Creep Life Prediction of 508-III Steel Based on Nonlinear Ultrasonic Guided Wave Using ANN

To ensure the safety of reactor pressure vessel (RPV), it is critical to detect and evaluate creep damage of the commonly used 508-III steel. Nonlinear ultrasonic guided wave is an effective tool to detect the creep damage as it is highly sensitive to microstructure evolution. In this study, an artificial neural network (ANN)-based creep life prediction model was developed based on the received nonlinear ultrasonic guided wave in 508-III steel. Creep damage samples were tested at stress of 90 MPa and 80 MPa, respectively. The nonlinear parameters were found to initially increase with respect to creep damage, followed by a decrease. There was a significant increase in the nonlinear parameters at 80% creep life. The developed ANN-based creep life prediction model was then used to establish a mapping relationship between nonlinear ultrasonic characteristics and creep strain. The results show that the model can effectively predict creep strain under different operating conditions. This study proposes a potential method for the assessment of creep damage and the prediction of the creep life.