Prediction and Classification of Large Deformations in Deep Tunnels Based on Stress Inversion Method

Large deformation disasters of tunnels greatly affect the construction time and quality of the tunnel. In order to best reduce the impact of large deformation disasters on tunnel construction and prevent the problems before they occur, this paper had proposed a method of large deformation disaster prediction and grading technology for deep buried tunnels with high stress, which mainly depends on stress inversion; this model was applied to predict the location and severity of the large deformation disasters of Anding tunnel. Lastly, the accuracy and characteristics of the prediction results were judged using the statistical data of large deformation disasters that occurred during the tunnel construction process. The results show that the errors between stress inversion and borehole measured data were basically within 25%; thus, it is feasible to obtain the in situ stress distribution characteristics of the tunnel by means of stress inversion. The large deformation disaster prediction method proposed in this paper had an accuracy of 62.5% in predicting the location of large deformation disasters, achieving good results in this respect; hence, it can play a very important role in further prevention and control of large deformation disasters.