Stress Threshold Determination Method and Damage Evolution Modelling Based on Micritic Bioclastic Limestone

The surrounding rock and slope of the exit of the No. 1 drainage tunnel of the Artashi Water Conservancy Project exhibited unpredicted deformation during excavation. The rock was determined to be micritic bioclastic limestone containing 55% bioclastic material, which has not been sufficiently studied. Conventional triaxial compression experiments on this material show that the deviatoric stress‒volumetric strain curves have no typical reversal arc and that the rock specimens rapidly expand at the peak stress. Thus, the method adopted by many researchers to determine the crack damage stress σ_cd by identifying the reversal point of the volumetric strain curve is not suitable for such rock. To overcome this drawback, the beginning of both axial and lateral cracks is considered to determine the crack damage stress σ_cd. Accordingly, a new method named the stress threshold determination method using axial and lateral crack strains (STDM-ALCS) is proposed to determine the micritic bioclastic limestone stress thresholds according to the crack evolution characteristics. This method can also be well applied to determine stress thresholds for common rocks such as granite, which proves the versatility of STDM-ALCS. The crack dissipation energy density is obtained by axial and lateral crack strains and divided into four stages according to the evolutionary characteristics of the crack and stress thresholds. By introducing damage variable D, defined as the ratio of the crack dissipation energy density to the peak dissipation energy density, the damage evolution characteristics at each stage are analysed. A five-element damage evolution model is developed by correlating the deviatoric stress ratio R d (ratio of deviatoric stress to peak deviatoric stress) with damage variable D in the four stages of rock damage. The damage evolution curves obtained by the proposed model satisfactorily fit with the experimental data in each stage, indicating that the model can effectively represent the damage evolution of this rock.