Abstract:Tunnel construction is a complete system engineering. The characteristics of surrounding rock information and the monitored deformation,which is significant for future tunnel construction,have not been used effectively. Therefore,Gaussian process regression was proposed to establish the model of mapping between the natural properties and deformation of exposed rock mass based on the prior information of distribution of excavated surrounding rock. With the mapping model,the value of convergence deformation at excavating face can be predicted,and the stability of surrounding rock can be evaluated in real time. The interactive analysis on the stability of surrounding rock was carried out with the predicted deformation value as the pilot criterion. With the calculated results of multivariate heterogeneous information such as the interpreted data from the advance geological forecast,classification of field surrounding rock and so on,the mechanism of dynamic permit of construction and design change were established. In addition,the probability of whole deformation failure of the certain section with the same surrounding rock grade was evaluated using the GPR-MCS algorithm.
贺 鹏,李术才,李利平,张乾青,许振浩. 基于数据挖掘的隧道围岩变形响应预测与动态变更许可机制[J]. 岩石力学与工程学报, 2017, 36(12): 2940-2953.
HE Peng,LI Shucai,LI Liping,ZHANG Qianqing,XU Zhenhao. Prediction of deformation response in surrounding rock of tunnels and permit mechanism of dynamic change based on data mining. , 2017, 36(12): 2940-2953.
BIENIAWSKI Z T. Engineering classification of jointed rock masses[J]. Transaction of the South African Institution of Civil Engineering,1973,15(12):335-344.
Your browse does not support frame!