Big data analysis and visualization of potential hazardous risks of the mine based on text mining
-
摘要: 基于大數據分析技術，構建了礦山安全隱患多維度分析模型，分析了隱患在時間和空間兩個維度上的分布規律；利用主題挖掘模型將眾多隱患信息歸類，得到了13個隱患主題；利用關聯規則挖掘模型探究了不同隱患之間的內在聯系，并利用R編程語言對上述結果進行可視化展示。通過對安全隱患的分析研究不僅充分利用了礦山隱患數據，避免了數據資源的浪費，同時也對礦山井下事故預防有一定的指導價值。Abstract: Compared with other production industries, metal mine is recognized as a high accident rate and the highest casualty rate due to the bad working environment. Therefore, safety production is the key concern of mining enterprises. With the attention of enterprises to safety problems and the increasing improvement of mine safety management system, many mines have established secure big data platform to effectively manage production and ensure the safety of underground operation, receiving the safety hazard information from daily safety inspection into the platform. However, due to the data of security risks are unstructured short texts with the operation of the enterprise, including the data recorded in the platform presents the characteristics of complex data content, large data scale, and non-standard data records. Moreover, due to the lack of an effective text analysis model, a small part of the security risk data is only used for simple analysis such as report analysis and data statistics, whereas more data is stored in a secure big data platform. Thus, the data did not play a guiding role in production, resulting in a waste of these valuable data resources. In order to explore the internal relationship between hidden danger data and the rule of hidden danger occurrence, based on big data analysis technology, this paper constructed a multi-dimensional analysis model of mine safety hidden danger. We analyzed the distribution law of hidden danger in two dimensions of time and space, used the topic mining model to classify hidden danger information, and obtained 13 hidden danger topics, using association rules to mine hidden danger. The model explores the internal relationship between different hidden dangers and uses an R programming language to visualize the above results. The results made full use of the mine hidden danger data and avoided the waste of data resources through the analysis and research of the hidden danger with a certain guiding value for preventing mine accidents.
-
Key words:
- mine safety /
- text mining /
- data of hidden danger /
- data analysis /
- data visualization
-
圖 1 基于大數據分析的隱患分析模型流程
Figure 1. Hidden danger analysis model process based on big data analysis
圖 3 基于Apriori算法礦山安全隱患關聯規則挖掘流程
Figure 3. Mining process of association rules for mine safety hazard based on Apriori algorithm
圖 10 礦山井下安全隱患關聯規則散點圖
Figure 10. Scatter diagram of association rules for underground safety hazards
圖 11 基于圖的礦山安全隱患關聯規則可視化
Figure 11. Visualization of mine hidden danger association rules based on graph
表 1 安全隱患高頻詞(部分)
Table 1. High frequency words of hidden danger (part)
Number Hidden danger vocabulary Word frequency Proportion/
%Number Hidden danger vocabulary Word frequency Proportion/
%1 Support 9493 5.27 11 Civilized production 2440 1.36 2 Roof 9174 5.10 12 Pavement 2327 1.29 3 Pumice 8756 4.86 13 Roadway’s sides 2232 1.24 4 Illumination 6145 3.41 14 Not in place 2190 1.22 5 Head-on 5237 2.91 15 Fan 2112 1.17 6 Much more 4931 2.74 16 Work 2099 1.17 7 Hydrops 2909 1.62 17 Distribution box 2011 1.12 8 Roof and sidewalls 2773 1.54 18 Fracture 1900 1.06 9 Facilities 2659 1.48 19 Explosive 1798 1.00 10 Rock bolt 2456 1.36 20 Jeep 1538 0.85 
下載: 導出CSV
表 2 不同年份共有隱患詞匯統計表(部分)
Table 2. Statistical table of common hidden danger vocabulary (part)
Hidden danger vocabulary Word frequency 2013 2014 2015 2016 2017 2018 2019 Roof 605 872 818 1080 1358 1716 1246 Illumination 451 547 405 489 938 1176 1106 Rock bolt 161 220 360 259 321 322 235 Pumice 593 850 1014 1326 1317 1748 1234 Distribution box 176 226 156 237 333 391 303 Head-on 484 477 387 618 765 1242 794 Support 704 781 849 1152 1274 2116 1687 Fan 221 254 167 210 363 302 313 Hydrops 280 280 278 296 459 592 484 … … … … … … … …
下載: 導出CSV
表 3 高頻隱患地點統計表(前20)
Table 3. Statistical table of high frequency hidden danger location (top 20)
Hidden danger location Quantity Hidden danger location Number Slope mouth 509 S13155 149 S12186 254 X06111 144 S14186 239 X08059 141 X07097 228 S18156 140 X07087 226 X08055 132 X07105 225 X05103 123 S13186 202 S15186 122 Assistant ramp 197 S10167 115 X09105 170 X05111 108 Main ramp 164 West ventilating shaft 105
下載: 導出CSV
表 4 BTM礦山安全隱患主題與隱患主題詞表
Table 4. BTM mine safety hidden danger theme and hidden danger keywords list
Number Safety hidden danger theme Hidden danger keywords 1 Hidden danger of support Support, roof, roadway’s sides, network degree, measures, not in place, invalid, fracture 2 Hidden danger of roof Roof, joint, caving, fragment, pumice, dangerous rock, crack, development 3 Hidden danger of transport Overload, ramp, violation, jeep, down, fire extinguisher, load-haul-dump unit 4 Hidden danger of rock bolt Rock bolt, network degree, not in time, follow-up, lack, long- cable, too long 5 Hidden danger of pipeline Wind belt, cable, set up, follow-up, damaged, hang, stringing, drop, water pipe 6 Hidden danger of ventilation and three prevention Fire extinguisher, fire water pipe, fire box, dust, airflow, oxygen, air quality, local ventilation 7 Hidden danger of operation Operation, grouting, excavation, scene, top brush, people, construction, not completely 8 Hidden danger of safety protection Safety hat, protect, protective fence, sign, carapace, measures, sign 9 Hidden danger of electromechanical Fan, distribution box, transformer, switch, ground wire, grounding electrode, cable 10 Hidden danger of blasting operation Smooth blasting, explosive, detonating tube, explosive box, lock, lying around 11 Hidden danger of road Pavement, out-of-flatness, silt, potholes, sundries, hydrops 12 Hidden danger of water disaster Hydrops, too much, deeper, ditch, water pump, puddles, drain 13 Hidden danger of environmental Silt, mud, clean up, poor, hydrops, sundries, purling
下載: 導出CSV
表 5 礦山安全隱患關聯規則挖掘(部分)
Table 5. Mining association rules of mine hidden danger (part)
Number Association rules Support Confidence Lift Count 1 {driver}=>
{safety hat}0.0050427 0.7208333 51.200060 173 2 {pry detection}=>
{top brush}0.0158860 0.9663121 49.332244 545 3 {pavement}=>
{potholes, uneven}0.0124173 0.9487751 41.891410 426 4 {network degree}
=>{bigger}0.0123299 0.9276316 22.348495 423 5 {roof and sidewalls, head-on}=>{pumice} 0.0139039 0.9173077 4.151725 477 6 {roadway’s sides, illumination, facility}=>{pumice} 0.0102603 0.9048843 4.095497 352 7 {lying around}=>
{explosive}0.0091235 0.8743017 22.317461 313 8 {landing}=>{fan} 0.0063544 0.5561224 11.322785 218 9 {pumice, ventilation facilities}
=>{illumination}0.0061503 0.5926966 4.800199 211 10 {residual explosive}=>{roof} 0.0123590 0.7054908 3.445306 424
下載: 導出CSV
www.77susu.com<span id="fpn9h"><noframes id="fpn9h"><span id="fpn9h"></span> <span id="fpn9h"><noframes id="fpn9h"> <th id="fpn9h"></th> <strike id="fpn9h"><noframes id="fpn9h"><strike id="fpn9h"></strike> <th id="fpn9h"><noframes id="fpn9h"> <span id="fpn9h"><video id="fpn9h"></video></span> <ruby id="fpn9h"></ruby> <strike id="fpn9h"><noframes id="fpn9h"><span id="fpn9h"></span> -
參考文獻
[1] Bi H T, Li G Q. Research on the closed circuit management system of safety production in gold mines. Gold, 2014, 35(8): 1 doi: 10.11792/hj20140801畢洪濤, 李國清. 黃金礦山安全生產閉環管理體系研究. 黃金, 2014, 35(8):1 doi: 10.11792/hj20140801 [2] Meng X F, Li K Y, Liu F. Study on coal mine safety risk pre-control based on three-levels nested management mode. China Saf Sci J, 2013, 23(4): 102孟現飛, 李克業, 劉飛. 基于3級嵌套安全管理模式的煤礦安全風險預控研究. 中國安全科學學報, 2013, 23(4):102 [3] Zhao D F, Shen Y Q, Zhao Z Q, et al. Risk classification method for accident potential based on development and control measures of accident. China Saf Sci J, 2012, 22(4): 71 doi: 10.3969/j.issn.1003-3033.2012.04.013趙東風, 申玉琪, 趙志強, 等. 基于事故發展與控制的隱患分級方法. 中國安全科學學報, 2012, 22(4):71 doi: 10.3969/j.issn.1003-3033.2012.04.013 [4] Martin E B, Morris A J. Non-parametric confidence bounds for process performance monitoring charts. J Process Control, 1996, 6(6): 349 doi: 10.1016/0959-1524(96)00010-8 [5] Martin E B, Morris A J. An overview of multivariate statistical process control in continuous and batch process performance monitoring. Trans Inst Meas Control, 1996, 18(1): 51 doi: 10.1177/014233129601800107 [6] Dunia R, Qin S J, Edgar T F. Identification of faulty sensors using principal component analysis. AIChE J, 1996, 42(10): 2797 doi: 10.1002/aic.690421011 [7] Li J, Cui B, Wang Q, et al. Information system for identification and control of major coal mine disasters // Innovative coal enterprise development and informatization Summit Forum. Lanzhou, 2010: 56李季, 翟勃, 汪群, 等. 煤礦重大災害辯識和控制信息系統 // 創新型煤炭企業發展與信息化高峰論壇論文集. 蘭州, 2010: 56 [8] Qin W J. The Research and Application of the Three-Dimensional Method in Coalmine Gas Explosion Hazard Identifiction [Dissertation]. Taiyuan: Taiyuan University of Technology, 2015秦文靜. 煤礦瓦斯爆炸危險源三維辨識研究及應用[學位論文]. 太原: 太原理工大學, 2015 [9] Zhang B L, Wang X Q, He Y R, et al. Construction of safety hazard identification and investigation system of coal mine based on ontology. Saf Coal Mines, 2018, 49(5): 239張寶隆, 王向前, 何葉榮, 等. 基于本體的煤礦事故隱患辨識排查系統構建. 煤礦安全, 2018, 49(5):239 [10] Ma X P, Dai W. Research status and application prospect of big data technology in coal industry. Ind Mine Autom, 2018, 44(1): 50馬小平, 代偉. 大數據技術在煤炭工業中的研究現狀與應用展望. 工礦自動化, 2018, 44(1):50 [11] Sun J P. Accident analysis and big data and Internet of Things in coal mine. Ind Mine Autom, 2015, 41(3): 1孫繼平. 煤礦事故分析與煤礦大數據和物聯網. 工礦自動化, 2015, 41(3):1 [12] Tan Z L, Chen X, Song Q Z, et al. Analysis for the potential hazardous risks of the coal mines based on the socalled text mining. J Saf Environ, 2017, 17(4): 1262譚章祿, 陳曉, 宋慶正, 等. 基于文本挖掘的煤礦安全隱患分析. 安全與環境學報, 2017, 17(4):1262 [13] Qian Y H. Application of data mining algorithm in gas safety prediction. Coal Technol, 2018, 37(5): 207錢宇虹. 數據挖掘算法在瓦斯安全預測中的應用. 煤炭技術, 2018, 37(5):207 [14] Shi J B, Shi J H. Application of correlation analysis on data mining in coal mine gas safety monitoring and early warning. China Energy Environ Prot, 2017, 39(8): 1石記斌, 石記紅. 關聯分析數據挖掘在煤礦瓦斯安全監測預警中的應用. 能源與環保, 2017, 39(8):1 [15] Lei Y B, Chen Z B, Zeng J C, et al. Research on causal chain of coal mine gas accidents based on association rule. Saf Coal Mines, 2016, 47(8): 240雷煜斌, 陳兆波, 曾建潮, 等. 基于關聯規則的煤礦瓦斯事故致因鏈研究. 煤礦安全, 2016, 47(8):240 [16] Liu H B, Li C H. Research on occupational health and safety management information system in intelligent mine. Coal Sci Technol, 2019, 47(3): 87劉海濱, 李春賀. 智慧礦山職業健康安全監管信息系統研究. 煤炭科學技術, 2019, 47(3):87 [17] Henriques V, Malekian R. Mine safety system using wireless sensor network. IEEE Access, 2016, 4: 3511 doi: 10.1109/ACCESS.2016.2581844 [18] Zhou X T. Mine safety monitoring, early warning and integrated management information system. World Nonferrous Met, 2017(19): 26周雪田. 礦山安全監測預警與綜合管理信息系統. 世界有色金屬, 2017(19):26 [19] Zhang D W. Analysis of coal mine safety hidden danger trends based on OLAM. Coal Eng, 2015, 47(5): 139 doi: 10.11799/ce201505045張大偉. 基于OLAM的煤礦企業安全隱患趨勢分析. 煤炭工程, 2015, 47(5):139 doi: 10.11799/ce201505045 [20] Yao Q G, Zhao L X, Zhang X M. Fuzzy comprehensive evaluation model of coal mine safety management information system. Min Saf Environ Prot, 2017, 44(6): 120 doi: 10.3969/j.issn.1008-4495.2017.06.027姚慶國, 趙麗霞, 張學睦. 煤礦安全管理信息系統模糊綜合評價模型. 礦業安全與環保, 2017, 44(6):120 doi: 10.3969/j.issn.1008-4495.2017.06.027 [21] Cheng J W, Yang S Q. Data mining applications in evaluating mine ventilation system. Saf Sci, 2012, 50(4): 918 doi: 10.1016/j.ssci.2011.08.003 [22] Tang X B, Xiang K. Hotspot mining based on LDA model and microblog heat. Libr Inf Serv, 2014, 58(5): 58唐曉波, 向坤. 基于LDA模型和微博熱度的熱點挖掘. 圖書情報工作, 2014, 58(5):58 [23] Blei D M, Ng A Y, Jordan M I, Lafferty J. Latent Dirichlet Allocation. J Mach Learn Res, 2003, 3: 993 [24] Ma T B, Xu F. Research and improvement for apriori algorithm of association rule mining. J Lanzhou Polytech Coll, 2010, 17(1): 13馬廷斌, 徐芬. 關聯規則挖掘中Apriori算法的研究與改進. 蘭州工業高等專科學校學報, 2010, 17(1):13 [25] Cui Y, Bao Z Q. Survey of association rule mining. Appl Res Comput, 2016, 33(2): 330 doi: 10.3969/j.issn.1001-3695.2016.02.002崔妍, 包志強. 關聯規則挖掘綜述. 計算機應用研究, 2016, 33(2):330 doi: 10.3969/j.issn.1001-3695.2016.02.002 [26] Krause S, Busch F. New insights into road accident analysis through the use of text mining methods // 2019 6th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS). Cracow, 2019: 1 [27] Zhang Y. Establishment and application of hidden danger automatic investigation and automatic report system to coal mine safety production accident. Coal Eng, 2014, 46(11): 150 doi: 10.11799/ce201411047張勇. 煤礦安全生產事故隱患自查自報系統建立與應用. 煤炭工程, 2014, 46(11):150 doi: 10.11799/ce201411047 [28] Ministry of Emergency Management of People’s Republic of China. Interim Provisions on the Investigation and Treatment of Hidden Dangers of Work Safety [EB/OL]. Ministry of Emergency Management of the People’s Republic of China (2008-01-10) [2020-10-23]. http://www.mem.gov.cn/gk/gwgg/zjl_01/200801/t20080110_233738.shtml中華人民共和國應急管理部. 安全生產事故隱患排查治理暫行規定[EB/OL]. 中華人民共和國應急管理部 (2008-01-10) [2020-10-23]. http://www.mem.gov.cn/gk/gwgg/zjl_01/200801/t20080110_233738.shtml [29] Ministry of Emergency Management of People’s Republic of China. Judgment Standards for Hidden Dangers of Major Production Safety Accidents in Metal and Nonmetal Mine (Trial) [EB/OL]. Ministry of Emergency Management of People’s Republic of China (2017-09-05) [2020-10-23].https://www.mem.gov.cn/gk/gwgg/201709/t20170905_241758.shtml中華人民共和國應急管理部. 金屬非金屬礦山重大生產安全事故隱患判定標準(試行)[EB/OL]. 中華人民共和國應急管理部 (2017-09-05) [2020-10-23].https://www.mem.gov.cn/gk/gwgg/201709/t20170905_241758.shtml -
點擊查看大圖
圖(11) / 表(5)
計量
- 文章訪問數: 3966
- HTML全文瀏覽量: 520
- PDF下載量: 119
- 被引次數: 0
