Water inrush warning system of deep limestone in Panxie Mining Area based on multi-source data mining
-
摘要: 深层灰岩水在长时间水岩耦合作用下各含水层的水化学成分有所不同,但随着地壳运动、采动影响等因素导致不同含水层产生水力联系。重大的突水事故都是深层高压灰岩水以浅层灰岩水为通道突入矿井发生的。依据对淮南煤田潘谢矿区9对矿井2015—2018年182个地面水文观测孔的水位数据及潘二矿突水后各水文观测孔水位变化的时空规律,得出水文观测孔的水位变化数据比水位高程数据更灵敏,潘谢矿区深层灰岩水由下向上对浅层灰岩水进行补给,通过聚类分析算法识别出矿井与深层灰岩水存在补给关系的浅层灰岩含水层区域;另一方面基于改进的随机森林算法对收集的7 000多条矿井水质化验资料进行分析,基于错分数据识别出与深层灰岩水水力联系紧密的各矿含水层信息。综合分析水位变化数据聚类分析结果,得出各矿井的突水风险区域。基于含水层分类显著因子、水化学空间分布特征,结合温度、流量、水位、水质等参数的高精度传感器,构建快速准确突水预警系统,对矿井出水点进行智能监测,为实施防治水措施提供快速、可靠的依据,可以极大地避免矿井发生突水事故和减少突水事故产生的损失。Abstract: Under the long-term water-rock interaction, the hydrochemical composition of deep limestone water in each aquifer is different. Such factors as crustal movement and mining influence have led to the hydraulic connection between aquifers. Major water inrush accidents often occur when the deep high-pressure limestone water bursts into the mine with the shallow limestone water as a channel. On the basis of water level data of 182 surface hydrological observation holes in nine coal mines in Panxie Mining Area of Huainan Coalfield from 2015 to 2018, and the temporal and spatial law of water level change of each hydrological observation hole after the water inrush occurred in Pan’er Mine, we found that the water level change data of hydrological observation holes are more sensitive than the water level elevation data, and the deep limestone water in Panxie Mining Area recharged from bottom to top to the shallow limestone water. The shallow limestone water area with a recharge relationship between aquifer and deep limestone water is identified by clustering analysis algorithm. On the other hand, the water quality test data of more than 7 000 mines are analyzed based on the improved random forest algorithm, and the information on aquifers closely related to the water and hydraulic of deep limestone is identified based on the misclassification data. With the results of clustering analysis of water level change data being analyzed comprehensively, the water inrush risk area of each mine is obtained. Based on the significant classification factors and the hydrochemical characteristics of spatial distribution of each aquifer, we constructed a fast and accurate water inrush warning system by using high precision sensors of temperature, pressure, water level and water quality and other high-precision sensors, as so to monitor the water inrush point of a mine on-line. It provides a fast and reliable basis for water prevention and control measures in construction, and can greatly avoid water inrush accidents in mines, reducing the losses arising from sudden water accidents.
-
表 1 突水前后地面水文观测孔水位变化统计
Table 1 Statistical table of water level changes in surface hydrological observation holes before and after water inrush
煤矿 新生界松散层含水层 太原组灰岩含水层 奥陶系、寒武系灰岩含水层 水位变化/m 水位高程/m 水位变化/m 水位高程/m 水位变化/m 水位高程/m 潘二矿 –0.1~1.7/0.4 7.9 –20~84/49 –253 313.0~377.0/346.0 –40 潘一矿 –0.4~1.3/0.3 6.2 10~146/58 –193 353.0 –33 潘北矿 0.4~0.4/0.4 11.0 22~83/58 –71 77.0~185.0/101.0 –25 潘三矿 0~1.0/0.4 –4.3 0~72/5 –145 –5.8~77.0/55.0 –29 丁集矿 –0.2~1.4/0.7 11.0 0~1 27 26.0 –13 顾桥矿 –0.1~2.4/0.7 3.7 4~5 –8 4.0~5.1/4.6 –8 顾北矿 0.2~2.7/1.2 14.0 0~6/2 2 1.9~2.6/2.2 2 张集矿 0.1~3.4/1.4 11.0 –5~6/1 –102 0.1~4.7/2.4 –7 谢桥矿 0~3.2/1.2 11.0 –16~2/–2 –82 0.8~2.3/1.7 2 注:–0.1~1.7/0.4表示最小~最大值/平均值,其他同。 表 2 水样各参数对于分类的贡献值
Table 2 Contribution rate of each parameter of water samples to classification
项目 贡献值 Cl− 矿井类别 Na++K+ Ca2+ 采样年限 SO4 2− HCO3 − Mg2+ CO3 2− pH 奥灰水 0.20 0.27 0.18 0.20 0.17 0.26 0.14 0.12 0.07 0.07 砂岩水 0.09 0.08 0.05 0.07 0.04 0.04 0.05 0.06 0.01 0.01 松散层水 0.16 0.10 0.23 0.11 0.10 0.10 0.16 0.08 0.05 0.04 太灰水 0.24 0.19 0.19 0.11 0.15 0.11 0.11 0.07 0.04 0.05 MDA 0.15 0.13 0.12 0.10 0.09 0.09 0.09 0.07 0.03 0.03 MDG 115 105 101 74 64 71 76 58 32 30 表 3 随机森林分类结果混淆矩阵
Table 3 Confusion matrix of random forest classification results
项目 奥灰水 砂岩水 太灰水 松散层水 奥灰水 143 5 0 10 砂岩水 2 594 9 21 太灰水 0 13 146 3 松散层水 18 3 2 264 -
[1] 王国法,王虹,任怀伟,等. 智慧煤矿2025情景目标和发展路径[J]. 煤炭学报,2018,43(2):295−305.WANG Guofa,WANG Hong,REN Huaiwei,et al. 2025 scenarios and development path of intelligent coal mine[J]. Journal of China Coal Society,2018,43(2):295−305. [2] 徐静,谭章禄. 智慧矿山系统工程与关键技术探讨[J]. 煤炭科学技术,2014,42(4):79−82.XU Jing,TAN Zhanglu. Smart mine system engineering and discussion of its key technology[J]. Coal Science and Technology,2014,42(4):79−82. [3] 吴立新,殷作如,钟亚平. 再论数字矿山:特征、框架与关键技术[J]. 煤炭学报,2003,28(1):1−7.. doi: 10.3321/j.issn:0253-9993.2003.01.001WU Lixin,YIN Zuoru,ZHONG Yaping. Restudy on digital mine:Characteristics,framework and key technologies[J]. Journal of China Coal Society,2003,28(1):1−7.. doi: 10.3321/j.issn:0253-9993.2003.01.001 [4] 毛善君. “高科技煤矿”信息化建设的战略思考及关键技术[J]. 煤炭学报,2014,39(8):1572−1583.MAO Shanjun. Strategic thinking and key technology of informatization construction of high–tech coal mine[J]. Journal of China Coal Society,2014,39(8):1572−1583. [5] 武强. 我国矿井水防控与资源化利用的研究进展、问题和展望[J]. 煤炭学报,2014,39(5):795−805.WU Qiang. Progress,problems and prospects of prevention and control technology of mine water and reutilization in China[J]. Journal of China Coal Society,2014,39(5):795−805. [6] 储婷婷. 潘谢矿区地下水常规水化学分析及判别模式的建立[D]. 合肥: 中国科学技术大学, 2014.CHU Tingting. Conventional hydrochemical analysis and establishment of discriminant model for groundwater system in Panxie mining area[D]. Hefei: University of Science and Technology of China, 2014. [7] 李垣志,牛国庆,刘慧玲. 改进的GA–BP神经网络在矿井突水水源判别中的应用[J]. 中国安全生产科学技术,2016,12(7):77−81.LI Yuanzhi,NIU Guoqing,LIU Huiling. Application of improved GA–BP neural network on identification of water inrush source in mine[J]. Journal of Safety Science and Technology,2016,12(7):77−81. [8] 徐星,郭兵兵,王公忠. 人工神经网络在矿井多水源识别中的应用[J]. 中国安全生产科学技术,2016,12(1):181−185.XU Xing,GUO Bingbing,WANG Gongzhong. Application of artificial neural network for recognition of multiple water sources in mine[J]. Journal of Safety Science and Technology,2016,12(1):181−185. [9] 杨建,刘基,靳德武,等. 有机–无机联合矿井突水水源判别方法[J]. 煤炭学报,2018,43(10):2886−2894.YANG Jian,LIU Ji,JIN Dewu,et al. Method of determining mine water inrush source based on combination of organic−inorganic water chemistry[J]. Journal of China Coal Society,2018,43(10):2886−2894. [10] 闫鹏程. 基于激光诱导荧光技术的煤矿突水水源识别模型研究[D]. 淮南: 安徽理工大学, 2016.YAN Pengcheng. Research on source determination of mine water inrush based on laser induced fluorescence technology[D]. Huainan: Anhui University of Science and Technology, 2016. [11] 刘剑民,王继仁,刘银朋,等. 基于水化学分析的煤矿矿井突水水源判别[J]. 安全与环境学报,2015,15(1):31−35.LIU Jianmin,WANG Jiren,LIU Yinpeng,et al. Hydrochemistry analysis based on the source determination of coal mine water−bursts[J]. Journal of Safety and Environment,2015,15(1):31−35. [12] 王心义,赵伟,刘小满,等. 基于熵权–模糊可变集理论的煤矿井突水水源识别[J]. 煤炭学报,2017,42(9):2433−2439.WANG Xinyi,ZHAO Wei,LIU Xiaoman,et al. Identification of water inrush source from coalfield based on entropy weight−fuzzy variable set theory[J]. Journal of China Coal Society,2017,42(9):2433−2439. [13] 黄平华,陈建生. 基于多元统计分析的矿井突水水源 Fisher识别及混合模型[J]. 煤炭学报,2011,36(增刊1):131−136.HUANG Pinghua,CHEN Jiansheng. Fisher indentify and mixing model based on multivariate statistical analysis of mine water inrush sources[J]. Journal of China Coal Society,2011,36(Sup.1):131−136. [14] 武强,崔芳鹏,赵苏启,等. 矿井水害类型划分及主要特征分析[J]. 煤炭学报,2013,38(4):561−565.WU Qiang,CUI Fangpeng,ZHAO Suqi,et al. Type classification and main characteristics of mine water disasters[J]. Journal of China Coal Society,2013,38(4):561−565. [15] 张俊,李余生,林曼利,等. 淮南张集矿水文地球化学特征及水源识别[J]. 水文地质工程地质,2014,41(6):32−37.ZHANG Jun,LI Yusheng,LIN Manli,et al. Hydrogeochemistry characteristics and source identification: A case study in the Zhangji Coal Mine in Huainan[J]. Hydrogeology and Engineering Geology,2014,41(6):32−37. [16] 张丹丹. 淮南矿区水化学特征及水源判别模型研究[D]. 淮南: 安徽理工大学, 2017.ZHANG Dandan. Study on water chemistry characteristics and water source discrimination model in Huainan mining area[D]. Huainan: Anhui University of Science and Technology, 2017. [17] 王浪. 采动影响下谢桥煤矿地下水动态变化特征及模拟研究[D]. 合肥: 合肥工业大学, 2016.WANG Lang. The groundwater dynamic change characteristics and simulation studies under the influence of mining in Xieqiao mine[D]. Hefei: Hefei University of Technology, 2016. [18] 刘勇. 基于随机森林分类器的Φ–OTDR传感系统模式识别研究[D]. 北京: 北京交通大学, 2018.LIU Yong. Investigation on pattern recognition based on random forest classifier for phase−sensitive Φ−OTDR sensing system[D]. Beijing: Beijing Jiaotong University, 2018. [19] 王琳琳. 说话人识别中的时变鲁棒性问题研究[D]. 北京: 清华大学, 2013.WANG Linlin. Research on time–varying robustness in speaker recognition[D]. Beijing: Tsinghua University, 2013. [20] 齐瑞. 淮南潘北矿水化学特征分析及突水水源判别模型[D]. 淮南: 安徽理工大学, 2018.QI Rui. Huainan Panbei mine water chemical characteristics analysis and discriminating model of water inrush source[D]. Beijing: Anhui University of Science and Technology, 2018. [21] 孟磊,丁恩杰,吴立新. 基于矿山物联网的矿井突水感知关键技术研究[J]. 煤炭学报,2013,38(8):1397−1403.MENG Lei,DING Enjie,WU Lixin. Research on key technologies of water inrush perception based on mine IoT[J]. Journal of China Coal Society,2013,38(8):1397−1403.