Validity of Successive Inspections to the Piping With Multiple Circumferential Stress Corrosion Cracks

Abstract

This paper describes the validity of successive inspections in a chain of the integrity assessment applying Fitness-For-Service (FFS) Codes to piping with stress corrosion cracks. Integrity assessment of primary recirculation system (PLR) piping with stress corrosion cracks is one of the recent interests in Japanese boiling water reactor (BWR) plants. Girth weld joints with detected stress corrosion cracks are required to secure the integrity during subsequent 5 years operation and to conduct three times of successive inspections for PLR piping. If the safety factor included in FFS Codes works effectively, this, like frequent successive inspections are deemed unnecessary. However, since there is some uncertainty in the measurement of crack size and in crack growth estimation in the integrity assessment of piping with stress corrosion cracks, a possibility of failure cannot be denied when not conducting successive inspections to confirm the validity of the integrity assessment. In this study, piping failure probability when conducting and not conducting successive inspections using a probabilistic fracture mechanics (PFM) code, and the validity of successive inspections were examined quantitatively from the failure probabilities. The effect of crack sizing accuracy, which has great influence on the validity of inspections, was also examined. The validity of successive inspections was examined for 12, 16, and 24 inch pipes which were employed in PLR piping of BWR-5. It became clear that successive inspections were very effective in decreasing failure in 12 and 24 inch pipes, but not as effective in decreasing failure in 16 inch pipes. The reason is suggested that the crack growth rate in a 16 inch pipe is high compared with the other two sized pipes and opportunity of renewal of crack growth estimation at successive inspections is few. When successive inspections are carried out, there are several times of opportunity for improvement of underestimated crack growth behavior unless a crack size is measured smaller than the actual size in all successive inspections. Consequently, the influence of the crack sizing error on failure probability of pipes becomes very small.