Improved Bayesian update method on flaw distributions reflecting non-destructive inspection result

Katsuyama, Jinya; Miyamoto, Yuhei*; Lu, K.; Mano, Akihiro; Li, Y.

Proceedings of ASME 2020 Pressure Vessels and Piping Conference (PVP 2020) (Internet), 8 Pages, 2020/08

https://doi.org/10.1115/PVP2020-21430

We have developed a probabilistic fracture mechanics (PFM) analysis code PASCAL4 for evaluating failure frequency of reactor pressure vessels (RPVs). It is known that flaw distributions have an important role in failure frequency calculation in PFM analysis. Previously, we proposed likelihood function to obtain more realistic flaw distributions applicable for both case when flaws are detected and when there is no flaw indication as the inspection results based on Bayesian update methodology. Here, it can be applied to independently obtain posterior distributions of flaw depth and density. In this study, we improve the likelihood function to enable them to update flaw depth and density simultaneously. Based on the improved likelihood function, an example is presented in which flaw distributions are estimated by reflecting NDI results through Bayesian update and PFM analysis. The results indicate that the improved likelihood functions are useful for estimating flaw distributions.