工程爆破

2020, v.26;No.118(06) 1-8

[打印本页] [关闭]
本期目录(Current Issue) | 过刊浏览(Past Issue) | 高级检索(Advanced Search)

基于PSO-LSSVM模型的露天矿爆破振动效应预测
Prediction of blasting vibration effect in open-pit mine based on PSO-LSSVM model

岳中文;吴羽霄;魏正;王贵;王渊;李鑫;周奕硕;
YUE Zhong-wen;WU Yu-xiao;WEI Zheng;WANG Gui;WANG Yuan;LI Xin;ZHOU Yi-shuo;College of Mechanics and Architectural Engineering,China University of Mining and Technology(Beijing);Inner Mongolia Kinergy Blasting Co., Ltd.;

摘要(Abstract):

为寻找一种适合露天矿爆破振动效应的预测模型,结合内蒙古康宁爆破公司德隆和华武煤矿爆破工程的50组实测数据,建立最小二乘支持向量机(LSSVM)模型分别对爆破峰值振动速度v和振动主频f两个评价爆破振动效应的指标进行预测,并结合粒子群算法(PSO)寻优获得LSSVM中正则化参数γ和核函数宽度系数σ的最佳参数组合。结果表明:PSO-LSSVM模型对爆破峰值振动速度和振动主频预测的MRE为3.31%和6.38%,RMSE为0.98%和2.02%。与BP神经网络、LSSVM模型对比,此模型具有更好的泛化能力和更高的预测精度,为多因素影响下爆破振动效应预测提供了一种新的思路。
In order to find a suitable model for predicting the blasting vibration effect of open-pit mines, the least square support vector machine(LSSVM) model is established to predict the blasting peak vibration velocity and blasting vibration main frequency which evaluate the blasting vibration index combining with 50 sets of measured data of the blasting engineering of Delong and Huawu Coal Mine of Inner Mongolia Corning Blasting Company. In addition, particle swarm optimization(PSO) is used to optimize the optimal combination of regularization parameters γ and kernel function width coefficients σ in LSSVM. The results show that the average relative error of the PSO-LSSVM model to predict the blasting vibration velocity and blasting vibration main frequency are 3.31% and 6.38%, and the mean square error is 0.98% and 2.02%. Compared with BP neural network model and LSSVM model, PSO-LSSVM model has better generalization ability and higher prediction accuracy. The research provides a new idea for the prediction of blasting vibration effect under the influence of multiple factors.

关键词(KeyWords): 露天矿爆破;最小二乘支持向量机;粒子群算法;爆破振动效应;惯性权重因子
open-pit blasting;least squares support vector machine(LSSVM);particle swarm optimization(PSO);blast vibration effect;inertia weight factor

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金资助项目(51974318)

作者(Authors): 岳中文;吴羽霄;魏正;王贵;王渊;李鑫;周奕硕;
YUE Zhong-wen;WU Yu-xiao;WEI Zheng;WANG Gui;WANG Yuan;LI Xin;ZHOU Yi-shuo;College of Mechanics and Architectural Engineering,China University of Mining and Technology(Beijing);Inner Mongolia Kinergy Blasting Co., Ltd.;

参考文献(References):

扩展功能
本文信息
服务与反馈
本文关键词相关文章
本文作者相关文章
中国知网
分享