Systematic Review of Bearing Component Failure: Strategies for Diagnosis and Prognosis in Rotating Machinery

Abstract

The rapid advancement of industrial technologies has underscored the importance of effective diagnosis and prognosis in equipment maintenance to ensure safe operations. This is particularly critical in rotating machinery (RM), where bearing components play a pivotal role in determining the health and lifespan of systems such as wind turbines and high-speed trains. With the expanding application of bearings, the literature on predicting remaining useful life (RUL) and fault classification has become increasingly vital. This review paper provides a comprehensive overview of the current research on diagnosis and prognosis strategies for bearing-related faults and anomalies. Initially, it presents updated fault statistics in RMs, offering a detailed analysis and summary of fault cases, their effects, and identification techniques. The paper then delves into theoretical insights into fault frequencies and the latest failure detection strategies for bearing elements, emphasizing current and vibration analysis. The review further assesses advancements in sensor technologies for data acquisition and examines the most utilized bearing data repositories for fault classification and run-to-failure analysis. Additionally, it also explores the recent literature on fault diagnosis strategies for bearings, categorizing the approaches into model-based, knowledge-based, and pattern recognition frameworks. Regarding prognosis, the paper reviews methodologies grounded in statistical, model-based, and data-driven strategies. By highlighting contributions from the past five years, this paper aims to provide a thorough analysis of methodologies in the diagnosis and prognosis of bearing faults, offering valuable insights and directions for future research in this critical field.