A fracture mechanics numerical code, FRACOD, has previously been developed to predict the explicit fracturing process such as fracture initiation and fracture propagation in rock masses. It has been used in studying borehole breakouts, tunnel stability, and rock mass stability for nuclear waste disposal. FRACOD has recently been further developed to predict sub-critical crack growth and time-dependent rock behaviour. This paper describes the theories of sub-critical crack growth and numerical procedures implemented in the code. This new function in FRACOD is validated against creep test results of coal and hard rock samples. The coal samples contain numerous fractures (cleats) and represent a highly fractured rock mass; whereas the hard rock sample is basically intact rock which only contains microcracks. The model results are compared with the creep test results of the two types of rock samples and the capability of the FRACOD in modelling timedependency is discussed.
Time-dependency is an important consideration when studying the long-term stability of large slopes, underground caverns, and nuclear waste repositories etc. In brittle rocks, time-dependency is mainly dominated by sub-critical crack growth. Flaws, defects, voids and other small cracks tend to grow slowly when the stresses are less than the material strength or the stress intensity factor is less than the fracture toughness. Traditionally, rock mass time-dependency has been modelled by the principles of continuum and damage mechanics. However, when the instability of the rock mass is dominated by explicit fracturing, it is more realistic to use theories of fracture mechanics and sub-critical crack growth to predict its time dependent behaviour.
2 THEORETICAL BACKGROUND OF FRACOD
FRACOD is a two-dimensional code based on Displacement Discontinuity Method (DDM) principles. It predicts the explicit fracturing process including fracture sliding/opening, fracture initiation and fracture propagation in rocks.