Mid Sweden University

In most industrial applications, Fault Detection (FD) is an important function. Understanding equipment's conditions help engineers and operators avoiding the inevitable catastrophe that could lead to a significate impact on system reliability and safety. For electromechanical devices such as asynchronous motors, which play an essential role in production lines, mastering vibrations for Fault Detection is interesting. Downtime, costly maintenance, and energy waste are damages resulting from faults. The paper presents an overview of different techniques used in asynchronous motors fault detection. Vibrations signals processing and techniques for extracting characteristic features from them are largely developed. Spectrum analysis of the envelope signal is employed to process vibration signals resulting from rolling bearings. The paper also proposes rotors broken bars detection using Park Transformation. It was shown that Park's transformation is an effective tool to detect broken bars even at the incipient stage, which gives to it the merit to be a complementary technique to vibration analysis. © 2020 IEEE.