The cavitation erosion failure of pumps or valves induces the low efficiency and reduced service life of nuclear reactors. This paper reviews works regarding the cavitation erosion of pumps and valves in the nuclear power industry and academic research field. The cavitation erosion mechanisms of materials of pumps and valves are related to the microstructure and mechanical properties of the surface layer. The cavitation erosion resistance of austenitic stainless steel can be ten times higher than that of ferritic steel. The cavitation erosion of materials is related to the hardness, toughness, and martensitic transformation capacity. Erosion wear and erosion–corrosion research is also reviewed. Erosion wear is mainly influenced by the hardness of the material surface. Erosion–corrosion behavior is closely connected with the element composition. Measures for improving the cavitation erosion of pumps and valves are summarized in this paper. The cavitation erosion resistance of metallic materials can be enhanced by adding elements and coatings. Adhesion, inclusion content, and residual stress impact the cavitation erosion of materials with coatings.

The very high cycle fatigue (VHCF) property of TC21 titanium alloy blunt-notched specimens were investigated by using an ultrasonic fatigue test machine with a frequency of 20 kHz. S–N of blunt-notched specimens illustrated a continuous decrease characteristic with a horizontal line over the 10⁵–10⁹ cycle regimes. However, the fatigue life showed a large scatter for blunt-notched specimens. Blunt-notch significantly reduced the fatigue property in the high cycle and very high cycle regimes compared with that of smooth specimens. The crack initiation modes for blunt-notched specimens in the very high cycle regime can be divided into three types: (i) surface initiation, (ii) subsurface with flat facet, and (iii) subsurface with “facet + fine granular area”. The crack initiation mechanism of blunt-notched specimens is discussed in view of the interaction of notch stress gradient distribution and heterogeneous microstructure. Furthermore, the fatigue limit model based on the theory of critical distance (TCD) was modified for the very high cycle regime, and the scatter of the fatigue property of the blunt-notched specimens were well predicted by using this model.

Cavitation damage has not been well predicted because of its complex relationship of many mechanical and microstructural factors. An artificial neural network approach of the back-propagation network was used to predict cavitation damage of stainless steels, 316L and 420, in terms of the significant influence of cavitation time, roughness, and residual stress on cavitation damage. Mean depth of erosion was used to quantitatively describe cavitation damage of 316L and 420. Prediction accuracy was improved by analyzing the effects of the number and type of input nodes, the number of nodes in the hidden layer, and different activation functions on prediction accuracy. The best performance was in the model with the input nodes of cavitation time and roughness, eleven nodes in the hidden layer, and the activation function of logsig.