Semantic matching model of potential safety hazards in hydroelectric project construction

doi:10.16265/j.cnki.issn1003-3033.2024.12.0795

Abstract

Abstract:

In order to assist in the development of safety hazard management measures for hydropower project construction, the safety hazard texts accumulated during the construction inspection of hydropower projects were collected. Entities and relationships from the semi-structured safety hazard texts were extracted using Python. A knowledge graph of safety hazards was constructed and imported into the neo4j graph database for storage. A Sentence-Bidirectional Encoder Representations from Transformer (BERT) model based on bidirectional coding was built for the semantic matching of construction hazards in hydropower projects. The deep semantic features of target hazards and historical hazards were learned, and the historical safety hazards most similar to target hazards were recommended. Using the Cypher query statement, the governance measures corresponding to the historical security risk were searched. The results show that the Sentence-BERT model has an accuracy of 96.48% in identifying architecturally and historically similar safety hazards, which is significantly better than BERT, Word2vec-Deep Semantic Similarity Model (Word2vec-DSSM), and BERT-DSSM models. Among 150 randomly selected target safety hazard data, the accuracy rate of testing historical similar safety hazard suggestions reaches 92%, and the retrieval effect of hazard management measures is demonstrated through the hazard knowledge graph, which verifies the applicability and effectiveness of the method.

Key words: hydropower project construction, safety hazard, semantic matching, management measures, intelligent recommendation, knowledge graph

CLC Number:

X948

CHEN Shu, WANG Dianxue, YANG Yingliu, CAO Kunyu, NIE Benwu. Semantic matching model of potential safety hazards in hydroelectric project construction[J]. China Safety Science Journal, 2024, 34(12): 40-47.

Figures/Tables 10

Table 1

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 2

Table 3

Fig.6

Fig.7

References 25

[1]	樊启祥, 林鹏, 魏鹏程, 等. 水电工程安全事故发生机制与管理对策[J]. 中国安全科学学报, 2019, 29(1):144-149. doi: 10.16265/j.cnki.issn1003-3033.2019.01.024
	FAN Qixiang, LIN Peng, WEI Pengcheng, et al. Hydropower engineering safety accident occurrence mechanism and management strategies[J]. China Safety Science Journal, 2019, 29(1):144-149. doi: 10.16265/j.cnki.issn1003-3033.2019.01.024
[2]	王仁超, 张毅伟, 毛三军. 水电工程施工安全隐患文本智能分类与知识挖掘[J]. 水力发电学报, 2022, 41(11):96-106.
	WANG Renchao, ZHANG Yiwei, MAO Sanjun. Intelligent text classification and knowledge mining of hidden safety hazards in hydropower engineering construction[J]. Journal of Hydroelectric Engineering, 2022, 41(11):96-106.
[3]	靳嵩, 朱艳, 吴可嘉, 等. 基于BERT的海上船舶安全隐患分类[J]. 船舶工程, 2023, 45(增1):381-384.
	JIN Song, ZHU Yan, WU Kejia, et al. Classification of ship safety hazards of marine vessel based on BERT[J]. Ship Engineering, 2023, 45(S1):381-384.
[4]	陈述, 习俊博, 王建平, 等. 水电工程施工安全隐患关联规则挖掘[J]. 中国安全科学学报, 2021, 31(8):75-82. doi: 10.16265/j.cnki.issn1003-3033.2021.08.011
	CHEN Shu, XI Junbo, WANG Jianping, et al. Mining association rules of near-misses of hydropower projects construction[J]. China Safety Science Journal, 2021, 31(8):75-82. doi: 10.16265/j.cnki.issn1003-3033.2021.08.011
[5]	陈述, 刘文濯, 邵波, 等. 水电工程施工安全隐患时空分布特征研究[J]. 中国安全科学学报, 2022, 32(11):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.11.2227
	CHEN Shu, LIU Wenzhuo, SHAO Bo, et al. Research on spatial and temporal distribution characteristics of safety nazards in hydropower projects[J]. China Safety Science Journal, 2022, 32(11):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.11.2227
[6]	肖琪耀, 贾宝山, 徐以诺, 等. 基于深度学习模型的煤矿安全隐患数据主题挖掘[J]. 中国安全生产科学技术, 2024, 20(4):49-55.
	XIAO Qiyao, JIA Baoshan, XU Yinuo, et al. Topics mining on potential safety hazard data of coal mine based on deep learning models[J]. Journal of Safety Science and Technology, 2024, 20(4):49-55.
[7]	宁桂峰, 高龙, 刘利平. 基于遗传算法的采煤工作面隐患数据关联规则挖掘[J]. 采矿与岩层控制工程学报, 2024, 6(2):136-145.
	NING Guifeng, GAO Long, LIU Liping. Mining hidden danger data association rules of coal mining face based on genetic algorithm[J]. Journal of Mining and Strata Control Engineering, 2024, 6(2):136-145.
[8]	陈述, 鲁世立, 王建平, 等. 缆机安全运行本体知识表示及复用方法[J]. 中国安全科学学报, 2024, 34(3):93-100. doi: 10.16265/j.cnki.issn1003-3033.2024.03.1321
	CHEN Shu, LU Shili, WANG Jianping, et al. Representation and reuse of ontology knowledge for safe operation of cable crane[J]. China Safety Science Journal, 2024, 34(3):93-100. doi: 10.16265/j.cnki.issn1003-3033.2024.03.1321
[9]	臧洁, 周万林, 王妍. 融合多头注意力机制和孪生网络的语义匹配方法[J]. 计算机科学, 2023, 50(12):294-301. doi: 10.11896/jsjkx.221000083
	ZANG Jie, ZHOU Wanlin, WANG Yan. Semantic Matching method integrating multi-head attention mechanism and siamese network[J]. Computer Science, 2023, 50(12):294-301. doi: 10.11896/jsjkx.221000083
[10]	阮光册, 钟静涵, 张祎笛. 基于深度学习的术语识别研究综述[J]. 数据分析与知识发现, 2024, 8(4):64-75. doi: 10.11925/infotech.2096-3467.2023.0158
	RUAN Guangce, ZHONG Jinghan, ZHANG Yidi. Review of term recognition studies based on deep learning[J]. Data Analysis and Knowledge Discovery, 2024, 8(4):64-75. doi: 10.11925/infotech.2096-3467.2023.0158
[11]	管立本, 李实. 融合多粒度语义信息和知识图谱的中文医疗问答匹配模型[J]. 计算机工程与应用, 2024, 60(14):152-161. doi: 10.3778/j.issn.1002-8331.2305-0453
	GUAN Liben, LI Shi. A Chinese medical Q&A matching model based on multi-granularity semantic information and knowledge graph[J]. Computer Engineering and Applications, 2024, 60(14):152-161. doi: 10.3778/j.issn.1002-8331.2305-0453
[12]	许童羽, 赵冬雪, 周云成, 等. 基于word2vec和Attention-Seq2Seq的水稻病虫害智能问答方法研究[J]. 沈阳农业大学学报, 2019, 50(3):378-384.
	XU Tongyu, ZHAO Dongxue, ZHOU Yuncheng, et al. Research on method of intelligent Q&A for rice pests and diseases based on word2vec and Attention-Seq2Seq[J]. Journal of Shenyang Agricultural University, 2019, 50(3):378-384.
[13]	李琳, 段围, 周栋, 等. 基于深度语义匹配的法律条文推荐方法[J]. 软件学报, 2022, 33(7):2618-2632.
	LI Lin, DUAN Wei, ZHOU Dong, et al. Law article recommendation approach based on deep semantic matching[J]. Journal of Software, 2022, 33(7):2618-2632.
[14]	张晓健, 张栋梁, 李明超, 等. 面向质量检测的混凝土坝施工规范智能检索[J]. 水力发电学报, 2023, 42(4):114-125.
	ZHANG Xiaojian, ZHANG Dongliang, LI Mingchao, et al. Intelligent retrieval of concrete dam construction specification for quality inspection[J]. Journal of Hydroelectric Engineering, 2023, 42(4):114-125.
[15]	罗香玉, 杜浩, 华颖, 等. 一种煤矿顶板灾害防治知识图谱构建方法[J]. 工矿自动化, 2024, 50(6):54-60.
	LUO Xiangyu, DU Hao, HUA Ying, et al. A method for constructing a knowledge graph of coal mine roof disaster prevention and control[J]. Journal of Mine Automation, 2024, 50(6):54-60.
[16]	张吉祥, 张祥森, 武长旭, 等. 知识图谱构建技术综述[J]. 计算机工程, 2022, 48(3):23-37. doi: 10.19678/j.issn.1000-3428.0061803
	ZHANG Jixiang, ZHANG Xiangsen, WU Changxu, et al. Survey of knowledge graph construction techniques[J]. Computer Engineering, 2022, 48(3):23-37. doi: 10.19678/j.issn.1000-3428.0061803
[17]	王莉, 王建平, 许娜, 等. 基于知识图谱的地铁工程事故知识建模与分析[J]. 土木工程与管理学报, 2019, 36(5):109-114,122.
	WANG Li, WANG Jianping, XU Na, et al. Knowledge graph-based metro engineering accidents knowledge modeling and analysis[J]. Journal of Civil Engineering and Management, 2019, 36(5):109-114,122.
[18]	郭天颖, 茆晓阳, 段齐骏, 等. 基于知识图谱的地震救援装备智能管理方法[J]. 中国安全科学学报, 2024, 34(7):239-245. doi: 10.16265/j.cnki.issn1003-3033.2024.07.2014
	GUO Tianying, MAO Xiaoyang, DUAN Qijun, et al. Intelligent management and scheduling approach for earthquake rescue equipment based on knowledge graph[J]. China Safety Science Journal, 2024, 34(7):239-245. doi: 10.16265/j.cnki.issn1003-3033.2024.07.2014
[19]	HU Caiping, SNU Xuekui, DAI Hua, et al. Research on log anomaly detection based on sentence-BERT[J]. Electronics, 2023, 12(17):DOI: 10.3390/ELECTRONICS12173580.
[20]	叶文豪, 王东波, 沈思, 等. 基于孪生网络的基金与受资助论文相关性判别模型构建研究[J]. 情报学报, 2020, 39(6):609-618.
	YE Wenhao, WANG Dongbo, SHEN Si, et al. Research on constructing a model of correlation discrimination between funds and funded papers based on siamese network[J]. Journal of the China Society for Scientific and Technical Information, 2020, 39(6):609-618.
[21]	苏锦钿, 洪晓斌, 余珊珊. 基于多模型集成的语义文本相似性判断[J]. 华南理工大学学报:自然科学版, 2022, 50(4):1-9.
	SU Jindian, HONG Xiaobin, YU Shanshan. Semantic textual similarity justification based on multi-model ensemble[J]. Journal of South China University of Technology:Natural Science Edition, 2022, 50(4):1-9.
[22]	CAO Kunyu, CHEN Shu, ZHANG Xinkai, et al. Identification of causative factors for fatal accidents in the electric power industry using text categorization and catastrophe association analysis techniques[J]. Alexandria Engineering Journal, 2024,102:290-308.
[23]	夏占杰, 张贝克, 高东. 基于数据增强的HSE检查纪要命名实体识别[J]. 中国安全科学学报, 2022, 32(12):53-62. doi: 10.16265/j.cnki.issn1003-3033.2022.12.2727
	XIA Zhanjie, ZHANG Beike, GAO Dong. Named entity recognition of HSE inspection minutes based on data enhancement[J]. China Safety Science Journal, 2022, 32(12):53-62. doi: 10.16265/j.cnki.issn1003-3033.2022.12.2727
[24]	钱杨舸, 秦小林, 张思齐, 等. 基于深度学习的文本语义匹配综述[J]. 软件导刊, 2022, 21(12):252-261.
	QIAN Yangge, QIN Xiaolin, ZHANG Siqi, et al. Survey of texts semantic matching based on deep learning[J]. Software Guide, 2022, 21(12):252-261.
[25]	JADHAV S S, THEPADE S. Fake news identification and classification using dssm and improved recurrent neural network classifier[J]. Applied Artificial Intelligence, 2019, 33(12):1058-1068.

编号	目标隐患记录	历史隐患记录	标签
1	顶拱挂网施工,汽车吊吊装范围未警戒防护	下岔洞,7号尾水下游钢筋台车吊装无警戒	1
2	5号尾调溜渣井下部未施工时未设置警示警戒	7号尾水连接管二层竖井下部警戒线缺失	1
︙	︙	︙	︙
3 126	安装间上游排架区,一配电箱有一闸多接现象	进水口EL744平台,二级配电箱箱门脱落	0
︙	︙	︙	︙
6 250	主厂南厂顶交通洞及5-2廊道南侧排架竹跳板破损	开挖道路临边警示带破损	0

模型	A	P	R	F₁值
Word2vec-DSSM	86.61	87.72	86.62	86.64
BERT-DSSM	84.55	83.34	84.49	84.12
BERT	90.43	89.55	91.57	91.24
Sentence-BERT	96.48	95.80	97.23	96.51

目标隐患记录	历史隐患记录	相似度值
左岸4号帷幕灌浆洞洞口K0+10位置,氧气瓶和乙炔瓶混放,未到达相关规定的安全距离,存在安全隐患	氧气瓶与乙炔气瓶未按安全距离摆放,存在爆炸危险	0.976 5
钢筋绑扎区域无上下通道,施工作业人员沿斜放钢筋上下	7号机背坡备仓过程中,该段无通道,人员在已安装的钢筋上行走	0.963 2
施工现场防护栏底部挡脚板缺失,存在物体打击风险	654马道防护挡脚板缺失,且文明施工较差	0.982 7
电焊机插头使用钢筋代替,存在触电风险	抗力体马道电焊机电焊线插头用钢筋代替,存在安全隐患	0.968 8
炸药临时存在点未警戒,无专人看守	0+230处炸药临时存放无人看守,无警示灯,警示牌	0.944 6
砼泵管使用K1800塔机长短泵管混吊入仓易发生坠物伤害风险	下游围堰防渗墙浇筑溜槽稳定性不够,存在倾倒危险	0.653 1