基于PSO-ELM的爆破块度预测研究
|
摘要
针对影响爆破块度因素之间的复杂非线性关系,利用粒子群算法(PSO)优化ELM(极限学习机)的输入权值与隐含层阈值,建立PSO-ELM爆破块度预测模型。以别斯库都克露天煤矿的实测数据为例,选取岩石抗拉强度、岩石抗压强度、炮孔间距、排距、最小抵抗线、超深、炸药单耗7个因素作为预测模型的输入因子,选取爆破块度的平均尺寸X50作为预测模型的输出因子。结果表明:PSO-ELM模型预测值与实测值的平均相对误差为5.6%,优于ELM模型的11.4%,具有更好的预测精确度;经PSO优化后的ELM模型,受隐含层节点数影响降低,稳定性增加,PSO-ELM模型更适用于爆破块度的预测。
Aimed at complicated nonlinear relationship among factors affecting blasting fragmentation,the PSO-ELM blasting fragmentation prediction model was built by particle swarm optimization(PSO)to optimize input weights and hidden layer nodes of ELM(extreme learning machine).Taking the measured data of Biesikuduke open-pit coal mine as an example,the tensile strength of rock,the compressive strength of rock,the hole distance,the row distance,the minimum resistance line,the super-deep and the specific charge were selected as the input factors of the prediction model,and the average size of blasting fragmentation was selected as the output factor of the prediction model.The results showed that the average relative error between the measured and predicted values of the PSO-ELM model was 5.6%,which was better than that of 11.4% between the measured and predicted values of ELM model,and had better prediction accuracy.The influence of number of hidden layer nodes on the ELM model optimized by PSO was reduced,and the stability of this model was increased.Therefore,the PSOELM model was more suitable for the blasting fragmentation prediction.
Aimed at complicated nonlinear relationship among factors affecting blasting fragmentation,the PSO-ELM blasting fragmentation prediction model was built by particle swarm optimization(PSO)to optimize input weights and hidden layer nodes of ELM(extreme learning machine).Taking the measured data of Biesikuduke open-pit coal mine as an example,the tensile strength of rock,the compressive strength of rock,the hole distance,the row distance,the minimum resistance line,the super-deep and the specific charge were selected as the input factors of the prediction model,and the average size of blasting fragmentation was selected as the output factor of the prediction model.The results showed that the average relative error between the measured and predicted values of the PSO-ELM model was 5.6%,which was better than that of 11.4% between the measured and predicted values of ELM model,and had better prediction accuracy.The influence of number of hidden layer nodes on the ELM model optimized by PSO was reduced,and the stability of this model was increased.Therefore,the PSOELM model was more suitable for the blasting fragmentation prediction.
引文
[1]汪雪清,单仁亮.人工神经网络在爆破块度预测中的应用研究[J].岩土力学,2008(11):529-531.
[2]史秀志,王洋,黄丹,史采星.基于LS-SVR岩石爆破块度预测[J].爆破,2016,33(3):36-40.
[3]Shi X Z,Zhou J,Wu B B,et al.Support vector machines approach to mean particle size of rock fragmentation due to bench blasting prediction[J].Transactions of Nonferrous Metals Society of China,2012,22(2):432-441.
[4]史秀志,郭霆,尚雪义,姬露露.基于PCA-BP神经网络的岩石爆破平均粒径预测[J].爆破,2016,33(2):55-61.
[5]祝文化,朱瑞赓,夏元友.爆破块度预测的神经网络方法研究[J].武汉理工大学学报,2001(1):60-62.
[6]段宝福,费鸿禄.神经网络模型在台阶爆破块度预测中的应用[J].工程爆破,1999,5(4):25-29.
[7]郝全明,杨振增.BP神经网络在岩层爆破参数优化中的应用[J].煤炭技术,2014,33(12):20-22.
[8]杨成祥,冯夏庭.支持向量回归在岩土工程中的应用[J].东北大学学报(自然科学版),2006,27(5):563-566.
[9]聂军,史秀志,陈新,等.基于GEP的露天台阶爆破块度预测模型[J].爆破,2015(2):82-88.
[10]Huang G B,Zhu Q Y,Siew C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[11]皮骏,马圣,贺嘉诚,等.基于IGA-ELM网络的滚动轴承故障诊断[J/OL].航空学报:1-13[2018-07-16].
[12]Kennedy J,Eberhart R,Shi Y H.Swarm Intelligence[M].San Francisco,Cal,USA:Morgan Kaufmann Publishers,2001:287-318.
[13]李爱国,覃征,鲍复民,等.粒子群优化算法[J].计算机工程与应用,2002,38(21):1-3.
[2]史秀志,王洋,黄丹,史采星.基于LS-SVR岩石爆破块度预测[J].爆破,2016,33(3):36-40.
[3]Shi X Z,Zhou J,Wu B B,et al.Support vector machines approach to mean particle size of rock fragmentation due to bench blasting prediction[J].Transactions of Nonferrous Metals Society of China,2012,22(2):432-441.
[4]史秀志,郭霆,尚雪义,姬露露.基于PCA-BP神经网络的岩石爆破平均粒径预测[J].爆破,2016,33(2):55-61.
[5]祝文化,朱瑞赓,夏元友.爆破块度预测的神经网络方法研究[J].武汉理工大学学报,2001(1):60-62.
[6]段宝福,费鸿禄.神经网络模型在台阶爆破块度预测中的应用[J].工程爆破,1999,5(4):25-29.
[7]郝全明,杨振增.BP神经网络在岩层爆破参数优化中的应用[J].煤炭技术,2014,33(12):20-22.
[8]杨成祥,冯夏庭.支持向量回归在岩土工程中的应用[J].东北大学学报(自然科学版),2006,27(5):563-566.
[9]聂军,史秀志,陈新,等.基于GEP的露天台阶爆破块度预测模型[J].爆破,2015(2):82-88.
[10]Huang G B,Zhu Q Y,Siew C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[11]皮骏,马圣,贺嘉诚,等.基于IGA-ELM网络的滚动轴承故障诊断[J/OL].航空学报:1-13[2018-07-16].
[12]Kennedy J,Eberhart R,Shi Y H.Swarm Intelligence[M].San Francisco,Cal,USA:Morgan Kaufmann Publishers,2001:287-318.
[13]李爱国,覃征,鲍复民,等.粒子群优化算法[J].计算机工程与应用,2002,38(21):1-3.