Evaluation of intelligent approaches for motor diagnosis under changing operational conditions

Abstract

The diagnosis of electric machines, such as induction motors, is one of the key tasks that needs to be performed to guarantee their right operation as electromechanical energy converters in most industrial facilities. The ability to reliably identify a mechanical fault occurrence in an induction motor, before it became catastrophic, can reduce risks related to the productive chain. Recently, different intelligent approaches have been proposed to develop feature-based methods for the automatic fault diagnosis in induction motors. This article provides an evaluation of different machine learning techniques for fault identification that come from different families. The datasets are formed to allow the performance analysis of the results when the classifier is trained with data obtained from some particular operational conditions, and then it is tested under different operating conditions, as it is usual in industry. The input information is obtained from current signals of an induction motor with one broken rotor bar.