For developing fatigue design curve of cast stainless steel that is used in piping material of nuclear power plants, a low-cycle fatigue test rig was built. It is capable of performing tests in pressurized high temperature water environment of PWR. Cylindrical solid fatigue specimens of CF8M were used for the strain-controlled environmental fatigue tests. Fatigue life was measured in terms of the number of cycles with the variation of strain amplitude at 0.04%/s strain rates. The disparity between target length and measured length of specimens was corrected by using finite element method. The corrected test results showed similar fatigue life trend with other previous results.

Environmental fatigue crack propagation of CF8M and CF8A steels used in the domestic nuclear power plants (NPPs) were investigated on the simulated pressurized water reactor (PWR) condition (temperature: 316°C, pressure: 15MPa). The test equipment for environmental fatigue (high temperature-high pressure loop, autoclave, load frame, and measurement system) was designed. As-received and 60-year aged specimens were used in the test. To compare with environmental fatigue test, another test in the air condition was performed. The fracture surfaces of specimens were difficult to verify the fracture modes such as striation, inter-granular crack and cleavage and so on. As the ferrite content of CF8M is increased, more particles on the fracture surface were peeled.

No abstract

No abstract

CF8M cast austenitic stainless steel is used for several components such as primary coolant piping, pump casing, and valve bodies in light water reactors. Thermal aging results in spinodal decomposition of the delta-ferrite leading to increased strength and decreased toughness. Material properties depend on multiple variables including chemical composition, ferrite content, aging time, aging temperature and so on. Various algorithms have been developed to predict the results of these multiple variables, and studies to predict the results using artificial neural networks have been performed actively. In this study, we performed the neural network training using the measured data by Argonne national lab. and others and obtained the Charpy impact energy considering the various aging conditions through the experiments. The results of ferrite content and Charpy impact energy considering thermal embrittlement in CF8M material using the trained neural network are well corrected with the measured results. The material property change of CF8M by thermal embrittlement can be effectively predicted using the result of this study.