温廷新;孔祥博;

• 1:辽宁工程技术大学系统工程研究所

摘要(Abstract)：

为对矿井突水水源进行识别以减少矿井突水事故的发生,提出了粒子群(PSO)结合RBF核参数优化的SVM模型,并使用核主成分分析法(KPCA)对选取水源特征指标进行高效降维.根据水源离子敏感性选取了8种水化学指标(K⁺、Na+、Na⁺、Mg+、Mg⁽²⁺⁾、Ga(2+)、Ga⁽²⁺⁾、HCO_3(2+)、HCO_3^-、Cl-、Cl^-、F-、F^-、SO_4-、SO_4^(2-))作为突水水源识别特征参数.使用基于最大方差关联度准则的核参数选择方法并结合粒子群算法构造参数优化算法,使用参数优选后的支持向量机模型对90组突水水源识别训练数据进行模型训练,用其余32组数据进行测试,模型实测效果与Logistic模型、PCA-Fisher模型以及PSO-SVM模型进行对比,结果表明:采用径向基核函数优化的支持向量机模型能够选取较优参数,模型实测平均准确率为93.75%,误差明显低于其他模型,证明了该模型能精准且高效地识别矿井突水水源.

关键词(KeyWords)： 矿井突水;支持向量机;参数优化;径向基核函数;粒子群算法

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金（71371091）

作者(Authors): 温廷新;孔祥博;

参考文献(References)：

• [1]杨建,刘基,靳德武,王强民.有机-无机联合矿井突水水源判别方法[J].煤炭学报,2018,43(10):2 886-2 894.YANG Jian,LIU Ji,JIN Dewu,WANG Qiangmin.Method of determining mine water inrush source based on combination of organic-inorganic water chemistry[J].Journal of China Coal Society,2018,43(10):2 886-2 894.
• [2]李凤莲,冯琳,张雪英,等.模糊综合评判法的改进及在水源判别中的应用[J].太原理工大学学报,2015,46(4):444-447.LI Fenglian,FENG Lin,ZHANG Xueying,et al.Improvement of the fuzzy comprehensive evaluation and its application in water-bursting source discrimination[J].Journal of Taiyuan University of Technology,2015,46(4):444-447.
• [3]鲁金涛,李夕兵,宫凤强,等.基于主成分分析与Fisher判别分析法的矿井突水水源识别方法[J].中国安全科学学报,2012,22(7):109-115.LU Jintao,LI Xibing,GONG Fengqiang,et al.Recognizing of mine water inrush sources based on principal components analysis and fisher discrimination analysis method[J].China Safety Science Journal,2012,22(7):109-115.
• [4]连会青,刘德民,尹尚先.水化学综合识别模式在矿井水源判别中的应用[J].煤炭工程,2012(8):107-109+113.LIAN Huiqing,LIU Demin,YIN Shangxian.Application of hydrochemistry comprehensive identification mode to distinguish mine water resources[J].Coal Engineering,2012(8):107-109+113.
• [5]张帝,孟磊,董飞,刘晓文,等.矿井突水水源识别的GA-SVM方法研究[J].煤炭技术,2018,37(4):144-147.ZHANG Di,MENG Lei,DONG Fei,LIU Xiaowen,etc.Study on GA-SVM for mine water inrush source identification[J].Coal Technology,2018,37(4):144-147.
• [6]成荣秋,吴燕清.基于KPCA-FDA方法的矿井突水水源判别研究[J].煤炭技术,2018,37(11):194-195.CHENG Rongqiu,WU Yanqing.Identifying of mine water inrush sources based on KPCA-FDA method[J].Coal Technology,2018,37(11):194-195.
• [7]冯亚娟,崔宁,王丹.矿井突水水源Logistic识别及混合模型[J].辽宁工程技术大学学报(自然科学版),2015,34(11):1 228-1 233.FENG Yajuan,CUI Ning,WANG Dan.Logistic identify and mixing model of mine water inrush sources[J].Journal of Liaoning Technical University(Natural Science),2015,34(11):1 228-1 233.
• [8]邵良杉,李相辰.基于MIV-PSO-SVM模型的矿井突水水源识别[J].煤炭科学技术,2018,46(8):183-190.SHAO Liangshan,LI Xiangchen.Indentification of mine water inrush source based on MIV-PSO-SVM[J].Coal Science and Technology,2018,46(8):183-190.
• [9]周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-246.ZHOU Shaolei,LIAO Jian,SHI Xianjun.Kernel parameter selection of RBF-SVM and its application in fault diagnosis[J].Journal of Electronic Measurement and Instrumentation,2014,28(3):240-246.
• [10]温廷新,孙雪,孔祥博,等.基于PSOBP-Ada Boost模型的瓦斯涌出量分源预测研究[J].中国安全科学学报,2016,26(5):94-98.WEN Tingxin,SUN Xue,KONG Xiangbo,etc.Research on prediction of gas emission quantity with sub sources basing on PSOBP-Ada Boost[J].China Safety Science Journal,2016,26(5):94-98.
• [11]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(S1):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(S1):131-136.
• [12]白越,王经明.微震监测技术在煤矿突水监测中的应用[J].辽宁工程技术大学学报(自然科学版),2010,29(4):549-552.BAI Yue,WANG Jingming.Application of monitoring techniques of micro-seismic to water inrush forecast in coal mines[J].Journal of Liaoning Technical University(Natural Science),2010,29(4):549-552.