The Back Analysis Of Surrounding Rock Parameters And Numerical Simulation Of Xiaomuling NO.1 Tunnel

With the development of the society,the rapid development of the national economy,the increasing number of population,infrastructure construction is becoming more and more important.People pay more and more attention to the use of underground space,such as in the traffic,the tunnel has become an important part of highway and railway etc.The construction of underground engineering,such as tunnel,involves a series of construction process,such as excavation and support of rock and soil.Therefore,the physical and mechanical parameters of rock and soil are of great significance to the design and construction of the project.However,due to the complex underground environment,such as a large number of test points,high cost,long cycle,the impact of construction and other factors,the mechanical parameters are difficult to obtain and the accuracy is not enough.In order to solve this problem,many domestic and foreign experts and scholars find inverse analysis for geotechnical engineering.This paper takes the Xiaomuling No.1 tunnel as the research object,through the method of BP neural network of genetic algorithm optimization,to displacement back analysis of tunnel,and then get the rock and soil parameters inversion.Using the parameters to build a tunnel model and doing the numerical simulation for the construction of the tunnel.Through the analysis of tunnel surrounding rock in construction process in support of the displacement deformation and stress distribution,to get the change of displacement of tunnel deformation and stress in each stage of tunnel construction and the whole process curve.The main work of this paper is as follows:(1)The original measurement data of the tunnel monitoring section were processed,and then the data were analyzed by regression analysis.Considering the spatial effect of tunnel excavation,the deformation of the monitoring section is calculated by the displacement release rate,and the vault settlement and clearance convergence are 18.62 mm and 9.24 mm.(2)The inversion of rock and soil mechanics parameters of artificial neural network,and the genetic algorithm optimization,orthogonal test scheme for monitoring section credentials created sample displacement back analysis,established corresponding displacement back analysis system,through the input calculated from the displacement of rock and soil mechanical parameters back calculate tunnel(E,C,?,?)for(0.67 GPa,0.15 MPa,26.5,0.35).(3)According to the numerical simulation of the step construction method,we get the surrounding rock displacement and stress analysis of each construction stage,and draw the related curve,vault displacement of 17.84 mm were occurred to the last stable,empty net convergence displacement 9.21 mm.(4)According to the numerical simulation of the full-section construction method,we get the surrounding rock displacement and stress analysis of each construction stage,and draw the related curve,vault displacement of 20.98 mm were occurred to the last stable,empty net convergence displacement 17.99 mm.(5)By comparing the different changes caused by step construction method and full section construction method,we get the conclusion that using the step excavation is more conducive to the stability of the tunnel.