Gear faults diagnosis based on the geometric indicators of electrical signals in three-phase induction motors

Although it is commonly used in research and industrial domains, vibration analysis present some notable drawbacks such as important costs and sensitivity issues resulting from the technical difficulties of the sensor's position within the system. Therefore, there is an increasing interest for fault detection using motor current signal analysis (MCSA) in recent years. Some works have been focused on gear fault detection using the induction machine's electrical current. Hence, we propose a novel approach for gear fault diagnosis centered around a set of indicators (Park and Concordia transforms, Frenet-Serret equations,...) as part of a constructed library of fault indicators. These indicators rely on the geometric representations of the trajectories pursued by the electrical quantities of the three-phase motor. This work shows good results that enable the clustering of the healthy and faulty conditions. (C) 2019 Elsevier Ltd. All rights reserved.