化工进展 ›› 2016, Vol. 35 ›› Issue (S2): 110-115.DOI: 10.16085/j.issn.1000-6613.2016.s2.018

• 化工过程与装备 • 上一篇    下一篇

基于贝叶斯网络的粉尘爆炸多米诺效应分析

夏晨曦, 韩辉, 李伟敏   

  1. 江苏省安全生产科学研究院, 江苏 南京 210042
  • 出版日期:2016-12-31 发布日期:2016-12-22
  • 通讯作者: 夏晨曦(1987-),男,工程师,研究方向为过程安全。E-mail:xcxi09@163.com。

Domino effect analysis of dust explosions using Bayesian networks

XIA Chenxi, HAN Hui, LI Weimin   

  1. Jiangsu Academy of Safety Science & Technology, Nanjing 210042, Jiangsu, China
  • Online:2016-12-31 Published:2016-12-22

摘要: 粉尘爆炸机理的复杂性及引起粉尘爆炸因素的多样性使得粉尘爆炸成为学者们研究的热点课题,目前研究成果集中在不同种类粉尘的爆炸特性、粉尘爆炸机理、预防和减轻粉尘爆炸的安全措施及粉尘爆炸定量风险分析等方面,但在建立粉尘爆炸相关模型及风险评价方面的研究较少。本文在剖析粉尘爆炸机理的基础上结合多米诺效应原理建立了粉尘爆炸多米诺效应计算模型,该模型在计算出初始粉尘爆炸事故的条件概率后,利用贝叶斯网络灵活的特性和自动推理引擎,可以计算出潜在多米诺效应传播的途径。通过实例应用该模型的计算结果,可以得出发生粉尘爆炸多米诺效应的概率变化情况及粉尘爆炸事故最可能发生的传播途径,为粉尘爆炸事故的预防控制及风险评价提供了方向。

关键词: 粉尘爆炸, 事件树, 升级概率, 多米诺效应, 贝叶斯网络

Abstract: Because of the complexity of the dust explosion mechanism and the diversity of the causes of dust explosion,dust explosion is becoming a hot spot of research,the outcomes are mainly shown in different kinds of dust explosion,dust explosion mechanism,applications of preventive and mitigating safety measures and dust explosion quantitative risk analysis,etc.,Nevertheless,the attempts made to model and assess the risk of domino effect analysis of dust explosions have been very few.This paper present the calculation model of dust explosion based on the dust explosion mechanism.After calculating the accidents conditional probability of initial dust explosion,taking advantage of the flexible structure and robust reasoning engine of BN,the likely propagation route of potential domino effects along with the probabilities thereof can be calculated.Through the application of the model calculation results,which can be concluded that dust explosion probability of domino effect and the changes of dust explosion accidents most likely route of transmission.The model provides a direction for the prevention of dust explosion accident and risk assessment.

Key words: dust explosions, event tree, escalation probability, domino effect, Bayesian network

中图分类号: 

京ICP备12046843号-2;京公网安备 11010102001994号
版权所有 © 《化工进展》编辑部
地址:北京市东城区青年湖南街13号 邮编:100011
电子信箱:hgjz@cip.com.cn
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn