<<
[an error occurred while processing this directive] | [an error occurred while processing this directive] >>
Rock Creep Simulation Based on the Drucker-Prager Criterion
(1 MOE Key Laboratory of Geomechanics and Embankment Engineering, Nanjing 210098; 2 State Key Laboratory for Geomechanics and Deep Underground Engineering, China University of Mining and Technology, Xuzhou 221008)
Abstract The stress states at different parts of surrounding rocks and the working face differ during tunnel construction. In conventional simulation, one Drucker-Prager (D-P) model and set of test parameters, which will only match with one stress state, are often used to predict the strain states of the whole model, which is impractical and may cause errors. Based on the ABAQUS′s creep model, in which the D-P yield criterion is coupled with the time hardening rate, this paper analyzes the prediction errors induced by several different series of D-P yield criteria and hardening parameters. The results show that: 1) apart from the yield criterion, the error is also greatly related to the ratio of the plastic factor (hardening parameter) [λ] under different stress state types, which means that the change rate of error is proportional to the value of[λ]; 2) if[λ≠1], there will be an error balance point P with the error of 0; 3) the absolute errors on the left side of point P decrease with an increase of the stress, while the opposite is true on the right side; 4) with a linear hardening model, a correction formula for the creep error can be established under different stress states.
Abstract：
The stress states at different parts of surrounding rocks and the working face differ during tunnel construction. In conventional simulation, one Drucker-Prager (D-P) model and set of test parameters, which will only match with one stress state, are often used to predict the strain states of the whole model, which is impractical and may cause errors. Based on the ABAQUS′s creep model, in which the D-P yield criterion is coupled with the time hardening rate, this paper analyzes the prediction errors induced by several different series of D-P yield criteria and hardening parameters. The results show that: 1) apart from the yield criterion, the error is also greatly related to the ratio of the plastic factor (hardening parameter) [λ] under different stress state types, which means that the change rate of error is proportional to the value of[λ]; 2) if[λ≠1], there will be an error balance point P with the error of 0; 3) the absolute errors on the left side of point P decrease with an increase of the stress, while the opposite is true on the right side; 4) with a linear hardening model, a correction formula for the creep error can be established under different stress states.
2016, Vol. 53 