Naïve Bayes classifier for temporary short circuit fault detection in stator winding

This paper is proposing Naïve Bayes classifier detection system to identify the symptom of stator winding deterioration. The proposed system is based on probabilistic classifier with strong independence assumption of each fault case. The temporary short circuit case is defined as non permanent short circuit fault with high impedance. This fault case is representing the early stage of stator insulation break down. The laboratory experiment is performed to simulate the fault cases consist of induction motor with stator modification and current measurement system. The detection system is trained to identify the temporary short circuit occurrence consist of transient starting, steady state and ending of temporary short circuit. The system is also tested using non trained data to clarify the detection performance.