단시간 크리프 파단 데이터의 통계적 변동성과 확률론적 크리프 수명예측 방법에 관한 연구

Abstract

An accurate assessment of creep rupture behavior is important to the structural design of creep sensitive components at elevated temperature. The creep design approach is based on life assessments and design codes involving the analysis of creep rupture data. By their very nature, experimental creep data take a long time to accumulate and there is usually a requirement to extrapolate long durations beyond the range of the data. The accuracy of such extrapolations and design strengths, and the determined confidence limits have both essential safety and economic implication. Currently, it is well known that a considerable amount of variability is obtained in experimental results relating to creep rupture properties such as the creep rupture time, initial strain on loading, the minimum creep rate and the total creep strain to failure under constant creep stress and temperature. Various aspects of the variability in creep rupture data, the statistical and probabilistic modeling of creep rupture data have been investigated by many researchers. In this work, a series of statistical studies were performed on short-term creep rupture data in STS304 stainless steels. In order to clearly investigate the variability of short term creep rupture data in inter-specimens and to develop the methodology for probabilistic creep life prediction, the variability of short term creep rupture time was investigated by Weibull statistical analysis. Finally, creep life prediction methods considering statistical variability of short term creep property were presented. The main results of this study can be summarized as follows: In the investigation of the statistical characteristics of creep rupture time using few experimental data for a relatively short period of time, this study represented a possibility in the prediction of creep rupture time at high temperatures for a long period of time as a statistical and probabilistic way.