A coupled elasto-plastic damage model for fine-grained sandstone under triaxial compression and lateral extension loading conditions

Abstract

A series of triaxial compression tests under confining pressures of 0, 3, 5, 10, 15 and 20 MPa and lateral extension tests under initial confining pressures of 10, 15 and 20 MPa are conducted. Compared to the test results under loading and unloading conditions, lateral deformation is more sensitive than the axial deformation under unloading process. According to the characteristic of volume strain, the stress–strain curves can be divided into five phases: initial nonlinear compaction, linear elastic, stable crack growth, unstable crack growth and post-peak stage. The experimental results show that dominant failure mode of both tests is shear failure. The failure mode becomes more complex at the confining pressure of 20 MPa in the lateral extension test, consisting of shear and tensile failures. Based on experimental results, a coupled elasto-plastic damage model is proposed using a nonlinear yield function in the p–q plane, using a non-associated plastic potential function in the irreversible thermodynamic framework. The isotropic damage evolution model is formulated with plastic coupling. The numerical simulation results using the coupled model agree well with the experimental data of the fine-grained sandstone, indicating that this model can describe the main mechanical behaviours of the rock under complex loading conditions.

Keywords:

• Triaxial compression test
• lateral extension tests
• fine-grained sandstone
• coupled elasto-plastic damage model

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Financial supports provided by the National Key R&D Program of China (2018YFC1508501), Postgraduate Research & Practice Innovation Program of Jiangsu Province (Grant No. KYCX17_0471, 2017B700X14), the Natural Science Foundation of China (Grant Nos. 11572110, 1177211) and Qinglan project are gratefully acknowledged.