive

Abstract

Maintaining the efficiency of AC motors in site equipment is important, given the increasing cost of energy. Reduction of motor efficiency from baseline manufacturer data can go undetected until total failure of the equipment is experienced. This paper introduces motor current signature analysis methods used to detect the early onset of motor efficiency reduction in AC motors controlled by modern Sensorless-Vector Variable Speed Control inverters. A step increase in the resistance of one stator winding is simulated in stages. Off-line processing of motor current data signals using data analysis methods developed for the MATLAB platform is used to identify imbalances caused by subtle stator resistance increases. Initial results indicate that small increases in stator resistances can be observed in the motor current signals received after data processing techniques have been used on the measured signals. The test results are presented herein along with details on the research work to be continued.