• 论文
主办单位:煤炭科学研究总院有限公司、中国煤炭学会学术期刊工作委员会
基于PCA-RA的滨海矿井水源识别技术研究
  • Title

    Technological yesearch on water source identiftcation of coastal coalmines based on PCA-RA

  • 作者

    陈绍杰 刘久潭汪锋周景奎唐鹏飞高宗军

  • Author

    CHEN Shaojie, LIU Jiutan, WANG Feng, ZHOU Jingkui, TANG Pengfei, GAO Zongjun

  • 单位

    山东科技大学 能源与矿业工程学院山东能源龙口矿业集团 梁家煤矿山东科技大学 地球科学与工程学院

  • Organization
    School of Construction Management, Jiangsu Vocational Institute of Architectural Technology; State Key Laboratory for Geomechanics & Deep Underground Engineering,China University of Mining & Technology
  • 摘要

    煤炭开采过程中矿井水害的发生,严重威胁着煤矿的安全生产,确定水源数量和类型对于矿井水害的防治具有重要意义。为了明确滨海煤矿矿井水补给来源的数量和类型,以龙口梁家煤矿为例,对矿区内不同水体(矿井水、第四系水和地表塌陷区积水)分别进行取样,并基于水化学和主成分分析-残差分析(PCA-RA)进行矿井水源识别。结果表明:梁家煤矿区内第四系水、地表塌陷区积水和矿井水中的主要化学组分含量差别较大,且受到了海水入侵作用的影响。就均值而言,矿井水中的阳阴离子质量浓度分别存在着ρ(Na +)>ρ(Ca 2+)>ρ(Mg 2+)>ρ(K +)和ρ(HCO -3)>ρ(Cl -)>ρ(SO 2-4)>ρ(Cl -)的关系。不同水体中,Cl -和Na +均为优势阴、阳离子,水化学类型以Na-Cl型为主,并且3种水体之间存在着一定的水力联系。选取的水化学数据适合进行PCA,但仅根据特征值大于1或是累计方差贡献率大于85%来确定主成分的数量,并不能很好地表征原始数据的全部信息。基于水化学和PCA-RA方法,确定了梁家煤矿矿井水共有5个补给来源,即海水、富HCO3基岩水、塌陷区积水、混合水和第四系水。PCA-RA法能有效地处理和表征原始水质数据信息,可更加合理地确定矿井水的补给来源类型和数量。研究结果可为滨海煤矿区的水害防治提供科学依据。

  • Abstract

    The occurrence of mine water hazards during coal mining has seriously threatened the safety of coal mines.Determining the quantity and type of water sources is of great significance to the prevention of water hazards.In order to determine the quantity and type of mine water supply sources in coastal coal mine, taking Liangjia Coal Mine in Longkou as an example, different water bodies (mine water, quaternary water and accumulated water in the subsidence area) in the mining area were sampled respectively, and mine water sources were identified based on hydrochemistry and principal component analysis residual analysis (PCA-RA).The results show that the contents of main chemical components in Quaternary water, accumulated water in the subsidence area and mine water in Liangjia Coal Mine area are quite different, and are affected by seawater intrusion.In terms of mean value, the mass concentrations of cations and anions in mine water have the relationships of Na+>Ca2+>Mg2+>K+and HCO-3>Cl->SO2-4>Cl-, respectively.In different water bodies, Cl-and Na+are dominant anions and cations, and the hydrochemical type is mainly Na-Cl type, and there is a certain hydraulic connection among the three water bodies.The selected water chemistry data is suitable for PCA, but the number of principal components is only determined based on the criterion that the characteristic value is greater than 1 or the cumulative variance contribution rate is greater than 85%, which cannot well represent all the information of the original data.Based on the method of water chemistry and PCA-RA, it is determined that there are five recharge sources for mine water in Liangjia Coal Mine, namely seawater, HCO3-rich bedrock water, accumulated water in the subsidence area, mixed water and Quaternary water.The PCA-RA method can effectively process and characterize the information of the original water quality data, and it can be more reasonable to determine the type and quantity of the recharge source of mine water.The research results can provide a certain scientific reference and basis for the prevention of water hazards in the coastal coal mine area.

  • 关键词

    滨海煤矿区主成分分析残差分析水源识别梁家煤矿

  • KeyWords

    coastal coal mining area; principal component analysis; residual analysis; water source identification; Liangjia Coal Mine

  • 基金项目(Foundation)
    国家自然科学基金资助项目(51774194);“泰山学者工程”资助项目(tsqn201812067);山东省自然科学基金重大基础研究资助项目(ZR2018ZC0740)
  • 引用格式
    陈绍杰, 刘久潭,汪锋,等.基于PCA-RA的滨海矿井水源识别技术研究[J].煤炭科学技术,2021,49(2):217-225.doi:10.13199/j.cnki.cst.2021.02.025
    CHEN Shaojie,LIU Jiutan,WANG Feng,et al.Water source identification of coastal coal mine based on PCA-RA:a case study of Liangjia Coal Mine, Longkou[J].Coal Science and Technology,2021,49(2):217-225.doi:10.13199/j.cnki.cst.2021.02.025
  • 相关专题
相关问题

主办单位:煤炭科学研究总院有限公司 中国煤炭学会学术期刊工作委员会

©版权所有2015 煤炭科学研究总院有限公司 地址:北京市朝阳区和平里青年沟东路煤炭大厦 邮编:100013
京ICP备05086979号-16  技术支持:云智互联