基于主题挖掘采煤工作面事故致因模型探索
1. 北京赛迪工业和信息化工程监理中心工程公司, 北京 100089
2. 泰山职业技术学院, 机电技术工程系, 山东泰安 271001
| 摘 要： | 采煤工作面作为煤炭第一生产现场，是煤矿事故多发地。为避免或减少事故的发生，需要及时排除发现采煤工作面事故隐患，但以往事故致因多采用定性分析为主的方法，缺少有效的事故致因分析模型，以至于大量事故隐患数据潜在的价值没有得到应用。为此，采用LDA主题挖掘算法，构建事故致因主题挖掘模型，采用困惑度（Perplexity）指标来确定最佳主题个数，并利用语义网络图分析事故隐患之间的内在关系，直观地反映了采煤工作面主要事故致因及致因因素间的关联。以山东某矿2013年8月~2019年9月18925条事故隐患数据为例，通过对主题模型挖掘，抽取事故致因主题，并绘制语义网络图，应用社会网络分析方法分析事故隐患要素之间的关联关系。结果表明：管理缺陷、人员不安全行为、设备不安全状态及环境不安全状态是导致采煤工作面事故的主要因素，它们之间的相互作用导致了事故的发生。 |
| 关 键 词： | 采煤工作面; 事故隐患; 事故致因; 主题挖掘; 社会网络分析 |
| DOI： | 10.57237/j.ssrf.2022.01.006 |
View:
Downloads:
收稿日期: 2022-10-11
接受日期: 2022-11-04
在线出版日期: 2022-11-23
Research on Cause of Coal Face Accident Based on Topic Model Mining
1. Beijing CCID Industry & Information Engineering Supervision Center Co., Ltd, Beijing 100089, China
2. Department of Mechanical and Electrical Engineering, Taishan Polytechnic, Tai’an 271001, China
| Abstract: | As the first production site of coal, mining working face is the place where coal mine accidents occur frequently. In order to avoid or reduce the occurrence of accidents, it is necessary to eliminate and discover the hidden danger of accidents in mining face in time. However, in the past, the cause of accidents mainly adopts the method of qualitative analysis, and lacks the effective analysis model of the cause of accidents, so that the potential value of a large number of hidden danger data of accidents has not been applied. Therefore, the LDA topic mining algorithm is used to build the accident cause topic mining model, and the Perplexity index is used to determine the best number of topics, and the semantic network graph is used to analyze the internal relationship between the hidden dangers of accidents, which directly reflects the main accident cause and the correlation between the cause factors of coal mining face. Taking 18925 accident hidden danger data of a mine in Shandong Province from August 2013 to September 2019 as an example, the topic of accident causes was extracted by mining the topic model, and the semantic network diagram was drawn. The social network analysis method was applied to analyze the association relationship between the factors of accident hidden danger. The results show that management defects, unsafe behavior of personnel, unsafe state of equipment and unsafe state of environment are the main factors leading to coal face accidents, and their interaction leads to the occurrence of accidents. |
| Keywords: | Coal Working Face; Hidden Danger of Accident; Cause of Accident; Theme Mining; Tocial Network Analysis |
参考文献
| [1] | Meng Fanqiang. Safety Warning Model of Coal Face Based on FCM Fuzzy Clustering and GA-BP Neural Network [J]. Symmetry, 2021, 13 (6): 1082. |
| [2] | 张东, 聂百胜, 王龙康. 我国煤矿安全生产事故的致灾因素分析 [J]. 中国安全生产科学技术, 2013, 9 (5): 136-140. |
| [3] | Kai W, Jiang S, Zhang W, et al. Study on PCPR Security System Construction in Coal Mine [J]. Procedia Engineering, 2011, 26: 2044-2050. |
| [4] | 诸利一, 吕文生, 杨鹏, 等. 2007—2016年全国煤矿事故统计及发生规律研究 [J]. 煤矿安全, 2018, v. 49; No. 528 (07): 237-240. |
| [5] | 孙玲, 宫立昊. 2019年国内煤矿安全事故统计分析及对策研究 [J]. 决策探索 (中), 2020, No. 642 (02): 22-23. |
| [6] | 陈娟, 赵耀江. 近十年来我国煤矿事故统计分析及启示 [J]. 煤炭工程, 2012 (03): 145-147. |
| [7] | 许多康. 国有煤矿事故统计分析与防控对策 [J]. 山东煤炭科技, 2019 (12): 190-193. |
| [8] | 李波, 巨广刚, 王珂, 等. 2005—2014年我国煤矿灾害事故特征及规律研究 [J]. 矿业安全与环保, v. 43; No. 227 (03): 111-114. |
| [9] | 康国峰. 煤矿安全隐患排查与治理分级分类管理机制探索 [J]. 煤矿安全, 2011, 42 (010): 158-160. |
| [10] | 王丹, 刘琳, 张小曼. 灰色关联度法在煤矿本质安全评价中的改进及应用 [J]. 中国安全生产科学技术, 2013, 9 (001): 151-157. |
| [11] | Meng, Fanqiang and Li, Chunxia.Safety Warning of Coal Mining Face Based on Big Data Association Rule Mining [J].ournal of Computational Methods in Sciences and Engineering, 2022, 22 (4): 1035-1052. |
| [12] | 夏春艳. 数据挖掘技术与应用 [M]. 冶金工业出版社, 2014. |
| [13] | 牛毅, 樊运晓, 高远. 基于数据挖掘的化工生产事故致因主题抽取 [J]. 中 国安全生产科学技术, 2019, 15 (10): 165-170. |
| [14] | 徐守坤, 周佳, 李宁, 等. 基于word2vec和LDA的文本主题 [J]. 计算机工程与设计, 2018, 39 (9): 2764-2769. |
| [15] | 何天文, 王红. 基于语义语法分析的中文语句困惑度评价 [J]. 计算机应用研究, 2017, 34 (12): 3638-3546. |
| [16] | 谭章禄, 王兆刚, 胡翰, 姜萱, 彭胜男.基于文本聚类的煤矿安全隐患类型挖掘研究 [J]. 中国安全科学学报, 2019, 29 (3): 145-148. |
