Boundary elements and damage mechanics to analyze excavations in rock mass


Institute of Materials and Structural Models, Faculty of Engineering, Central University of Venezuela
M. Cerrolaza & R. Garcia
Opción D
1998

RESUMEN
 
No disponible

ABSTRACT
 
      This paper presents and discusses a nonlinear model based on damage mechanics theory, together with the boundary element method. The mechanical model proposed, as well as the computational tool developed in this research, have proved to be very efficient and simple, providing reliable information on the deformation and stresses in the rock mass. The model estimates the behavior of the damaged rock mass undergoing an excavation process, by assuming a nonlinear, heterogeneous and anisotropic medium. An isotropic model of damage is used, combined with a load relaxation process at the excavation wall, which is discretized via two-noded linear boundary elements. The computational approach is also described. Three illustrative examples which show the power, simplicity and versatility of the proposed approach are included and discussed in detail.

ÍNDICE
 
No disponible

CONCLUSIONES
 
      A coupled numerical model based on the boundary element method and damage mechanics theory is developed. The model proposed is able to predict the nonlinear behavior of an excavated rock mass, by using an isotropic damage model. The model works well since the BEM allows the modelling of infinite problems defined by nontrivial geometries and subjected to complex boundary conditions.
The examples presented and discussed herein demonstrate the power and versatility of the proposed approach. They show that the technique reproduces accurately either elastic or nonelastic behavior, as well as an anisotropic and heterogeneous medium. The results obtained when the present model is applied lead to an optimum support design and, as a consequence, to safer and more reliable tunnel structures. Moreover, one of the advantages of the model is that it combines the simplicity of the BEM for modelling complex geometries with the possibility defining subregions to reproduce complex geological structures composed of stratums of different materials as well as faults, cavities and other geological discontinuities.
Nevertheless, it should be pointed out that the proposed approach has its own limitations. It is unable to properly model the behavior of rock in the presence of macrocracks. This case would have to be treated by using the well-known fracture-mechanic approaches.
From a computational point of view, the codes developed (preprocessor, analysis and postprocessor ) have proved to be very useful when analyzing real engineering tunnel excavations. They can be used in other types of problem, either internal or external, composed of different materials and displaying a non-linear response.
A promising research line is open and, of course, further and intensive research should be done, in order to optimize the algorithms currently used. Further extensions of the model are under development, as is three dimensional analysis and the use of other nonisotropic damage models.