Stability of roadways and intersections

A major factor determining the efficiency and economy of underground coal mining is stability of the openings. Gate roadways and intersections are particularly susceptible to ground control problems due to the inherent wide roof span, excessive stress and variable intersection shape. Adverse conditions such as high horizontal stress and abutment pressures from longwalls also have a deleterious effect on ground control. As many past studies which dealt with site specific problems failed to address the fundamental mechanisms of failure, there are still many stability problems in underground coal mines. This research was carried out to determine the actual failure mechanism of roadways and intersections and includes case studies from Australian underground coal mines. The present work provides an in depth study of the behaviour of the roof, pillar and floor in roadways and at intersections leading to the design of appropriate support systems under various conditions. The research was conducted in three major parts to achieve the objectives of the project. Laboratory testing was carried out on various rocks to determine the mechanical properties of rocks and prepare input data for numerical modelling. Numerical modelling was conducted using the Finite Element Method to consider the effect which different variables have on the behaviour of roadways and intersections. Finally, a field investigation program was undertaken to obtain in-situ data for validation of numerical results. Each part of the research was performed in such a way that they were interactive with one another. Theoretical, empirical and numerical research relating to the stability of roadways and intersections were reviewed and also general features of the Finite Element Method (FEM) were described. The 3-D FE code, NASTRAN, was introduced and its limitations were discussed. A number of new modelling techniques were developed and applied in the program in order to gain the most realistic stability assessment of underground structures. These techniques enabled consideration of the post-failure behaviour, empirical failure criteria and complete stress-strain characteristics of rocks as well as allowing the assessment of bedding plane effects and appropriate loading techniques for underground structures. Two underground coal mines in New South Wales, Australia were chosen for field investigations and sample collection. Laboratory testing was designed for acquisition of complete information on the mechanical properties of coal measures rocks such as coal, shale, mudstone and three types of sandstone. More than 250 standard samples were prepared and tested using uniaxial and triaxial compressive, point load, direct shear and Brazilian tests. The results from the triaxial tests were examined against the Mohr­ Coulomb, Bieniawski and Hoek and Brown failure criteria. In addition to the above criteria a new failure criterion was proposed. Moreover, rock mass classification was used to determine the in-situ properties of strata units based on the classifications index and laboratory results. The conventional and proposed failure criteria were employed in the finite element analysis (FEA) of a roadway model to determine the limitations of these failure criteria in stability analysis of underground structures. The stability of roadways (2-D models) was investigated by means of FEA of general and site-specific models. The mechanism of interaction between the roof, pillar and floor under various conditions was comprehensively studied. The significance of high horizontal stress, post-failure behaviour of rocks and abutment pressure from longwall faces on the stability of main and gate roadways was addressed. A general approach was suggested for the design of an optimum (safe and economic) support system by evaluating three potential modes of failure (structural failure criteria); arch failure over the roof line, shear failure on the bedding planes and guttering over the rib-lines. The stability of pillars and the floor was considered by determining the extent and degree of yield in the elements surrounding the roadway. The results obtained from FEA of roadways employing the new techniques were in good agreement with field data. Further application of the finite element technique was extended to determine the influence of certain parameters on the behaviour of four-way and three-way intersections. Individual parameters such as depth of cover, the ratio of horizontal to vertical stress and the width of opening were varied during a parametric analysis using 3-D models of intersections. The influence of the above factors on the stress and displacement patterns around the intersections was determined. In addition, site-specific models of intersections based on field data were constructed and analysed. Results from this research were very promising and were validated by empirical investigations carried out both during this project and in the past by other investigators. On the basis of the results of the investigation, a new procedure was suggested for the design of an optimum support system at intersections.