Intelligent Optimization Method And Application Of Initial Supporting Parameters Of Surrounding Rock Of Highway Tunnel

The selection of initial supporting parameters of surrounding rock in tunnel engineering is not only related to the safety of tunnel structure,but also affects the project schedule and cost.Because the geological environment of highway tunnel engineering is relatively complex,there are many design parameters,there are great uncertainties and other problems,the support parameters of the general design scheme are conservative.How to optimize the supporting parameters of surrounding rock under the premise of ensuring the stability of surrounding rock is a key technical problem to be solved urgently in highway tunnel construction.Aiming at the above problems in the selection of surrounding rock support parameters,an intelligent optimization scheme of surrounding rock support parameters based on PSO-SVM algorithm was proposed.Midas GTS NX is used for numerical simulation of tunnel construction process,and the above algorithm is used for machine learning of simulation data,so as to achieve the purpose of automatically calculating the most economical and reasonable initial tunnel support parameters under the condition of given surrounding rock parameters and tunnel depth.The main research work and achievements of this paper are as follows:(1)The specification given the tunnel vault subsidence value as a primary support stability evaluation index,set up the initial supporting parameters optimization of surrounding rock of tunnel engineering construction have a single objective programming model,the initial supporting parameters optimization is in under the premise of guarantee the stability of surrounding rock in tunnel construction process,to achieve the maximization of the tunnel vault subsidence,The problem is solved by modern optimization theory.(2)The corresponding relationship between the stability of surrounding rock and surrounding rock parameters,support parameters and tunnel depth was obtained by numerical simulation of uniform test scheme,and the nonlinear intelligent model between the above parameters was established by PSO and SVM coupling algorithm to achieve the prediction of surrounding rock deformation.(3)By introducing particle swarm optimization(PSO)and SVM coupling algorithm,combined with the field construction and finite element simulation of Faling Tunnel in Jihuang Highway,Anhui Province,the above constrained single objective planning problem was solved.The results show that this algorithm has a good optimization effect,and gives full play to the advantages of support vector machine(SVM)fitting precision,small sample learning and particle swarm optimization(PSO)algorithm,which has fast convergence speed and strong search ability,and provides a solution for the optimization of surrounding rock support parameters.