Intelligent Manufacturing Management Systems

7Intelligent Machine Learning and Deep Learning Techniques for Bearings Fault Detection and Decision-Making Strategies

Jagadeesha T.*, Thutupalli Srinivasa Advaith, Choppala Sarath Wesley, Grandhi Sri Sai Charith and Doppalapudi Manohar

Department of Mechanical Engineering, National Institute of Technology Calicut, Kerala, India

Abstract

Conditional monitoring of machine elements is an important aspect which leads to the huge boost in the production outputs from the manufacturing firms as we see in the industrial world today. It aids in predicting the failure of machines beforehand thereby drastically reducing the production lead times and leading to maintaining the optimal production run. As a result, many intelligent fault detecting systems have become popular in the recent times whose goal is to accurately and precisely predict and classify the machine faults.

Advanced machine learning algorithms have been proven to effectively classify the machine faults in accordance with the industry standards. In this work, standard Case Western Reserve University bearing fault dataset is used to build efficient and optimized machine learning models that can classify the bearing faults accurately. Bearings are a crucial part of rotating machinery and the monitoring of their health is vital to maintain optimal production in manufacturing firms. The dataset consists of the vibrational amplitude of the shaft which is mounted on rolling element bearings on both ends of the shaft. The vibrations ...