摘要
在介绍RBF神经网络基本思想的基础上,建立了爆破振动预测模型,用RBF神经网络方法对质点振幅、主振频率及振动持续时间进行预测。用阳泉煤矿主井爆破开挖工程中所监测到的振动数据对模型进行了训练,并对27组数据进行了预测,实测结果和模型预测结果的对比表明,RBF神经网络预测模型能反映影响因素与特征量之间的非线性关系,适用于爆破振动特征参量预测。
Based on introducing the basic idea of RBF neural network, the prediction model of blasting vibration was built. The vibration amplitude, main vibration frequency, and vibration duration time of the particles were predicted with the RBF neural network. Blasting vibration data measured in main shaft exca- vation of the Yang Quan coal mine was used to train the model and 27 sets of data were predicted by using the RBF trained. The comparison of measured results and predicted results showed that RBF neural net- work prediction model could reflect the non-linear relationship between influencing factors and characteris- tic parameters, and it could be suitable for characteristic parameters prediction of blasting vibration.
出处
《工程爆破》
北大核心
2012年第3期29-32,共4页
Engineering Blasting
关键词
RBF神经网络
爆破振动
强度预测
误差分析
RBF neural network
Blasting vibration
Strength prediction
Error analysis
