An advanced detection approach based on support vector machine during tunnelling

A new tunnel advanced detection method is proposed using disc cutter of tunnel boring machine(TBM) as a center electrode with DC resistivity principle, and unfavorable geology front or side of tunnel face is predicted. The numerical simulation of electrical resistivity is carried out using finite element method, meanwhile, the electric field distribution is calculated and discussed about abnormal characteristics under different ground conditions. Then the classifier based on support vector machine(SVM) algorithm is built to differentiate the position of abnormal geology body: front or side of tunnel face. The K-cross validation is used to choose the optimal parameters of SVM. According to the results, it can be said that the proposed method is useful and reliable means to predict the position of anomaly and provide the reference for site geological prediction.