Preliminary analysis of bearing current faults for predictive maintenance

Abstract

Bearings faults, one of the most common mechanical failures in the electrical machine, have been the diagnostic interest for a long time. The questions related to bearing currents have become an important issue due to a growing number of motors running with variable-speed drives. This paper introduces a bearing current problem presenting the leading causes and damages of induction machine bearing currents. As the world is moving towards Industry 4.0 standards and fault prediction in production is becoming an extremely crucial topic, this paper presents a pre-processing of training datasets and possibilities for predictive maintenance. The experimental results of fault implementation in a laboratory environment to collect training datasets are discussed.