Prediction Of Stress Corrosion Crack Growth Rate In Nuclear Power Plant Structural Materials

Intergranular stress corrosion cracking(IGSCC) in LWR coolant circuits is one of the most important factor that affects the safety of the nuclear power plants. Considering this problem is serious and widely spread, also it causes different kinds of safety and econonic problems, it is necessary to establish a prediction model. If this can be done successfully, tremendous economic and safety benefits would accrue to the operators, because it would allow the reactors to be operated in a manner to reduce both the risk and cost of damage and to enhance safety.A crack growth rate(CGR) database for 304 SS was developed from data reported in the literature and an artificial neural network(ANN) was developed. The results show that the ANN reveals the underlying relationships that map the dependencies of the dependent variable(CGR) on the various input independent variables. Based on the prediction results from the ANN, the coupled-environment fracture model(CEFM) was customized to take into consideration the impact of the degree of sensitization on crack propagation rate and was then used to predict IGSCC CGR in Type 304 SS in simulated Boiling Water Reactor(BWR) primary coolant circuits. Comparison of the ANN-predicted CGR and the CGR predicted by the CEFM as a function of the independent variables reveals good agreement between these two approaches. A sensitivity analysis revealed that IGSCC in sensitized Type 304 SS in high temperature aqueous environments is primarily electrochemical in character, but with significant contributions from mechanics and microstructure.A crack growth rate database for alloy 600 was developed from data reported in the literature and an artificial neural network was developed and trained using this database. The results show that the ANN reveals the dependencies of the dependent variable(CGR) on the various input independent variables. The CEFM model was customized to predict the CGR of alloy 600 in simulated Pressurized Water Reactor(PWR). The prediction results perfectly explain the effect of stress intensity factor and yield streng on the CGR. A sensitivity analysis has been done based on the fuzzy curve for alloy 600. The results indicate that IGSCC in alloy 600 in simulated PWR environment is firstly mechanical in character, but with almost the same contribution from electrochemical.Using the CEFM model, the effect of the standard deviation(SD) of input parameters on the distribution of CGR for 304 SS in simulated BWR environment has been studied. The results indicate that only the SD of electrochemical potential can change the distribution form of CGR. However, if the distribution of all the input parameters obey normal distribution, then the distribution of CGR obey normal distribution too.