An Induction Motor Fault Diagnosis Using Space Phasor Method and Artificial Neural Network

Abstract

This paper showcases a unique technique for the fault identification in induction motor using the Space Phasor Method (SPM), Artificial Neural Networks (ANN) and Ant Colony Optimization (ACO). The method aims to classify four distinct motor conditions: healthy, interturn fault, bearing fault, and rotor-bar crack. 3-phase space phasor is derived from the motor stator currents using SPM. Further Fourteen statistical features are extracted from the space phasor current. To enhance the feature selection process, ACO is applied, resulting in a significant reduction in the number of input parameters for the ANN. This optimization leads to an improvement in classification accuracy, achieving a 100% accuracy rate. The proposed methodology demonstrates high effectiveness in fault diagnosis of induction motor, offering a robust, efficient, and cost-effective solution for real-time motor monitoring. The results suggest that the integration of SPM, ACO, and ANN holds significant potential for improving fault diagnosis in induction motor.