EVALUATION OF THE PROBABILITY OF FAILURE WITH UNCERTAIN DISTRIBUTION PARAMETERS
作者:
HANPING HONG,
期刊:
Civil Engineering Systems
(Taylor Available online 1996)
卷期:
Volume 13,
issue 2
页码: 157-168
ISSN:0263-0257
年代: 1996
DOI:10.1080/02630259608970193
出版商: Taylor & Francis Group
关键词: Reliability analysis;point estimate method;probabilities;probability of failure;time invariant;time variant;uncertain distribution parameters
数据来源: Taylor
摘要:
Formulations are presented for estimating probability of failure considering uncertainty of distribution parameters in time invariant and time variant reliability analyses. Based on the formulations the probability of failure can be calculated by recursively using the first order reliability method. Also, a more efficient approximate analysis procedure by using the point estimate method to estimate the probability of failure is given. In this analysis procedure, the point estimate method is used to discretize the uncertain distribution parameters in the time invariant reliability analysis, and to discretize the time-independent random variables and uncertain distribution parameters in the time variant reliability analysis. The probability of failure is then obtained by weighting the probability of failure conditioned at each of discrete points. The conditional probability of failure can be calculated by using first order reliability method, second order reliability method or any other convenient methods. The use of point estimate method to treat uncertain distribution parameters in calculating probability of failure is less computer time consuming than the one of recursively using FORM. Illustrative numerical examples of calculating probability of failure are presented.
点击下载:
PDF (409KB)
返 回