A current monitoring system for diagnosing electrical failures in induction motors

Induction motors are critical components in industrial processes. A motor failure may yield an unexpected interruption at the industrial plant, with consequences in costs, product quality, and safety. Many of these faulty situations in three phase induction motors have an electrical reason. Among different detection approaches proposed in the literature, those based on stator current monitoring are advantageous due to its non-invasive properties. One of these techniques resorts to spectrum analysis of machine line current. Another non-invasive technique is the Extended Park's Vector Approach, which allows the detection of inter-turn short circuits in the stator winding. This article presents the development of an on-line current monitoring system that uses both techniques for fault detection and diagnosis in the stator and in the rotor. Based on experimental observations and on the knowledge of the electrical machine, a knowledge-based system was constructed in order to carry out the diagnosis task from these estimated data.