摘要
针对矿井水害的突水水源判别问题,采用KPCA方法对原始数据降维,通过布谷鸟搜索算法(CS)优化支持向量机(SVM)的惩罚因子C和核参数g,建立基于KPCA-CS-SVM的矿井突水水源判别模型,以淮南新庄孜矿各含水层共45个突水样本数据作为研究对象,选取7个主要影响因素作为突水水源的判别依据,对KPCA-CS-SVM水源判别模型进行学习训练及预测分析,并与KPCA-GA-SVM、KPCA-PSO-SVM模型的判别效果进行对比.研究结果表明:KPCA-CS-SVM模型的突水水源判别结果与实际结果保持一致,且其预测准确率、判别速度、稳定性高于其他2个模型.研究结论从新的角度对突水水源进行判别,有助于矿井突水的灾害预判.
Aiming at the problem of water inrush source identification of mine water disaster,KPCA method is used to reduce the dimension of original data,and cuckoo search algorithm(CS)is used to optimize the penalty factor C and kernel parameter g of support vector machine(SVM),and a mine water inrush source discrimination model based on KPCA-CS-SVM is established.Seven main influencing factors are selected as the basis for the discrimination of water inrush sources.The KPCA-CS-SVM model is used for learning,training and predictive analysis,and compared with the discrimination effect of the KPCA-GA-SVM and KPCA-PSO-SVM models.The results show that the discrimination results of KPCA-CS-SVM model are consistent with the actual results,and the prediction results are consistent.The accuracy,discrimination speed and stability are higher than that of the other two models.The research conclusion is helpful to predict the disaster of mine water inrush by distinguishing the source of water inrush from a new angle.
作者
毛志勇
崔鹏杰
黄春娟
韩榕月
MAO Zhiyong;CUI Pengjie;HUANG Chunjuan;HAN Rongyue(School of Business Administration,Liaoning Technical University,Huludao 125105,China)
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2021年第2期104-111,共8页
Journal of Liaoning Technical University (Natural Science)
基金
国家自然科学基金(71771111)
关键词
矿井突水
水源判别
核主成分分析
布谷鸟算法
支持向量机
mine water inrush
water source discrimination
kernel principal component analysis
cuckoo search
support vector machines
作者简介
毛志勇(1976-),男,陕西汉中人,硕士,副教授,主要从事数据挖掘、信息系统、采矿工程等方面的研究.
