arrow
返回

Sequential Bayesian inference for Weibull distribution parameters with initial hyperparameter optimization for system reliability estimation

delete2022-08-01
delete18
delete
OA
AI
S
Susumu Shuto *
T
Takashi Amemiya
DOI:10.1016/j.ress.2022.108516delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
In this study, we investigated the sequential Bayesian inference of the Weibull distribution parameters of the components of a system the remaining useful life of which can be estimated using lifetime estimation or survival function of certain types of components. Failure observations are collected sequentially and analyzed as censored data. Here, we propose a sequential Bayesian inference with optimized prior distribution (SBOPD) to provide an effective way to estimate the parameters of the Weibull distribution of the components. In this framework, the Bayesian inference of the parameters of the Weibull distribution is updated by a certain number of new failure observations. Two measures are used to estimate more accurately and effectively in the preliminary stages of the system life cycle; the posterior distribution of the previous update is used as the prior information of the new update, and the initial prior distribution is optimized using prior knowledge of the parameters. With optimized hyperparameters for the initial prior distribution, parameter estimation with SBOPD provided a better estimate at the preliminary stages of the system life than conventional methods.
Keyword:
System reliability
Component reliability
Weibull distribution
Bayesian inference
Weibull parameter estimation
Prior distribution
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

R
Reliability Engineering and System Safety
IF:
11
论文数:
9.0K
被引数:
4.2W

机构

Y
Yokohama National University
学者数:
4.2K
论文数: 3.4K
被引数: 3.2K
引用论文

引用论文

Recombinant interleukin-2 and lymphokine-activated killer cells in renal cancer patients: I. phenotypic and functional analysis of the peripheral blood mononuclear cells
err1990-05-01
err0
PREAI
errClaudio Fortis; Elisabetta Ferrero; Carlo Besana; Mauro Biffi; Silvia Heltai; Laura Galli; Anna Borri; Alisa Schoenheit; Claudio Rugarli
err分享
err收藏
学者 查看更多内容