基于抛掷爆破预测的BP神经网络参数优化方法
|
摘要
研究了不同隐含节点个数、训练函数、转移函数、学习率对BP神经网络预测结果准确性和收敛速度的影响;对比分析了参数优化后的BP模型与径向基神经网络、支持向量机预测模型的结果;并应用某露天矿抛掷爆破作业的实测数据进行了相关实验。实验结果表明:最优BP模型的拓扑结构为10-6-3;最佳的训练函数为LM函数,正切和线性函数的组合为最优的转移函数,最佳的网络学习率为0.77;参数优化后BP模型的最远抛掷距离、抛掷率、松散系数的预测结果与测试样本的标准差最小,分别为9.567 8,0.036 3,0.041 4,即参数优化后的BP模型预测结果最优。
A method of optimizing parameters of BP cast blasting model was described.The accuracy of predicting result and BP model convergence rate with different numbers of hidden layers,different training functions and transfer functions,different learning rates were analyzed particularly.The forecast result of parameter optimization model was contrasted with those of RBF(Radial-Basis Function),SVM(Support Vector Machine) models.The experiment was done by using a lot of surveying data of some open pit cast blasting.The result indicates that the optimizing BP topology structure is 10-6-3(10 input variables,6 hidden layers,3 output variables);LM(Levenberg-Marquart) algorithm is the best training function;tansig and purelin functions combination is the best transfer function;the best learning rate is 0.77;test sample simulation standard deviation of cast blasting ratio,max cast distance,coefficient of volumetric expansion based on parameter optimization BP model is 9.567 8,0.036 3,0.041 4 respectively and is the smallest contrasting with RBF and SVM;the prediction result of parameter optimization BP model is the best one.
A method of optimizing parameters of BP cast blasting model was described.The accuracy of predicting result and BP model convergence rate with different numbers of hidden layers,different training functions and transfer functions,different learning rates were analyzed particularly.The forecast result of parameter optimization model was contrasted with those of RBF(Radial-Basis Function),SVM(Support Vector Machine) models.The experiment was done by using a lot of surveying data of some open pit cast blasting.The result indicates that the optimizing BP topology structure is 10-6-3(10 input variables,6 hidden layers,3 output variables);LM(Levenberg-Marquart) algorithm is the best training function;tansig and purelin functions combination is the best transfer function;the best learning rate is 0.77;test sample simulation standard deviation of cast blasting ratio,max cast distance,coefficient of volumetric expansion based on parameter optimization BP model is 9.567 8,0.036 3,0.041 4 respectively and is the smallest contrasting with RBF and SVM;the prediction result of parameter optimization BP model is the best one.
引文
[1]李祥龙,何丽华,栾龙发,等.露天矿煤矿高台阶抛掷爆破爆堆形态模拟[J].煤炭学报,2011,36(9):1457-1462.Li Xianglong,He Lihua,Luan Longfa,et al.Simulation model formuckpile shape of high bench cast blasting in surface coal mine[J].Journal of China Coal Society,2011,36(9):1457-1462.
[2]傅洪贤,李克民.露天煤矿高台阶抛掷爆破参数分析[J].煤炭学报,2006,31(4):442-445.Fu Hongxian,Li Kemin.Analysis of high bench castblasting parame-ters in surface coal mines[J].Journal of China Coal Society,2006,31(4):442-445.
[3]马军,郭昭华.露天煤矿应用抛掷爆破技术研究[J].露天采矿技术,2005(5):43-45.Ma Jun,Guo ZhaoHua.Study of using cast blasting in surface coalmine[J].Open Cast Mining Technology,2005(5):43-45.
[4]潘井澜.北美露天煤矿开采中抛掷爆破法的应用[J].世界煤炭技术,1993(10):19-22.Pan Jinglan.Cast blasting used in North America’s surface coalmine[J].World Coal Mine Technology,1993(10):19-22.
[5]徐全军,张庆明,恽寿榕.爆破地震峰值的神经网络预报模型[J].北京理工大学学报,1998,18(4):472-475.Xu Quanjun,Zhang Qingmin,Yun Shourong.Neural network modelfor forecasting peak acceleration of blasting vibration[J].Journal ofBeijing Institute of Technology,1998,18(4):472-475.
[6]张艺峰,姚道平,谢志招.基于BP神经网络的爆破振动峰值及主频预测[J].工程地球物理学报,2008,5(2):222-226.Zhang Yifeng,Yao Daoping,Xie Zhizhao.The prediction of blastingvibration peak value&main frequency by BP neural network[J].Chinese Journal of Engineering Geophysics,2008,5(2):222-226.
[7]李祥龙.高台阶抛掷爆破技术与效果预测模型研究[D].北京:中国矿业大学(北京),2010:31-52.Li Xianglong.Predicting the effect of high bench cast blasting[D].Beijing:China University of Mining and Technology(Beijing),2011:31-52.
[8]姚金阶,朱以文.岩体爆破参数设计的神经网络模型[J].爆破,2005,22(2):34-36.Yao Jinjie,Zhu Yiwen.The neural network model of rock blastingparameters design[J].Blasting,2005,22(2):34-36.
[9]韩亮.黑岱沟露天煤矿高台阶抛掷爆破爆堆形态的研究[D].北京:中国矿业大学(北京),2010:20-45.Han Liang.Analysis of shape of high bench cast blasting of Heidai-gou in surface coal mine[D].Beijing:China University of Miningand Technology(Beijing),2010:20-45.
[10]郭连军,王智静,牛成俊.爆破优化的神经网络模型[J].工程爆破,1996,2(2):11-15.Guo Lianjun,Wang Zhijing,Niu Chengjun.The neural networkmodel for blasting optimization[J].Engineering Blasting,1996,2(2):11-15.
[11]张成良,杨阳,梁开水,等.岩壁梁爆破参数优化的神经网络模型[J].工程爆破,2006,12(1):22-25.Zhang Chengliang,Yang Yang,Liang Kaishui,et al.Neural networkmodel of blasting parameters optimization of crane beam at rockwall[J].Engineering Blasting,2006,12(1):22-25.
[12]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),2011:10-45.Liu Xiliang.The cast blasting effort forecast and analysis based onGA-BP neural network[D].Beijing:China University of Miningand Technology(Beijing),2011:10-45.
[13]Galaviz P,Melin P,Trujill L.Improvement of the back propagationalgorithm using(1+1)evolutionary strategies[J].Studies in Com-putational Intelligence,2010,312:287-302.
[14]蔡自兴,徐光佑.人工智能及其应用(第四版)[M].北京:清华大学出版社,2010:40-70.Cai Zixing,Xu Guangyou.Artificial intelligence:principles&appli-cations[M].Beijing:Tsinghua University Press,2010:40-70.
[15]张德丰.Matlab神经网络应用设计[M].北京:机械工业出版社,2009:121-130.Zhang Defeng.Matlab neural network application[M].Beijing:China Machine Press,2009:121-130.
[2]傅洪贤,李克民.露天煤矿高台阶抛掷爆破参数分析[J].煤炭学报,2006,31(4):442-445.Fu Hongxian,Li Kemin.Analysis of high bench castblasting parame-ters in surface coal mines[J].Journal of China Coal Society,2006,31(4):442-445.
[3]马军,郭昭华.露天煤矿应用抛掷爆破技术研究[J].露天采矿技术,2005(5):43-45.Ma Jun,Guo ZhaoHua.Study of using cast blasting in surface coalmine[J].Open Cast Mining Technology,2005(5):43-45.
[4]潘井澜.北美露天煤矿开采中抛掷爆破法的应用[J].世界煤炭技术,1993(10):19-22.Pan Jinglan.Cast blasting used in North America’s surface coalmine[J].World Coal Mine Technology,1993(10):19-22.
[5]徐全军,张庆明,恽寿榕.爆破地震峰值的神经网络预报模型[J].北京理工大学学报,1998,18(4):472-475.Xu Quanjun,Zhang Qingmin,Yun Shourong.Neural network modelfor forecasting peak acceleration of blasting vibration[J].Journal ofBeijing Institute of Technology,1998,18(4):472-475.
[6]张艺峰,姚道平,谢志招.基于BP神经网络的爆破振动峰值及主频预测[J].工程地球物理学报,2008,5(2):222-226.Zhang Yifeng,Yao Daoping,Xie Zhizhao.The prediction of blastingvibration peak value&main frequency by BP neural network[J].Chinese Journal of Engineering Geophysics,2008,5(2):222-226.
[7]李祥龙.高台阶抛掷爆破技术与效果预测模型研究[D].北京:中国矿业大学(北京),2010:31-52.Li Xianglong.Predicting the effect of high bench cast blasting[D].Beijing:China University of Mining and Technology(Beijing),2011:31-52.
[8]姚金阶,朱以文.岩体爆破参数设计的神经网络模型[J].爆破,2005,22(2):34-36.Yao Jinjie,Zhu Yiwen.The neural network model of rock blastingparameters design[J].Blasting,2005,22(2):34-36.
[9]韩亮.黑岱沟露天煤矿高台阶抛掷爆破爆堆形态的研究[D].北京:中国矿业大学(北京),2010:20-45.Han Liang.Analysis of shape of high bench cast blasting of Heidai-gou in surface coal mine[D].Beijing:China University of Miningand Technology(Beijing),2010:20-45.
[10]郭连军,王智静,牛成俊.爆破优化的神经网络模型[J].工程爆破,1996,2(2):11-15.Guo Lianjun,Wang Zhijing,Niu Chengjun.The neural networkmodel for blasting optimization[J].Engineering Blasting,1996,2(2):11-15.
[11]张成良,杨阳,梁开水,等.岩壁梁爆破参数优化的神经网络模型[J].工程爆破,2006,12(1):22-25.Zhang Chengliang,Yang Yang,Liang Kaishui,et al.Neural networkmodel of blasting parameters optimization of crane beam at rockwall[J].Engineering Blasting,2006,12(1):22-25.
[12]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),2011:10-45.Liu Xiliang.The cast blasting effort forecast and analysis based onGA-BP neural network[D].Beijing:China University of Miningand Technology(Beijing),2011:10-45.
[13]Galaviz P,Melin P,Trujill L.Improvement of the back propagationalgorithm using(1+1)evolutionary strategies[J].Studies in Com-putational Intelligence,2010,312:287-302.
[14]蔡自兴,徐光佑.人工智能及其应用(第四版)[M].北京:清华大学出版社,2010:40-70.Cai Zixing,Xu Guangyou.Artificial intelligence:principles&appli-cations[M].Beijing:Tsinghua University Press,2010:40-70.
[15]张德丰.Matlab神经网络应用设计[M].北京:机械工业出版社,2009:121-130.Zhang Defeng.Matlab neural network application[M].Beijing:China Machine Press,2009:121-130.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心