摘要
提出了一种用于识别锚杆杆侧刚度因子的小波神经网络模型.首先,利用均匀设计方法建立锚杆系统的样本库,采用数值模拟得到各样本的锚杆杆顶的动测信号;然后将小波分析和人工神经网络结合起来,利用小波分析技术提取的能表征系统状态的特征向量作为输入向量,利用广义回归神经网络识别杆侧刚度因子;最后,利用样本对所建立的小波神经网络进行训练.分析结果表明:训练后的神经网络能够较好地识别锚杆系统杆侧刚度因子,为锚杆系统的锚固质量评价提供了一个有效的智能化手段.
A new analytical method used in identification for bolt-surrounding rock structural system was put forward, which combined with advantages of wavelet analysis and artificial neural network. Firstly, the samples of bolt＇ s anchorage system were acquired by uniform design and the dynamic responses of these bolts were obtained by numerical simulation. Secondly, the features were extracted from the dynamic responses by wavelet analysis. A generalized regression network was used to estimate the values of bolt＇ s side rigidity factors. Finally, these features were fed into the artificial neural network for training. The results indicate that this wavelet neural network after training can best identify the bolt＇ s side rigidity factors and can be a useable intelligent mean to assess the quality of bolt＇ s anchoring system.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2009年第10期1333-1338,共6页
Journal of China Coal Society
基金
国家杰出青年科学基金资助项目(5062824)
国家自然科学基金联合资助项目(50679097)
关键词
小波分析
神经网络
锚杆
杆侧刚度因子
wavelet analysis
neural network
bolt
bolt＇ s side rigidity factor
