| 30 | 0 | 70 |
| 下载次数 | 被引频次 | 阅读次数 |
为加强瓶装液化石油气(LPG)事故防治,首先,基于2019—2024年间我国90起瓶装LPG事故典型案例,采用数理统计法分析其时间、空间、等级与类型的特征和规律;其次,构建事故STAMP模型,从政府监管、企业管理、人员行为、设备与环境4个层级的11项子系统处辨识18项事故致因;再次,采用卡方检验、让步比、Cramer's V系数分析相邻层级事故致因间的相关性,运用灰色关联分析(GRA)探讨各层级致因与事故间的关联程度。结果表明:瓶装LPG事故“频率高、地域广、后果重”特征显著;夏季和每日6:00—8:59为事故高发时间;华东为事故频发地区,餐饮场所和居民用户为事故的频发地点;“餐饮商户违规用气型”在4种事故类型中最常见,占比52.22%;相邻层级事故致因间存在14组显著因果关系,形成由经营者和用气者主导的2条关键路径,又以后者为主,结论认为:应统筹地方政府、瓶装LPG经营企业和用户企业等主体构建“协同—合作”双态防治模式,聚焦STAMP模型、2条关键路径和3项核心事故诱因采取针对性的安全控制策略。
Abstract:To strengthen the prevention and control of bottled Liquefied Petroleum Gas(LPG) accidents, firstly, the mathematical statistical methods were applied to analyze the characteristics and laws of 90 typical accident cases between 2019 to 2024 in China, from the perspective of time, space, level and type. Secondly, the STAMP model for bottled LPG accidents was constructed, then 18 typical accident causal factors were identified from 11 subsystems at 4 levels: government regulation, company management, human behaviors, equipment and environment. Thirdly, the correlation between accident causal factors at adjacent levels was analyzed by chi-square test, odds ratio, and Cramer's V coefficient, and the grey relation analysis method was used to explore the correlation between different accident causal factors at each level and accidents. The results indicate that the bottled LPG accidents have significant characteristics of high frequency, wide geographical coverage, and severe consequences. Summer and 6:00-8:59 are the peak periods for accidents. East China is the region with the highest frequency of accidents, while catering enterprises and residential households are the most frequent locations for accidents. The most common type of bottled LPG accidents among the four types is the violation of gas usage by catering merchants, which accounting for 52.22%. There are a total of 14 significant causal relationships between accident causal factors at adjacent levels, forming two key paths dominated by bottled LPG operators and users, with the latter being the main one. It is concluded that a collaborative-cooperation dual-mode accident prevention model should be established by coordinating local governments, bottled LPG operators, and user enterprises, then the targeted safety control measures should be implemented based on the STAMP model, 2 key accident paths and 3 core accident causal factors.
[1]Yuan X J, Li X, Zhang Y X, et al. The influence of ventilation conditions on LPG leak dispersion in a commercial kitchen[J].Energies, 2025, 18(11):2678.
[2]Tian B, Zhao Z F, Cui X J, et al. Characteristics of malignant urban gas accidents in China from 2013 to 2022[J]. Heliyon,2024, 10(14):e34568.
[3]中国城市燃气协会安全管理工作委员会.全国燃气事故分析报告(2024年·上半年报告)[R].北京:中国城市燃气协会,2024.
[4]田彬,张亦雯,崔晓君,等. 2016—2020年我国城镇燃气爆炸事故统计与规律分析[J].安全与环境学报,2023, 23(6):1993-2002.
[5]陈澳占,杜清,李文炜,等.2019—2022年餐饮燃气事故统计与分析[J].煤气与热力,2024, 44(3):36-42.
[6]高思达,郝琳,朱振兴,等.液化石油气(LPG)泄漏事故定量风险分析[J].化学工业与工程,2023, 40(4):78-85.
[7]Barbosa T F, Reis L, Raposo J, et al. LPG stored at the wildland-urban interface:recent events and the effects of jet fires and BLEVE[J]. International Journal of Wildland Fire, 2023, 32(3):388-402.
[8]Lyu S, Zhang S H, Huang X M, et al. Investigation and modeling of the LPG tank truck accident in Wenling, China[J].Process Safety and Environmental Protection, 2022, 157:493-508.
[9]Kang Y, Xia X X, Wu Z, et al. Visualization of integrated failure consequences of hazardous chemical leakage and explosion[J]. Journal of Loss Prevention in the Process Industries, 2024, 92:105464.
[10]郑小强,孙逸林.复杂社会技术系统视角下民用瓶装LPG爆炸事故致因分析[J].安全与环境学报,2024, 24(12):4808-4819.
[11]北京大学能源研究院,中国城市燃气协会液化石油气专业委员会.液化石油气行业发展报告[R].北京:中国城市燃气协会,2024.
[12]孙逸林,郑小强,刘丹秀,等.我国城镇燃气管网生产安全事故统计与宏观规律分析[J].武汉理工大学学报(信息与管理工程版),2023, 45(5):675-680,703.
[13]孙逸林,郑小强,刘险峰,等.重大城市燃气管道泄漏爆炸事故定量分析方法研究:以湖北十堰“6·13”事故为例[J].安全与环境学报,2022, 22(6):3312-3320.
[14]Hu S P, Fang C W, Wu J J, et al. Enhanced risk assessment framework for complex maritime traffic systems via data driven:a case study of ship navigation in Arctic[J]. Reliability Engineering and System Safety, 2025, 260:110991.
[15]Hulme A, Stanton N A, Walker G H, et al. Testing the reliability of accident analysis methods:a comparison of AcciMap,STAMP-CAST and AcciNet[J]. Ergonomics,2024,67(5):695-715.
[16]李显,焦宇,陈文涛,等.船舶修造企业火灾事故特征挖掘与致因分析[J].中国安全科学学报,2024,34(3):29-38.
[17]刘平,王诗禹,袁伟杰.基于改进HFACS模型的深基坑坍塌事故人因分析[J].项目管理技术,2025,23(4):67-75.
[18]秦宾宾,张清华,孙国玺,等.基于GRA-DBO-SVR的瓦斯含量预测方法[J].广东石油化工学院学报,2024,34(4):80-86.
[19]程硕,阳富强. 2011—2020年我国危险化学品事故统计及灰色关联分析[J].应用化工,2023,52(1):193-198.
[20]孙逸林,郑小强.城市燃气管网生产安全事故风险协同治理模式研究[J].灾害学,2024, 39(3):167-172.
基本信息:
DOI:10.26962/j.cnki.1991.2026.0063
中图分类号:TE88
引用信息:
[1]孙逸林,李洪兵.基于STAMP模型的瓶装液化石油气事故致因[J].广东石油化工学院学报,2026,36(04):16-23.DOI:10.26962/j.cnki.1991.2026.0063.
基金信息:
四川省哲学社会科学研究项目(SC25T056); 四川石油天然气发展研究中心课题(2025SY011); 四川省社会科学重点研究基地沱江流域高质量发展研究中心课题(TJGZL2025-12); 四川省哲学社会科学重点实验室“智慧应急管理重点实验室”课题(2025ZHYJGL-18)
2025-07-13
2025
2025-10-17
2026-03-27
2026
1
2026-08-25
2026-08-25
2026-08-25