Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : September 15, 2021 - September 17, 2021
The Rules on Fitness-for-Service (the FFS Codes) of the Japan Society of Mechanical Engineers stipulates limitation of the crack angle in the fracture evaluation of circumferential cracks in pipes. This limitation of the crack angle was set because of the impact of the crack size measurement error on the fracture strength mainly in the small-diameter pipe where the thickness becomes thin. The measured crack sizes vary in the magnitude direction with respect to the true sizes. Assuming that the crack size is measured small when the pipe diameter is small or the crack angle is large, the possibility of fracture cannot be ruled out even if the safety factor for loading is taken into consideration. However, according to the FFS Codes, when the pipes with cracks are used continuously, multiple successive inspections are required during the period of use, and the crack sizes are reviewed each time. In addition, usually, multiple crack size measurements are performed, and the maximum size among them is recorded as the crack size. In this way, considering that the maximum size of the measured crack sizes during the in-service and successive inspections is used for evaluation, the average of the measured sizes is larger than the true one, the standard deviation is smaller. From this point of view, it is advantageous to increase the number of inspections. On the other hand, since unnecessary inspections lead to an increase of exposure of personnel, it is important to set a target for the fracture probability according to the risk importance of the pipes and to grasp the number of inspections that satisfy the target. In this study, the reliability of fracture evaluation considering the measured average and standard deviation of crack sizes according to the number of measurements is evaluated by Monte Carlo simulation, and the appropriate number of measurements to satisfy the target is proposed according to the pipe thickness and measurement error.