城市盾构隧道施工地表沉降BP神经网络预测应用研究

基于成都轨道交通17号线明光站—九江北站区间的相关监测数据，采用了回归分析方法，利用Matlab拟合Peck法中的相关参数，对DK68+849，DK68+869断面的地表沉降进行了预测分析。同时，通过建立BP神经网络模型，调整相关的网络设置参数来调整模型的MSE值，最终得出最佳预测模型，利用训练出的BP神经网络模型对地表的最终沉降量进行了有效的预测。

Application of BP Neural Network in Prediction of Ground Settlement in Urban Shield Tunneling

Based on the monitoring data of the interval from Mingguang Station to Jiujiang North Station of Chengdu Metro Line 17, herein regression analysis was used to fit the parameters in the Peck method with Matlab so as to predict the ground settlement of the section DK68+849 and DK68+869. At the same time, BP neural network model was established with its MSE value to be regulated by means of adjusting the network settings, thus the optimal prediction model was obtained. This trained BP neural network model was used to predict the final subsidence of ground surface effectively.