Identification Of Tunnel Surrounding Rock Constitutive Model And Its Application In Tunnel Stability Evaluation

Rock creep is an important factor affecting tunnel stability.At present,indoor rock creep test is the most important means to understand creep characteristics.However,the indoor test is not consistent with the actual stress of rock mass.The inverse analysis method of rock mass mechanical parameters based on on-site monitoring information has become a new way to study rock mass creep.In this paper,the creep characteristics of rock and tunnel surrounding rock are studied by back analysis optimization algorithm and applied to the stability evaluation of surrounding rock.The main research contents are as follows:1)Research differential evolution simulated annealing algorithm(SADE).According to the nonlinear characteristics of rock creep,differential evolution algorithm(DE),which is good at solving nonlinear function optimization problems,is introduced to retrieve creep parameters.In order to make up for the shortage that a single differential evolution algorithm may fall into local optimum,Metropolis criterion with a certain probability of accepting deteriorating solution in SA is embedded to enhance the ability of the algorithm to jump out of local optimum solution.Two Benchmarks functions are used to test and compare the performance of SADE algorithm and DE algorithm.The results show that SADE algorithm has strong global search ability,high accuracy and fast convergence speed,and is an effective method for solving nonlinear constrained optimization problems.It is obviously superior to DE algorithm in convergence speed and stability.The algorithm is applied to the inversion of rock creep parameters,and the numerical experiment of rock triaxial compression verifies that the algorithm can better invert creep parameters,providing a reference for similar engineering research.2)The SADE algorithm is used to invert the creep constitutive model and creep parameters of rock.Firstly,the mechanical characteristics of rock creep are studied,and the constitutive equation corresponding to creep characteristics is introduced.Then,the creep equations of rock uniaxial creep and triaxial creep are derived by taking the original model as an example.Secondly,the creep parameters of rock uniaxial creep test are retrieved by SADE algorithm.Then,the creep parameters ofthe Nishihara model are retrieved by SADE algorithm.Finally,the creep constitutive equation of rock triaxial creep test is identified by SADE algorithm,which shows that the essence of rock model identification is still creep parameter identification,and the optimal model is selected by multiple inversion ratios.3)The SADE algorithm is used to invert the creep constitutive model and creep parameters of tunnel surrounding rock.Taking the simplest round roadway as an example,the objective function is established by deducing its analytical solution and the surrounding rock parameters are retrieved.At the same time,the equivalent elastic modulus of the surrounding rock of the tunnel is used as the characteristic quantity to identify the creep model of the surrounding rock,which can provide a mechanical model for the deformation prediction of the surrounding rock of the tunnel.4)Study an improved Burgers creep model.The classic Burgers model is composed of linear elements and cannot describe the nonlinear characteristics of rock creep.The Burger model is improved based on fractional calculus theory.The Newton body of Maxwell element is replaced by fractional software element.Aiming at the defect that Burgers model cannot describe accelerated creep,the nonlinear viscoplastic NVPB(Nonlinear Viscous Plastic Body)model is connected in series with Burgers model to obtain an improved Burgers creep model based on fractional calculus.The three-dimensional creep equation of the model is derived,and the model parameters are retrieved from the triaxial creep test of the green schist,which verifies that the model can describe the creep characteristics of the green schist at all stages.Through the sensitivity analysis of model parameters,it is known that the improved Burgers model can describe the size and growth speed of creep deformation in the second and third stages by fractional index ?,and the creep index n describes the growth speed of accelerated creep deformation.5)The customization process of the improved Burgers model in FLAC 3D is studied.This paper introduces the concrete steps and operation methods of the second development of the improved Burgers Creep Model,and gives the link map of the successful generation of userburger.dll,as well as the commands and instructions for interfacing and calling the link with FLAC3 D.Finally,a simple numerical example is used to verify the correctness of the second development.6)Combined with the above research results,for Jiuling Tunnel Project,the SADE algorithm is used to retrieve the surrounding rock parameters,and the equivalent creep model of surrounding rock is identified by taking the equivalent elastic modulus as the characteristic quantity.The improved Burgers model can better describe the creep characteristics of surrounding rock.The improved Burgers model is embedded into FLAC3 D by using the self-defined method of constitutive model,and creep characteristics of tunnel surrounding rock are analyzed during construction,which provides reference for similar engineering construction.