布沼坝露天矿爆破效果预测BP神经网络模型应用BP NEURAL NETWORK PREDICTION MODEL OF BLASTING EFFECT IN BUZHAOBA OPEN MINE
李洪超,刘殿书,李建丰,李鹏,邹志
LI Hong-chao,LIU Dian-shu,LI Jian-feng,LI Peng,ZOU Zhi
摘要(Abstract)：
针对布沼坝露天矿西帮抢险治理工程,进行了35次生产爆破试验,完成了爆破参数的初步优化。同时以孔距、排距、孔深和抵抗线等作为模型的输入因子,大块率、爆堆的前冲距离、爆堆的后冲距离和挖掘机的铲装速率等作为模型的输出因子,以现场爆破试验数据为训练样本,建立爆破效果预测的BP神经网络模型。通过对网络仿真结果和现场实测数据进行比较分析,表明BP神经网络模型能够比较准确地预测出布沼坝露天矿西帮治理工程的爆破效果。
According to west slope emergency control project of Buzhaoba open mine,35 practical blasting tests had been carried out and the optimization of blasting parameters had been accomplished.The BP neural network prediction model was set up to predict blasting effect by taking hole spaces,peripheral hole spaces,hole depth and burden as input factors,boulder yield,forward and backward setting distance,and the rate of excavator shoveling as output factors and data of industrial tests as training sample.By comparing the simulation results with the measured data,it showed that blasting effect could be more accurately predicted by BP neural network model in the west slope control project of Buzhaoba open mine.
关键词(KeyWords)：
BP神经网络模型;爆破试验;爆破效果预测
BP neural network model;Blasting test;Prediction of blasting effect
基金项目(Foundation):
作者(Author):
李洪超,刘殿书,李建丰,李鹏,邹志
LI Hong-chao,LIU Dian-shu,LI Jian-feng,LI Peng,ZOU Zhi
参考文献(References)：
- [1]李洪超.布沼坝露天煤矿西帮抢险治理工程爆破技术研究[D].昆明:昆明理工大学,2010.
- [2]Lv S R,Lv S J.Applying BP Neural Network Model to Forecast Peak Velocity of Blasting Ground Vibration[J].Procedia Engineering,2011(26):257-263.
- [3]孙文彬,刘希亮,谭正龙,等.基于抛掷爆破预测的BP神经网络参数优化方法[J].煤炭学报,2012,37(S1):59-64.
- [4]张戈平.浅谈降低深孔梯段爆破大块率的措施[J].工程爆破,2009,15(2):48-51.
- [5]王万富,刘殿书,黄永辉,等.海拉尔季节性冻土爆破参数试验研究[J].工程爆破,2013,19(5):21-24.
- [6]柳小桐.BP神经网络输入层数据归一化研究[J].机械工程与自动化,2010(3):122-123.
- [7]喻寿益,王吉林,彭晓波.基于神经网络的铜闪速熔炼过程工艺参数预测模型[J].中南大学学报(自然科学版),2007,38(3):523-527.