The Study Of Model Simulation Of Data Monitoring Prediction For Highway Tunnel Based On Neural Network

Artificial neural network is a mathematic simulation model of Neuro System. It can reveal the non-linear relation of the data sample. It is an adaptive dynamic system made up of much more processing units. So, it is in the nature of great adaptation, self-organization, good performance of learning, reflex association fault-tolerance and nice capacity of resisting disturbance. It is easy to make model of complex systems.This paper takes three tunnels of three bid sections of the ZaXiao highway back to ZhenAn as a research object,and construct the model of monitoring data prediction using bp neural network technology to realize the prediction of tunnel deformation trend.First of all,this paper builds the FEM analysis calculate model based on the software ANSYS,gives a reference for layout of tunnel safety monitoring sensor.By comparing the theroy result with the monitoring data on the spot,they are approximate,so it can verify the FEM analysis model.In the next place,on the basis of the analytical study of bp neural network theory,this paper construct the model of monitoring data prediction based on bp neural network after having a general comparision with the model of different network configuration and different algorithm. By comparing the results which are obtained by the predictive model with the testing data witnessed inspections,we will get that the predictive effects of the model are good and it does well in the generalization performance.The reserch results in this paper can predict the deformation trend of structural parameter of the tunnel.It is of great significance in appling correctly the deformation laws of tunnel surrounding rock and timely proceeding maintain, and gives an effective reference for the construction and operation safety of tunnel,so it is of certain useful value in the project.