Tunnel Surrounding Rock Deformation Prediction And Its Application In Engineering

In the tunnel construction, deformation stability of surrounding rock and tunnel engineering security is contacted together cheek by jowl, tunnel engineering accident of the main manifestation is the surrounding rock instability, which allows us to tunnel surrounding rock deformation monitoring seriously more and more, the deformation of surrounding rock stability judgment has become the main problem of the tunnel technology.At present, deformation prediction methods include empirical coefficient method, safety coefficient method, numerical method. In the actual engineering, the above method is widely used, but for the tunnel engineering, numerical simulation method, the tunnel surrounding rock’s complexity and unpredictability, makes the method only can be used as the theoretical reference for practical engineering. Therefore, more and more scholars focus on system method, system method used is mainly grey system theory and neural network method, this article analyzes the two kinds of method in tunnel surrounding rock deformation in application to undertake treatise.In this paper, to create a new "Ian yu" railroad tunnel in the "hua ma" as the main example, according to the actual monitoring of "hua ma" tunnel deformation data description and analysis, using the grey system theory and neural network two methods for surrounding rock deformation forecast. According to forecasting time different, this article selects different initial measured data, the selection of measured data as the basis, using two kinds of prediction methods on the tunnel surrounding rock deformation prediction. The forecast data are the list, according to the predicted and measured data contrast, choose the best predict the number of days, and according to the comparison graph of surrounding rock deformation stability time judgement.In this paper by using the grey theory and neural network method on the deformation of tunnel surrounding rock stability time was predicted at the same time, through the example of forecasting analysis and discuss, discuss advantages and disadvantages of the two methods, the two methods is made in detail. Grey forecasting model GM (1,1) is the main characteristics of simple model, using less data model. This greatly reduces the difficulty of modeling and computing, but grey theory prediction model for volatility data forecast effect is good, and the prediction accuracy over time and continue to decline. Neural network for nonlinear system prediction with good performance, but the training of the network requires a large amount of measured data, and neural network for parameter sensitivity is high, need a large number of parameters calculation, selection of the optimal parameters, otherwise it is difficult to achieve our desired results.Finally, this article on the new "Ian yu" change railroad in the "hua ma" tunnel prediction section for three-dimensional numerical simulation. This paper uses the Ansys software simulation, simulation of tunnel surrounding rock deformation prediction of "hua ma" section, surrounding rock stress, surrounding rock elastic-plastic etc, according to the simulation results are analyzed and discussed, judgment of surrounding rock in tunnel prediction section horse two times before supporting the deformation and stress of safety. Simulation after the discovery, we predict the deformation of surrounding rock in tunnel surrounding rock stability time, is safe and reliable.