摘要
在计算初因事件频率的过程中,当初因事件获取的数据较少时,便要考虑不确定性。在概率安全评估(PSA)中,通常以概率分布的形式表示事件的不确定性。本文针对"贮罐内压过大"这个初因事件数据量获取少的问题,采用贝叶斯-马尔科夫链蒙特卡洛方法(Bayesian-MCMC方法)对其频率进行了不确定性分析,得到了初因事件频率的不确定分布图形,并分析了不确定分布图形的特点,同时与直接计算频率方法进行了结果比较,从而验证了该方法的正确性和有效性。
In calculation of the frequency of initiating events,the uncertainty of initiating events with small amounts of data should be considered.In probabilistic safety assessment（PSA）,the uncertainty of events is usually expressed in the form of probability distributions.This paper discusses an initiating event which is too large pressure within the tank with a small amount of data available.This paper applies the BayesianMarkov chain-Monte Carlo method（Bayesian-MCMC）to analyzing the uncertainty of the frequency,and obtains the uncertainty distributions of the frequency of the initiating event.The paper also analyzes the characteristics of the distributions.By comparing the outcome with that from the direct calculation frequency method,the study also validates the correctness and validity of this method.
出处
《安全与环境工程》
CAS
北大核心
2014年第6期155-159,共5页
Safety and Environmental Engineering
基金
总装预研基金项目(9140A19040311KG0205)
关键词
概率安全评估
Bayesian-MCMC方法
初因事件频率
少量数据
不确定分布
事件链
贮罐内压过大
probabilistic safety assessment
Bayesian-MCMC method
frequency of initiating event
small amounts of data
uncertainty distribution
event chain
too large pressure within the tank
