Modeling of stress corrosion cracking growth rates for key structural materials of nuclear power plant

Stress corrosion cracking in light water reactor is one of the most important factors threatening the safe operation of nuclear power plants. Due to the severity, generality and various safety and economic problems caused by this phenomenon, it is necessary to establish a model for predicting the stress corrosion cracking growth rates. This paper provides an overview of three main methods for predicting stress corrosion cracking growth rates in recent decades, i.e., empirical, deterministic and calculation methods, which are introduced in detail. Empirical models describe classical statistical analysis and emerging artificial neural network method, both of which are based on a large number of experimental test data mining. They are convenient and relatively accurate in predicting, but require extensive, time-consuming and expensive tests for different service environments. Deterministic models aim to establish a theoretical relationship between crack growth rate and various influencing parameters by studying the stress corrosion cracking mechanism. Many scholars have proposed different mechanisms to scientifically explain the stress corrosion cracking phenomenon and propose corresponding crack growth rate models. Calculation models reveal the mechanism of crack initiation and propagation in different layers of materials by means of finite element method based on fracture mechanics and multiscale method based on quantum mechanics. They provide new idea for future research on stress corrosion cracking and bridge the quantitative mechanism or model, but no specific stress corrosion cracking growth rate model is formed. The article concludes with the prospect, aim and direction for stress corrosion cracking mechanism and prediction model.