基于贝叶斯网络的石油罐区多米诺效应风险分析技术研究
发布时间：2018-08-30 18:02
【摘要】：石油和化工行业是我国的支柱产业之一,对国民经济起着重要的作用。随着我国石油和化工行业的发展,大型化工园区和储罐区不断出现。由于石油和化工行业的高危险性,一旦发生火灾、爆炸事故,将产生非常严重的后果。多米诺效应是指当初始事故发生时,引起临近的其他单元发生事故,导致事故的后果大于初始事件。一旦化工园区或储罐区发生多米诺效应,往往能造成灾难性的后果,所以对化工园区和储罐区的多米诺效应进行风险分析十分重要。本文主要研究内容如下：1、2011-2014我国危险化学品事故统计分析。2011-2014年我国共发生821起危险化学品事故,共造成死亡993人,受伤2960人。本文对821起事故,从事故发生的时间、地点、物理表象、环节和原因等几个方面进行了统计分析。此外,本文对2011-2014年我国发生的多米诺事故进行了列举分析,结果表明多米诺事故后果更加严重,并存在于我国危险化学品生产、储存、运输等环节。2、基于贝叶斯网络的多米诺效应场景定量分析。运用贝叶斯网络对多米诺效应进行分析,具有可以考虑多级多米诺效应、多米诺效应的协同效应、多事故情景等优势,可以使得多米诺效应的分析更加真实准确。此外,贝叶斯网络的证据推理还可以准确的识别多米诺效应事故情景。本文通过实例,对贝叶斯网络在多米诺效应概率以及多米诺效应场景上的应用进行了分析,并通过改变阈值,研究了阈值对多米诺效应概率的影响。3、基于贝叶斯网络的多米诺效应个人风险评估。由于贝叶斯网络的优势,在多米诺效应风险评估时运用贝叶斯网络可以使评估结果更加真实。本文建立了基于贝叶斯网络的多米诺效应个人风险评估方法。实例表明,运用贝叶斯网络对多米诺效应进行风险评估,相同风险等值曲线覆盖范围更大,评估结果更准确。当阈值改变时,风险评估的结果也会改变,所以阈值的选取在多米诺效应的风险评估中十分重要。
[Abstract]:Petroleum and chemical industry is one of the pillar industries in China, which plays an important role in the national economy. With the development of petroleum and chemical industry in China, large-scale chemical parks and storage tanks are emerging. Because of the high risk of petroleum and chemical industry, once fire and explosion occur, it will have very serious consequences. Domino effect refers to that when the initial accident occurs, it causes the accident to happen in other adjacent units, and the consequence of the accident is greater than that of the initial event. Once the domino effect occurs in the chemical park or the storage area, it will often cause disastrous consequences, so it is very important to analyze the risk of the domino effect in the chemical park and the storage area. The main contents of this paper are as follows: 1: 1. Statistical analysis of dangerous chemical accidents in China in 2011-2014. There were 821 hazardous chemical accidents in China in 2011-2014, resulting in 993 deaths and 2960 injuries. This paper makes a statistical analysis of 821 accidents in terms of the time, place, physical appearance, links and causes of the accidents. In addition, this paper enumerates and analyzes the Domino accident in China from 2011-2014. The results show that the Domino accident has more serious consequences and exists in the production and storage of hazardous chemicals in China. Transportation and other links. 2, based on Bayesian network domino effect scene quantitative analysis. The application of Bayesian network to the analysis of domino effect has the advantages of considering the multi-stage domino effect, the synergic effect of domino effect and the multi-accident scenario, so that the analysis of domino effect can be more true and accurate. In addition, Bayesian network evidence reasoning can also accurately identify domino effect accident scenarios. In this paper, the application of Bayesian network in domino effect probability and domino effect scenario is analyzed, and the threshold is changed. The influence of threshold on the probability of domino effect. 3. The personal risk assessment of domino effect based on Bayesian network. Because of the advantage of Bayesian network, using Bayesian network in Domino effect risk assessment can make the evaluation results more true. In this paper, a Bayesian network based Domino effect personal risk assessment method is established. The example shows that using Bayesian network to evaluate the domino effect, the coverage of the same risk equivalent curve is larger and the evaluation result is more accurate. When the threshold value changes, the result of risk assessment will also change, so the selection of threshold value is very important in the risk assessment of domino effect.
【学位授予单位】：北京化工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F426.22
，
本文编号：2213862
[Abstract]:Petroleum and chemical industry is one of the pillar industries in China, which plays an important role in the national economy. With the development of petroleum and chemical industry in China, large-scale chemical parks and storage tanks are emerging. Because of the high risk of petroleum and chemical industry, once fire and explosion occur, it will have very serious consequences. Domino effect refers to that when the initial accident occurs, it causes the accident to happen in other adjacent units, and the consequence of the accident is greater than that of the initial event. Once the domino effect occurs in the chemical park or the storage area, it will often cause disastrous consequences, so it is very important to analyze the risk of the domino effect in the chemical park and the storage area. The main contents of this paper are as follows: 1: 1. Statistical analysis of dangerous chemical accidents in China in 2011-2014. There were 821 hazardous chemical accidents in China in 2011-2014, resulting in 993 deaths and 2960 injuries. This paper makes a statistical analysis of 821 accidents in terms of the time, place, physical appearance, links and causes of the accidents. In addition, this paper enumerates and analyzes the Domino accident in China from 2011-2014. The results show that the Domino accident has more serious consequences and exists in the production and storage of hazardous chemicals in China. Transportation and other links. 2, based on Bayesian network domino effect scene quantitative analysis. The application of Bayesian network to the analysis of domino effect has the advantages of considering the multi-stage domino effect, the synergic effect of domino effect and the multi-accident scenario, so that the analysis of domino effect can be more true and accurate. In addition, Bayesian network evidence reasoning can also accurately identify domino effect accident scenarios. In this paper, the application of Bayesian network in domino effect probability and domino effect scenario is analyzed, and the threshold is changed. The influence of threshold on the probability of domino effect. 3. The personal risk assessment of domino effect based on Bayesian network. Because of the advantage of Bayesian network, using Bayesian network in Domino effect risk assessment can make the evaluation results more true. In this paper, a Bayesian network based Domino effect personal risk assessment method is established. The example shows that using Bayesian network to evaluate the domino effect, the coverage of the same risk equivalent curve is larger and the evaluation result is more accurate. When the threshold value changes, the result of risk assessment will also change, so the selection of threshold value is very important in the risk assessment of domino effect.
【学位授予单位】：北京化工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F426.22
，
本文编号：2213862
本文链接：https://www.wllwen.com/qiyeguanlilunwen/2213862.html
