中国安全科学学报 ›› 2019, Vol. 29 ›› Issue (4): 177-182.doi: 10.16265/j.cnki.issn1003-3033.2019.04.028

• 公共安全 • 上一篇    下一篇

化工园区储罐多米诺效应的脆弱性评估

蒋代1, 华敏1,2 副教授, 潘旭海1,2 教授, 蒋军成1 教授   

  1. 1 南京工业大学 安全科学与工程学院,江苏 南京 210009;
    2 江苏省危险化学品本质安全与控制技术重点实验室,江苏 南京 210009
  • 收稿日期:2018-11-16 修回日期:2019-01-24 出版日期:2019-04-28 发布日期:2020-11-04
  • 作者简介:蒋 代 (1993—),女,四川广安人,硕士研究生,研究方向为储罐区多米诺效应风险。E-mail: dai_jiang@foxmail.com。
  • 基金资助:
    国家重点研发计划项目(2016YFC0801500,2016YFC0800100)。

Vulnerability assessment of tanks in chemical industry park under domino effect

JIANG Dai1, HUA Miu1,2, PAN Xuhai1,2, JIANG Juncheng1   

  1. 1 College of Safety Science and Engineering, Nanjing Tech University, Nanjing Jiangsu 210009, China;
    2 Key Laboratory of Intrinsic Safety and Control Technology for Dangerous Chemicals in Jiangsu Province, Nanjing Jiangsu 210009, China
  • Received:2018-11-16 Revised:2019-01-24 Online:2019-04-28 Published:2020-11-04

摘要: 为减少化工园储罐区多米诺效应发生概率,找出最容易发生多米诺效应、且对事故影响最大的储罐,评估各储罐的脆弱性等级。首先基于不同领域脆弱性的内涵和多米诺效应的特点,提出多米诺效应情况下储罐脆弱性的定义;然后利用熵权物元可拓模型和风险矩阵法的客观性,并结合复杂网络模型和逼近理想排序法(TOPSIS)模型评估事故网络节点重要度,从初始事故的可能性与传播影响力2个方面,构建多米诺效应下储罐脆弱性分级评估模型;最后应用该模型评估某化工园储罐区各罐体脆弱性等级,并将结果与贝叶斯方法的评估结果对比。结果表明: 该模型考虑了储罐自身属性及影响扩展范围,得到的脆弱性分级结果更准确。

关键词: 多米诺效应, 脆弱性, 熵权物元可拓模型, 复杂网络模型, 评估模型

Abstract: In order to reduce the probability of domino effect in the chemical tank area, the vulnerability of each storage tank is evaluated to protect the tank that has a maximum accident probability of domino effect and the greatest impact on accidents. First, the definition of vulnerability of storage tanks under domino effect was proposed based on the connotation of vulnerability in different fields and the characteristics of domino effect. Second, by using the objectivity of entropy weight matter element extension model and risk matrix method in combination with the importance of complex network model and technique for order preference by similarity to ideal solution (TOPSIS) in evaluating the accident network node, a model for vulnerability grade evaluation of tanks under domino effect was constructed from the possibility of initial accident and the influence of propagation. Finally, the model was used to evaluate the vulnerability level of each tank in a chemical industry park, and the results were compared with that of the Bayesian method. The results show that this model takes the attributes of tanks and the scope of impact into consideration and proves to be more accurate with its vulnerability grading results.

Key words: domino effect, vulnerability, entropy weight matter element extension model, complex network model, evaluation model

中图分类号: