工程爆破

2020, v.26(05) 30-35

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

PCA-BP算法在地面爆破振动中的应用
Application of PCA-BP algorithm in ground blasting vibration

赵红梦;姜志侠;
ZHAO Hong-meng;JIANG Zhi-xia;School of Science, Changchun University of Science and Technology;

摘要(Abstract):

为了更加准确地预测地面爆破的质点峰值振动速度,提出应用一种PCA-BP算法,该算法首先利用主成分分析对爆心距、高程差、总药量、炮孔深度、单段最大药量等地面爆破振动影响因素进行研究,然后结合BP神经网络算法对其爆破质点峰值振动速度进行预测。结果显示:利用PCA-BP算法的预测结果更接近工程实测值,平均相对误差为7.748%,远小于用传统萨道夫斯基经验公式进行预测的平均相对误差32.654%,说明将PCA-BP算法应用到爆破振动工作中是比较可行的,对评估地面振动危害有一定的指导意义。
In order to more accurately predict the peak particle velocity of ground blasting vibration, a PCA-BP algorithm was proposed; The algorithm firstly uses principal component analysis to analyze the impact factors of ground blasting vibration,such as blast center distance, elevation difference, total charge, blasthole depth and the maximum charge of the segment, and then combined with the BP neural network algorithm to predict the peak particle velocity of the blasting particles.The results show that the prediction result byusing the PCA-BP algorithm is closer to the measured value of the project,and the average relative error is 7.748%, which is less than the average relative error of 32.654% predicted by the traditional Sadovsky empirical formula. It shows that it is more feasible to apply the PCA-BP algorithm to blasting vibration work, and there are certain guiding significance for evaluating the ground vibration hazards.

关键词(KeyWords): PCA-BP算法;BP神经网络;振速预测;质点峰值振动速度;主成分分析
PCA-BP algorithm;BP neural network;vibration velocity prediction;peak particle velocity;principal component analysis

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金资助项目(51378076)

作者(Author): 赵红梦;姜志侠;
ZHAO Hong-meng;JIANG Zhi-xia;School of Science, Changchun University of Science and Technology;

Email:

DOI:

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