Modelling the onset of fracture-induced instabilities for underground mining operations

The propagation of fractures in coal seams and the surrounding rock mass can lead to instability at the face of longwalls or roof and ribs of underground roadways, resulting in injuries or delays to mining operations. In practice, bolt and cable systems are widely used to prevent roof failures in roadways and have proven effective when implemented correctly. However, the design of such systems requires an understanding of the stresses and displacements in the surrounding rock, which also depend on the fractures that form throughout their lifespan. This presents a challenge to engineers as, with few exceptions, the empirical, analytical, and numerical methods available at present have limited capabilities for modelling the propagation of fractures through rock at the mine scale. The development of advanced numerical methods that can accurately simulate the propagation of fractures within a rock mass will provide the tools needed for designing more reliable and economical bolt and cable support systems.

The focus of this project has been to investigate a new approach for aiding the design of efficient bolt and cable roof support systems in rock through the development and application of a novel phase-field finite element code that is capable of computing stresses, displacements, and fracture paths.

The method on which the code is based incorporates a Phase-field in addition to the usual displacement field, which allows cracks to be represented as smeared regions of damaged material within a continuum. The method has been used previously to accurately reproduce simple crack paths and load-displacement relationships in homogeneous and isotropic materials with or without structural features.

The phase-field method is different from many existing approaches for modelling fractures as it does not require that a network of possible failure surfaces be specified a priori or that elements must break to form new surfaces. Instead, the phase-field method permits fractures to initiate at any point and propagate along any path within a body, without having to make modelling assumptions (such as a specific mesh configuration) that dictate where the failure will start and how it will propagate. Therefore, the fracture pattern is determined by the stress state and material properties alone. As a result, the phase-field method has the potential to solve full scale problems of practical interest and is not limited to investigations involving the micromechanics of fracture. An additional benefit is that phase-field elements fit within the standard finite element framework familiar to many engineers and may therefore be implemented within existing commercial software packages.

Project outcomes:

• A literature review on methods for simulating crack propagation in rock, and the limitations which are addressed through this research;
• The structure and the main features of the computer code developed ab initio, including novel element technologies;
• The material models implemented to simulate isotropic and anisotropic fracturing of rock;
• The validation of the code against published data and against the results of experiments performed at the University of Newcastle; and benchmark testing performed against a state-of-practice analysis method for a field-scale roadway problem.