The Research On Back Analysis Of Rock Parameters And Numerical Simulation Of Lanjiayan Tunnel Portal Construction

With the rapid development of national economy, our country has built a large number of railway highway.Tunnel and other underground engineering’s construction is facing great opportunities and challenges.The physical and mechanical parameters of the geological body which is related to the safety and economical efficiency of underground engineering have great significance to the design, construction and mechanical analysis of the underground engineering.However,the complexity of the underground environment leads that these important physical and mechanical parameters is difficult to determine. In order to solve this problem, this paper based on portal section of the Lanjiayan tunnel as the research object studies the displacement back analysis method on MEABP neural network and uses the parameters inversed in the numerical simulation of the main hole tunnel construction.After that,analysis of the surrounding rock’s in tunnel construction,and the supporting structure’s displacement and stress characteristics, the whole process of the tunnel deformation curve and displacement release rate of key point on the tunnel walls with time and location is operated.In this paper, the main research contents and achievements have the following several aspects:Firstly, as the spatial effects of tunnel excavation, the regressive analysis of the parallel pilot’s monitoring data should be realized and the final prediction of monitoring data would be done next.According to previous studies, the release rate on class V surrounding rock after excavation is referred on the research of Li Shucai in Shandong University.By that,the real deformation of monitored data on the section PDK51+326 of tunnel is calculated.Secondly,as the defects of BP neural network, the BP neural network is optimized by mind evolutionary algorithm(MEA).According to the scheme of the orthogonal experiment the numerical simulation of parallel pilot construction shapes sample set of displacement back analysis by FLAC2 D.The displacement back analysis system based on MEABP neural network is established after training.Then input the real deformation into the displacement back analysis system,mechanical parameters of surrounding rock to be inversed will be output automatically.Thirdly,adopt the parameters obtained to calculate the construction of the main cave tunnel and analyze the displacement and stress on the section K51+310.Some principles can be summarized.Fourthly,by the analysis of the whole process of horizontal displacement and vertical displacement curve of the vault, the hance, the arch foot and the center of them,some rules can be summarized as follows:the trend of the vault vertical displacement shows as the type of reversed S and affected slightly by down-stage excavation.On the contrary,the horizontal displacement of the arch foot are greatly influenced by the down-stage excavation and showed as the type of double S. The vertical displacement of hance and vertical displacement of thevault have almost the same trend,but its vertical displacement is less than vertical displacement of the vault at each step. Its horizontal displacement is greatly influenced by the down-stage excavation.The horizontal displacement curve is showed as “ 厂 ”,Finally horizontal displacement after the down-stage excavation becomes stable quickly and vertical displacement needs for a long time.Fifthly,according to the records of numerical simulation,the displacement release rate and displacement of the key point(arch, arch waist, arch foot and the centers of them) on the tunnel wall under the condition of numerical simulation are studied,the obtained conclusions follows such as some aspects :vault zone has a larger displacement release rate, and arch foot has a small displacement release rate, roughly in line with the quadratic parabola decreasing trend;when the location of the point is constant, displacement and the release rate of the measuring point change in line with S curve;And when the time is constant, the displacement and displacement release rate of key point has a quadratic parabola relationship roughly about the vertical angle.Finally as result of the particularity of the research question, displacement release rate function simplified as a two-dimensional function the vertical angle and time,according to the surface, displacement release rate is regular, and apply multiple regression analysis to the fitting.Then it is concluded that the tunnel displacement release rate function of tunnel change with position and time.By observe the error, it is found that near the zone t=0has a lager error, in the zone far away from t=0 the error is smaller, the result of regression is better.Sixthly, according to displacement release rate studied, combined with the measured sequence of the main cave K51+326 section arch top settlement, the method which is applied to modify the measured settlement sequence into the whole process of curve by adopting S growth model is studied.