Predictive Monitoring of Incipient Faults in Rotating Machinery: A Systematic Review from Data Acquisition to Artificial Intelligence

Predictive maintenance is one of the major tasks in today’s modern industries. All rotating machines consisting of rotating elements such as gears, bearings etc are considered as the backbone of any plant and condition based maintenance of these elements is at the top priority to keep them available all the time. Due to failure of these elements, whole system can lead to complete shutdown. A predictive monitoring program generally consists of four technical processes, i.e., data acquisition, pre-processing (denoising process), feature processing, and artificial intelligence. Over recent years, a significant amount of research work has been undertaken in each of the four processes. There has been a significant amount of literature available however, lack of a systematic review which encapsulate all four technical processes comprehensively. To fill this gap, this paper provides a review on predictive monitoring of incipient stage faults following its whole program, i.e., from data acquisition to artificial intelligence implementation. Firstly, various commonly used data acquisition methods are introduced and discussed. Then, advanced signal processing methods for the de-noising of signals are discussed. Afterwards, feature processing methods are summarized by explaining its major tasks and existing approaches. Finally, artificial intelligence based fault prediction approaches are discussed and summarized.