Intelligent Diagnosis of Broken Bars in Induction Motors Based on New Features in Vibration Spectrum

Abstract

This paper presents an intelligent method for diagnosing broken bars in induction motors. The method is based on training a neural network using new features extracted from vibration spectrum. These fault related features depend on slip. The exact value of slip can be determined using vibration spectrum; therefore, a vibration sensor is the only required sensor. The method has been able to diagnose correctly in all the laboratory tests.