Wireless monitoring of power quality disturbances in the supply to induction motor

Abstract

The lifespan of electrical machines depends both on the type of operating load and on the nature of the supply on which it is operating. When there are unexpected aberrations occurring in the latter, it leads to Power Quality(PQ) disturbances. Among several PQ disturbances, voltage sag and swell are predominant in the case of inverter driven induction motors (IM). Hence condition monitoring (CM) of induction motor should also include the monitoring of the supply. Presently, CM of induction motors is done using traditional wired systems. This paper illustrates a methodology to wirelessly monitor voltage sag and swell in the supply to induction motor. Wireless monitoring is employed using ZigBee based Wireless Sensor Network (WSN). This WSN samples the stator current of the induction motor at a frequency of 1.8 KHz. Wavelet transform is applied for the stator current for feature extraction. Daubechies wavelet (db9) has emerged to be the most effective in the proposed work after several comparisons between different wavelets. These wavelet coefficients are fed as input to Artificial Neural Network (ANN) and k-Nearest Neighbour (kNN) classifiers for final classification. It is seen that ANN has a better classification accuracy of 92.5% as compared to kNN with a classification accuracy of 86.36%.