Prediction of Residual Stress for Dissimilar Metals Welding at Nuclear Power Plants Using Fuzzy Neural Network Models

The factors that have an impact upon fatigue strength are residual stress, stress concentration, the mechanical properties of the material, and the macrostructure and microstructure. Residual stress is one of the most important factors but its effect on high-cycle fatigue is of more concern than the other factors. Residual stress is a tension or compression that exists in a material without any external load being applied, and the residual stresses in a component or structure are caused by incompatible internal permanent strains. Welding, which is one of the most significant causes of residual stress, typically produces large tensile stresses, the maximum value of which is approximately equal to the yield strength of materials that are joined by lower compressive residual stresses in a component. The residual stress of welding can significantly impair the performance and reliability of welded structures. The integrity of welded joints must be ensured against fatigue or corrosion during their long use in welded components or structures. In particular, stress-corrosion cracking usually occurs when the following three factors exist at the same time: susceptible material, corrosive environment, and tensile stress (including residual stress). Thus, residual stress becomes very critical for stress-corrosion cracking when it is difficult to improve the material corrosivity of the components and their environment under operating conditions [1]. Since the welding residual stress is a major factor in the generation of primary water stress corrosion cracking (PWSCC), the prediction of the welding residual stress is important for preventing PWSCC. Residual stresses may be measured by nondestructive techniques and locally destructive techniques. The nondestructive techniques include X-ray and neutron diffraction methods, magnetic methods, and ultrasonic techniques. The locally destructive techniques include hole drilling methods, ring core techniques, and sectioning methods. The selection of the optimum measurement technique should be based on consideration of the volumetric resolution, the material, the geometry, and the type of access. In recent years, there has been a rapid increase in efforts to predict residual stresses by numerical modeling of welding processes. Modeling of welding is technically A fuzzy neural network model is presented to predict residual stress for dissimilar metal welding under various welding conditions. The fuzzy neural network model, which consists of a fuzzy inference system and a neuronal training system, is optimized by a hybrid learning method that combines a genetic algorithm to optimize the membership function parameters and a least squares method to solve the consequent parameters. The data of finite element analysis are divided into four data groups, which are split according to two end-section constraints and two prediction paths. Four fuzzy neural network models were therefore applied to the numerical data obtained from the finite element analysis for the two end-section constraints and the two prediction paths. The fuzzy neural network models were trained with the aid of a data set prepared for training (training data), optimized by means of an optimization data set and verified by means of a test data set that was different (independent) from the training data and the optimization data. The accuracy of fuzzy neural network models is known to be sufficiently accurate for use in an integrity evaluation by predicting the residual stress of dissimilar metal welding zones.