岳中文,范皓宇,马鑫民

• 1:中国矿业大学(北京)力学与建筑工程学院

摘要(Abstract)：

煤矿巷道爆破效果受很多因素的影响,传统单一模式下的人工智能方法对煤矿巷道爆破效果预测不佳。因此以提高模型的预测精度为目的,通过建立PSO-SVM模型进行关键参数寻优。计算得出惩罚参数c与核函数参数g分别为14.0046和1.3622。以我国煤矿岩石巷道爆破工程为背景,从现场调研收集到的100余条巷道爆破工程实例中选取42组典型案例作为训练和测试样本,分别在RBF核函数的基础上应用传统SVM、Grid search-SVM和PSO-SVM模型对炮孔利用率进行预测对比,得到3种预测结果的准确率分别为66.67%、75%、91.67%;同时进一步验证了在PSO-SVM模型中4种不同核函数的预测准确率。结果表明:PSO-SVM模型中选取RBF核函数所得到的准确率最高,精度能够满足工程实际需求。

关键词(KeyWords)： 爆破参数;炮孔利用率;支持向量机;粒子群算法

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划专项资助(2016YFC0600903);; 高等学校学科创新引智计划项目(B14006)

作者(Author): 岳中文,范皓宇,马鑫民

参考文献(References)：

• [1] 杨仁树,马鑫民,李清,等.煤矿巷道掘进爆破智能设计系统及应用[J].煤炭学报,2013,38(7):1130-1135.[1] YANG Ren-shu,MA Xin-min,LI Qing,et al.Application on intelligent system for optimization design of blasting in mine tunnel excavation of coal mine[J].Journal of China Coal Society,2013,38(7):1130-1135.(in Chinese)
• [2] 林天舒.基于支持向量机煤矿巷道爆破参数智能优化关键技术研究[D].北京:中国矿业大学(北京),2015.[2] LIN Tian-shu.Research on key technologies of intelligent optimization of blasting parameters of coal mine roadway based on support vector machine[D].Beijing:China University of Mining and Technology (Beijing),2015.(in Chinese)
• [3] RENNIE Kaunda.New artificial neural networks for true triaxial stress state analysis and demonstration of intermediate principal stress effects on intact rock strength[J].Journa lof Rock Mechanics and Geotechnical Engineering,2014(6):338-347.
• [4] M Lefik.Some aspects of application of artificial neural network for numerical modeling in civil engineering[J].Building of the Polish Academy of Sciences Technical Sciences,2013,61(1):39-50.
• [5] 温廷新,于凤娥,邵良杉.基于灰色关联熵的煤与瓦斯突出概率神经网络预测模型[J].计算机应用研究,2018,35(11):3326-3329.[5] WEN Ting-xin,YU Feng-e,SHAO Liang-shan.Probabilistic neural network prediction model of coal and gas outburst based on grey relational entropy[J].Application Research of Computers,2018,35(11):3326-3329.(in Chinese)
• [6] 刘希亮.基于GA-BP神经网络抛掷爆破效果预测与分析[D].北京:中国矿业大学(北京),2011.[6] LIU Xi-liang.Prediction and analysis of throwing blasting effect based on GA-BP neural network[D].Beijing:China University of Mining and Technology(Beijing),2011.(in Chinese)
• [7] 赵国彦,孙贵东,戴兵,等.基于PSO改进的BP网络在爆破大块率优化中的应用[J].爆破,2017,34(2):15-19.[7] ZHAO Guo-yan,SUN Gui-dong,DAI Bing,et al.Application of BP network based on PSO improved in optimization of blasting boulder yield[J].Blasting,2017,34(2):15-19.(in Chinese)
• [8] 王宇涛,刘殿书,梁书锋,等.立井爆破效果预测及模糊综合评价[J].爆破,2014,31(3):10-14.[8] WANG Yu-tao,LIU Dian-shu,LIANG Shu-feng,et al.Vertical shaft blasting effect prediction and fuzzy comprehensive evaluation[J].Blasting,2014,31(3):10-14.(in Chinese)
• [9] 唐跃,徐曲,柯波,等.基于交叉验证的矿岩爆破块度SVM模型优选研究[J].爆破,2018,35(3):74-79.[9] TANG Yue,XU-Qu,KE Bo,et al.Study on optimization of SVM model of rock blasting fragmentation based on cross-validation[J].Blasting,2018,35(3):74-79.(in Chinese)
• [10] 孙苗,吴力,袁青,等.基于PSO-LSSVM算法的地下储气库洞室群爆破振动预测[J].爆破,2017,34(3):145-150.[10] SUN Miao,WU Li,YUAN Qing,et al.Blasting vibration prediction of underground gas storage caverns based on PSO-LSSVM algorithm[J].Blasting,2017,34(3):145-150.(in Chinese)