A review of prediction methods of cavitation erosion

Cavitation and its resulting erosion exert a significant influence on the performance and longevity of hydraulic machinery. This study provides a comprehensive review of numerical prediction methods for cavitation erosion, emphasizing computational approaches. Although traditional experimental techniques (high-speed imaging and acoustic measurements) allow direct observation, their high cost and limited applicability under complex operating conditions restrict their practicality. Therefore, computational fluid dynamics serves as an essential alternative for evaluating cavitation erosion. This review systematically examines mainstream prediction methods, including fluid–structure interaction (FSI), micro-jet models, and energy-balance approaches, and elaborates on their theoretical foundations, computational frameworks, and applicable domains. In recent years, energy-balance models and advanced turbulence simulations have enhanced prediction accuracy through measurable indicators such as local energy density and impact energy. These models are widely applied to investigate multi-bubble cavitation mechanisms and to predict local impact intensity, providing high computational efficiency under relatively simple flow conditions. Micro-jet models estimate wall loading through parameters such as jet velocity and peak pressure, making them effective for assessing erosion from near-wall single-bubble collapse; however, their performance is constrained in multi-bubble interactions and asymmetric collapse scenarios. FSI methods, employing variables such as wall pressure, material stress, and plastic deformation, enable high-fidelity multiphysics coupling and yield detailed simulations of single-bubble cavitation and fatigue damage, although at substantial computational cost. They are therefore more appropriate for localized mechanism analysis and high-precision design optimization. Despite substantial progress, challenges remain in multiphysics coupling, experimental validation, and data-driven model optimization. This review aims to provide theoretical insights that guide future research on cavitation-erosion modeling and its engineering applications.