Application of distance discriminant analysis method to classify the tunnel leakage

The prediction of possibility and classification of tunnel leakage are important issues in the process of tunnel construction and operation. Based on the principle of Mahalanobisdistance discriminant analysis (DDA), a distance discriminant analysis model to predict the possibility and classification of tunnel leakage is established, eight main control factors of tunnel leakage are chosen in the analysis, such as, the depth of tunnel, the situation of vegetation covering of tunnel area, annual average rainfall of tunnel area, and the situation of surface water of tunnel area. Some surveying data of tunnel leakage are taken as the training and testing samples, and the ration of mistake distinguish is considerably low zero. After the DDA model is trained, the tunnel leakage of Yuanliangshan tunnel and No.1 of Xiqi tunnel are used to verify this model, the results show that the DDA model can be used to predict the possibility and classification of tunnel leakage. The method offers a new way in classification of disaster of tunnel leakage.