Recent Trends in Physics-Based, Data-Driven, and Hybrid Approaches for Prognostics and Health Management of Rotating Machinery Elements

Rotating machinery elements, such as bearings and gears, are the most critical components of any industrial and power generation system. Diagnostics and prognostics are the primary elements of prognostics and health management (PHM) for rotating machines. The timely diagnostics and prognostics of rotating machine components are crucial in terms of safety and capital saving. With the ongoing development of sensor technologies, computational power, and Artificial Intelligence (AI) algorithms, the development of diagnostics and prognostics approaches has been drastically boosted during the last decade. There are various approaches used for the condition monitoring of rotating machinery elements, including physics-based, data-driven, and hybrid approaches. This chapter includes a review of all three approaches and indicates the current trends and challenges that persist with the outlined approaches. Unlike existing literature, which remains fragmented and domain-specific, the review exercise will result in a translational roadmap that can guide researchers and practitioners from isolated ideas to integrated intelligence and develop robust, interpretable, and deployable PHM solutions. Also, some future directions are highlighted to use these approaches in an integrated manner for developing reliable and practical solutions for the diagnostics and prognostics of rotating machines.