Study On The Stability Of Tunnels With Large Span In Soft Rock

With the quick development of the expressway in our country, the large cross section expressway tunnels will be an important part of the highway tunnels. Because of the large span and restriction of the construction technology, the tunnel construction accidents often occur, especially in the area of soft rock. Take the Hangxinjing expressway in the west area of the Zhejiang province as an example, most of the tunnels have three lanes, and the sandstone is the most important part of the rock around the tunnels. The sandstone brings risks to the construction of the tunnel, because of intenerating in the water. So the in-depth study on the stability of the tunnels with large span in soft rock is needed to provide the safety of the tunnel construction. The research on the stability of large cross section expressway tunnels is shown as following:The in-site monitor is an important part of NATM, and can't be taken place for the safety of the tunnel construction. Base on Hangxinjing expressway monitoring data, author introduces the important points of monitor, shows the principle of the data management, and also points out the difficulties of the monitor.For solving the problems in the monitor, and effectively estimating the stability of the tunnel, this thesis studies the factors which can affect the stability of the tunnel, and analyses the action mechanism. However, because most factors are uncertain, and the relationship between every factor is unable to express accurately with the accurate mathematical formula, so this thesis applies fuzzing mathematical theory to evaluate and solve the target's fuzzy and uncertainty in evaluation models, and sets up an evaluation on the stability of tunnels. While confirming every index weight, this thesis adopts Analytic Hierarchy Process to reduce the influence of the human factors.Since there are not enough alert definition value to the monitor, basing on the fuzzing mathematical theory, according to the fuzzy cluster, the fuzzy pattern recognition, the eigenvalue of class variable, and the corresponding class variable, the thesis combines the fuzzy cluster, recognition and statistic, and then sets up a model to determine the deformation speed of the tunnel lining. This model can comparatively accurately determine the largest deformation speed of the tunnel section, which is now constructing. Basing on this model, author codes a computer program, and gets a good result by checking it in-site.