Rotating Element Bearing Fault Diagnosis Using Discrete Cosine Transform and Supervised Machine Learning Algorithm

Abstract

Motors are the driving force of our industrial world, as they power approximately 85% of all rotating machines. This revolutionary invention has been through radical changes before entering into the commercial industries, and their present forms are very reliable, to say the least. However, despite being so robust, induction motors are not entirely fault-proof and are more vulnerable to the internal faults than the external ones. Among the internal faults, certain types of bearing faults are more frequent, and their effects range from various performance-related issues to hard motor breakdowns. Fortunately, the recent advancements in the fields of Digital Signal Processing and Machine Learning allow us to detect these bearing faults and Figure out their origins, which in turn enables us to preserve their health and take measures against breakdowns. Through vibration analysis, this paper proposes a powerful method to detect these faults and differentiate among them based on the location of their occurrence within the bearing. Utilizing a well-known signal processing technique called Discrete Cosine Transform and Decision Tree classifier, this method is capable of classifying the motor bearing states with a 99.4% accuracy.