A Data-Driven Approach Based Bearing Faults Detection and Diagnosis: A Review

Abstract

Monitoring the condition of rotating machines is essential for system safety, reducing costs, and increasing reliability. This paper tries to present a comprehensive review of the previously conducted research concerning bearing faults detection and diagnosis based on what is known as model-free or data-driven approaches. Mainly, two data-driven approaches are discussed, which are statistical-based approaches and artificial intelligence-based approaches. The employed condition monitoring techniques in diagnosing faults in different machinery are also deliberated. These include vibration, motor current signature, and acoustic emission signals analysis as they are widely utilized in condition monitoring based data-driven approaches. The advantages, limitations, and practical implications of each approach and technique are presented. However, it has been concluded that very few studies have adopted the statistical-based approach for bearings health monitoring. Thus, it is advised that more investigations have to be conducted in this regard, and hence it will be our next aim.