神经元网络在公路隧道支护设计中的应用

Application of artificial neural network in highway tunnel’s support design

选取有代表性的公路隧道支护设计工程实例,提取特征参数,归一化处理后,形成神经网络训练样本库。利用MATLAB程序设计语言编制BP神经网络可执行程序,经过训练,得到稳定的神经网络。应用神经网络进行工程实例预测,通过比较分析,证明了人工神经元网络在公路隧道支护设计中的可行性和可靠性。结果表明,神经元网络技术在公路隧道设计中应用的关键在于影响设计因素的确定和训练样本的选择。

Selecting representative design examples, from which characteristic parameters are extracted and regressed, thus the neural network training stylebook is formed. A BP neural network execute program is compiled in MATLAB. After training, it leads to steady neural network, which is used in practical project prediction. By comparison and analysis, the feasibility and reliability of artificial neural network in highway tunnel’s support design are verified. The key of artificial neural network’s application in design of highway tunnd’s shoring is to ensure influencing design factors and distilling disciplinal stylebooks.