爆破震动速度峰值的预测
|
摘要
结合在重庆测量的42组爆破数据,针对规程里所用的波速预测经验公式,采用经验公式法、基函数回归方法以及神经网络法对爆破地震效应进行了预测。基函数回归预测法要比经验公式预测法好,比经验公式迭代法稍差,但基函数回归法的使用要方便些。人工神经网络可用于爆破地震波的三向速度峰值预测,从检验样本值与预测结果值之间的相对误差可以看出,人工神经网络预测法的精度要高于基函数回归和经验公式法。同时,对于需要考虑影响震动强度多因素变量的情况,在神经网络中通过修改输入参量即可解决爆破多参量的问题。为爆破地震效应的预测及推广应用提供了有效途径。
Engineering blasting produces a strong ground vibration,it may affect damage of a structure,it's necessary to predict the intensity of vibration.Based on 42 groups of blasting data,the three different methods including the empirical formula method,the basis function regression method and the neural network method were presented to forecast the intensity of vibration.The basis function regression method was more effective than the empirical formula method,the neural network method could predict three-direction velocity peak values of blasting wave.It was found that the precision of the neural network method was the best;the neural network method could solve a multi-parameter problem by modifying the input variable;it provided an efficient path to predict the intensity of vibration.
Engineering blasting produces a strong ground vibration,it may affect damage of a structure,it's necessary to predict the intensity of vibration.Based on 42 groups of blasting data,the three different methods including the empirical formula method,the basis function regression method and the neural network method were presented to forecast the intensity of vibration.The basis function regression method was more effective than the empirical formula method,the neural network method could predict three-direction velocity peak values of blasting wave.It was found that the precision of the neural network method was the best;the neural network method could solve a multi-parameter problem by modifying the input variable;it provided an efficient path to predict the intensity of vibration.
引文
[1]陈士海.爆破灾害预测与控制[M].北京:煤炭工业出版社,2006.
[2]刘军,吴从师.建筑结构对爆破震动的响应预测[J].爆炸与冲击,2000,(5):20-23.
[3]Hong Hao,Guo-wei Ma,Yong Lu.Damage assessment of ma-sonry infilled RC frames subjucted to blasting induced groundexcitations[J].Engineering Structures,2002,(24):799-209.
[4]Chengqing Wu,Hong Hao.Numerical simulation of structuralresponses on a sand layer to blast induced ground excitations[J].Computers&Structure,2004,(5):722-739.
[5]崔波.框架结构对爆破地震效应的预测与损伤评估[D].重庆:重庆大学,2008.
[6]徐赵东,沈亚鹏.神经网络对结构地震反应的预测及试验研究[J].振动与冲击,2003,22(2):22-25.
[7]陈德志,朱瑞賡.基于BP神经网络的路堑爆破对邻近民房安全预测的研究[J].岩石力学与工程学报,2002,(2):51-54.
[2]刘军,吴从师.建筑结构对爆破震动的响应预测[J].爆炸与冲击,2000,(5):20-23.
[3]Hong Hao,Guo-wei Ma,Yong Lu.Damage assessment of ma-sonry infilled RC frames subjucted to blasting induced groundexcitations[J].Engineering Structures,2002,(24):799-209.
[4]Chengqing Wu,Hong Hao.Numerical simulation of structuralresponses on a sand layer to blast induced ground excitations[J].Computers&Structure,2004,(5):722-739.
[5]崔波.框架结构对爆破地震效应的预测与损伤评估[D].重庆:重庆大学,2008.
[6]徐赵东,沈亚鹏.神经网络对结构地震反应的预测及试验研究[J].振动与冲击,2003,22(2):22-25.
[7]陈德志,朱瑞賡.基于BP神经网络的路堑爆破对邻近民房安全预测的研究[J].岩石力学与工程学报,2002,(2):51-54.
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