Artificial Intelligence-based Fault Diagnosis Procedure for a Sustainable Manufacturing Industry

Abstract

Abstract All industries are fast transforming into smart industries as part of the sustainable developments in the fourth industrial revolution. Predictive maintenance is one of the most important aspects of such smart industries, to avoid unanticipated machine breakdowns and catastrophic failures. Machine vibration analysis is a common tool for predicting the state of machinery. Vibration analysis involves analysing vibration data collected from machinery and determining whether or not a fault exists. Despite the fact that different methods are utilized to handle data, artificial intelligence is capable of processing such data without the need for human intervention. Every day, a substantial amount of study is carried out in this field. New strategies, on the other hand, that yield greater classification accuracy have yet to be developed. With the use of artificial intelligence approaches, this research article attempts to offer an effective defect detection method for rolling element bearings. To illustrate the practical applications, the technique is used on real datasets which were developed by Case Western Reserve University, which is regarded as a gold standard for testing diagnostic algorithms.