基于人工神经网络技术的隧道地表沉降预测

对采用人工神经网络技术预测隧道地表沉降模型中进行了研究。采用MATLAB系统开发了一个多层反向传播神经网络模型,考虑了隧道的深度、隧道的直径、地下水位、土的弹性模量、土的剪切强度、土的侧压系数、土的重度和开挖间隙对地表沉降的影响。用世界多个隧道的地表沉降数据作为样本对模型进行了训练和测试。结果表明,利用该神经网络预测的沉降值与实测值比较吻合。

Prediction of Tunneling-induced Ground Subsidence with Artificial Neural Network

This paper researches a method to predict tunneling-induced ground subsidence with artificial neural network.Based on MATLAB system,a multi-layer back propagation neural network model was developed,in which the effects of the depth from surface to the tunnel axis,tunnel diameter,groundwater level,as well as the elastic modulus,shear strength,side pressure coefficient and unit weight of soil and the space between excavated wall and lining on the ground subsidence were considered.The developed prediction model is trained and tested with the data obtained from different tunnel projects in different countries.It shows that the predicted results are in well accordance with the field observations.