| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 769368 | Engineering Failure Analysis | 2006 | 13 Pages |
Prediction of the residual life of high temperature components is becoming a major issue. Key decisions, such as whether to repair or replace cracked components, have to be based on precise information. The cooperative program, HIDA Applicability, is aimed at improving these issues. This paper deals with the numerical application of probabilistic assessment to crack growth analysis. Based on the present studies, it has been shown that it is both possible and practical to use stratified samples to obtain increased accuracy or to reduce computation time. Stratified samples are particularly suitable when the desired probability of failure is very low.The ALIAS-HIDA software has been designed to calculate the probability of failure of cracked components under high temperature conditions. The Monte Carlo simulation technique was implemented successfully and the first tests seem to show it to be reliable. The ability of the software to deal with such calculations, which require a large number of random variables and simulations, was illustrated by a Creep Crack Growth calculation example. The HIDA test programme has provided numerous results, which have been supplemented by data from published and unpublished sources, and gathered in the HIDA database. Some information from the HIDA Database has been re-organized and analyzed in order to conduct probabilistic analysis. This task is still in progress, however, some results are available, and trends and technical issues are described for a 316 austenitic stainless steel.
