基于模糊贝叶斯网络的洞室事故人因分析
|
摘要
为量化水利工程施工安全风险,提高施工系统网络安全性,针对水利工程洞室坍塌事故,从人因角度出发,基于模糊贝叶斯网络理论对其进行风险量化。首先,利用修订HFACS框架辨识主要人为因素,然后利用专家知识构建事故树并将其转化为模糊贝叶斯网络,通过模糊集理论将基本事件的先验概率量化,并确定条件概率,最后利用GeNIe软件对模糊贝叶斯网络进行推理分析,最后得到事故的发生概率及各因素的敏感性。研究结果显示,利用模糊贝叶斯网络可预测事故发生概率也可用于探究事故发生原因。其中安全教育及风险监控不到位、隐患未整改的敏感性因子较大是主要风险事件。此结论与事故报告结论基本一致,说明模糊贝叶斯网络对水利安全风险预测是可行的。
In order to quantify the safety risks of hydraulic engineering construction and improve the network security of construction system, this paper focuses on the collapse accident of hydraulic engineering caverns from the perspective of human factors based on fuzzy Bayesian network theory. Firstly, the main human factors was identified by using the revised HFACS framework, and then the expert knowledge was used to construct the fault tree and transform it into a fuzzy Bayesian network. Next, the prior probability of basic events was quantified through fuzzy set theory and the conditional probability was determined. Finally, genie software was used to analyze the fuzzy Bayesian network and obtain the occurrence probability of the accident, as well as the sensitivity of each factor. The research results showed that the fuzzy Bayesian network could be used to predict the probability of accidents and to explore the cause of the accident. Among them, safety education and risk monitoring were not functionally in place, and the sensitivity factors of hidden dangers not rectified was large, which. were the main risk events. This conclusion is basically consistent with the conclusion of the accident report, indicating that the fuzzy Bayesian network is feasible for water security risk prediction.
In order to quantify the safety risks of hydraulic engineering construction and improve the network security of construction system, this paper focuses on the collapse accident of hydraulic engineering caverns from the perspective of human factors based on fuzzy Bayesian network theory. Firstly, the main human factors was identified by using the revised HFACS framework, and then the expert knowledge was used to construct the fault tree and transform it into a fuzzy Bayesian network. Next, the prior probability of basic events was quantified through fuzzy set theory and the conditional probability was determined. Finally, genie software was used to analyze the fuzzy Bayesian network and obtain the occurrence probability of the accident, as well as the sensitivity of each factor. The research results showed that the fuzzy Bayesian network could be used to predict the probability of accidents and to explore the cause of the accident. Among them, safety education and risk monitoring were not functionally in place, and the sensitivity factors of hidden dangers not rectified was large, which. were the main risk events. This conclusion is basically consistent with the conclusion of the accident report, indicating that the fuzzy Bayesian network is feasible for water security risk prediction.
引文
[1] 叶贵,陈梦莉,汪红霞.建筑安全事故人为因素分类研究[J].中国安全生产科学技术,2016,12(4):131-137.
[2]郑霞忠,李华飞,魏雄伟,等.贝叶斯网络在岩质边坡开挖风险概率评估中的应用[J].水电能源科学,2013,31(7):130-132.
[3]周忠宝,董豆豆,周经伦.贝叶斯网络在可靠性分析中的应用[J].系统工程理论与实践,2006(6):95-100.
[4]陈兆波,阴东玲,曾建潮,等.基于贝叶斯网络的煤矿事故人因推理[J].中国安全生产科学技术,2014,10(11):145-150.
[5] 王俊人. 地铁工程施工期间灾害风险分析[D].成都:西南交通大学,2016.
[6]毛宇辰,刘琨,石春,等.基于贝叶斯理论的地下厂房开挖进度动态预测[J].人民珠江,2018(9):62-66.
[7]LUXHOJJ T. Probabilistic causal analysis for system safety risk assessments in commercial air transport [C]// proceedings the workshop on investigating and reporting of incidents and accidents Williams-Proceedings the workshop on investigating and reporting of incidents and accidents. Williamsburg,VA,USA,2003: 17- 38.
[8] SALMON P M, CORNELISSEN M,TROTTER M J.Systems-based accident analysis methods:A comparison of Accimap,HFACS and STAMP[J].Safety Science,2012,50 (4) : 1158-1170.
[9] 孙开畅,周剑岚.水利水电工程施工安全风险管理[M].北京:中国水利水电出版社,2013.
[10]周建方,唐椿炎,许智勇.贝叶斯网络在大坝风险分析中的应用[J].水力发电学报,2010,29(1):192-196.
[11]孙开畅,徐小峰,张耀,等.水利工程施工安全人为因素重要度分析[J].人民长江,2016,47(9):80-83,114.
[12] 李娜,赵然杭,付海军.基于模糊数的事件树法在大坝风险分析中的应用研究[J].中国农村水利水电,2009(10):135-136,139.
[13]金朝光,林焰,纪卓尚.基于模糊集理论事件树分析方法在风险分析中应用[J].大连理工大学学报,2003(1):97-100.
[14] WICKENSC D. Engineering psychology and human performance [M]. 2nd ed. New York: HarperCollins Publishers Inc, 1992: 211- 257.
[15]LIOU T S, WANG M J. Ranking fuzzy numbers with integral value [J].Fuzzy Sets and Syst,1992,50(3):247-255.
[16] 任佳,高晓光.贝叶斯网络参数学习及对无人机的决策支持[M].北京:国防工业出版社, 2012.
[17] 张笛,梁峥,范存龙,等.基于贝叶斯网络的船舶自沉事故后果预测[J].中国航海,2018,41(1):53-59.
[18] 汪涛,廖彬超,马昕,等.基于贝叶斯网络的施工安全风险概率评估方法[J].土木工程学报,2010,43(S2): 384-391.
[19] 孙开畅,李权,徐小峰,等.施工高危作业人因风险分析动态贝叶斯网络的应用[J].水力发电学报,2017,36(5):28-35.
[20] YANG Y,ELFAYOUMY S.Anti-Spam filtering using neural networks andbayesian classifiers[C]/ / International Symposium on Computational Intelligence in Robotics and Automation.IEEE,2007: 272-278.
[21] 李典庆,唐文勇,张圣坤.船舶驾驶失效搁浅概率的混合事件树分析[J].上海交通大学学报, 2003, 37(8): 1146- 1150.
[2]郑霞忠,李华飞,魏雄伟,等.贝叶斯网络在岩质边坡开挖风险概率评估中的应用[J].水电能源科学,2013,31(7):130-132.
[3]周忠宝,董豆豆,周经伦.贝叶斯网络在可靠性分析中的应用[J].系统工程理论与实践,2006(6):95-100.
[4]陈兆波,阴东玲,曾建潮,等.基于贝叶斯网络的煤矿事故人因推理[J].中国安全生产科学技术,2014,10(11):145-150.
[5] 王俊人. 地铁工程施工期间灾害风险分析[D].成都:西南交通大学,2016.
[6]毛宇辰,刘琨,石春,等.基于贝叶斯理论的地下厂房开挖进度动态预测[J].人民珠江,2018(9):62-66.
[7]LUXHOJJ T. Probabilistic causal analysis for system safety risk assessments in commercial air transport [C]// proceedings the workshop on investigating and reporting of incidents and accidents Williams-Proceedings the workshop on investigating and reporting of incidents and accidents. Williamsburg,VA,USA,2003: 17- 38.
[8] SALMON P M, CORNELISSEN M,TROTTER M J.Systems-based accident analysis methods:A comparison of Accimap,HFACS and STAMP[J].Safety Science,2012,50 (4) : 1158-1170.
[9] 孙开畅,周剑岚.水利水电工程施工安全风险管理[M].北京:中国水利水电出版社,2013.
[10]周建方,唐椿炎,许智勇.贝叶斯网络在大坝风险分析中的应用[J].水力发电学报,2010,29(1):192-196.
[11]孙开畅,徐小峰,张耀,等.水利工程施工安全人为因素重要度分析[J].人民长江,2016,47(9):80-83,114.
[12] 李娜,赵然杭,付海军.基于模糊数的事件树法在大坝风险分析中的应用研究[J].中国农村水利水电,2009(10):135-136,139.
[13]金朝光,林焰,纪卓尚.基于模糊集理论事件树分析方法在风险分析中应用[J].大连理工大学学报,2003(1):97-100.
[14] WICKENSC D. Engineering psychology and human performance [M]. 2nd ed. New York: HarperCollins Publishers Inc, 1992: 211- 257.
[15]LIOU T S, WANG M J. Ranking fuzzy numbers with integral value [J].Fuzzy Sets and Syst,1992,50(3):247-255.
[16] 任佳,高晓光.贝叶斯网络参数学习及对无人机的决策支持[M].北京:国防工业出版社, 2012.
[17] 张笛,梁峥,范存龙,等.基于贝叶斯网络的船舶自沉事故后果预测[J].中国航海,2018,41(1):53-59.
[18] 汪涛,廖彬超,马昕,等.基于贝叶斯网络的施工安全风险概率评估方法[J].土木工程学报,2010,43(S2): 384-391.
[19] 孙开畅,李权,徐小峰,等.施工高危作业人因风险分析动态贝叶斯网络的应用[J].水力发电学报,2017,36(5):28-35.
[20] YANG Y,ELFAYOUMY S.Anti-Spam filtering using neural networks andbayesian classifiers[C]/ / International Symposium on Computational Intelligence in Robotics and Automation.IEEE,2007: 272-278.
[21] 李典庆,唐文勇,张圣坤.船舶驾驶失效搁浅概率的混合事件树分析[J].上海交通大学学报, 2003, 37(8): 1146- 1150.