Stability Analysis of a Non-pillar-Mining Approach Using a Combination of Discrete Fracture Network and Discrete-Element Method Modeling

This paper presents a study on the stability of a mine entry of an innovative approach to non-pillar coal mining—gob-side entry retaining using cantilever beam theory (GERMC). This method is on the rise in China and it has shown to increase the coal seam rate of recycling and mining productivity. The method utilizes the caved-in material resulting from the mining activities to fill the gob area and ensure the stability of the roof of the already mined area. The caved-in material will form the gob-side wall of the retained entry. This characteristic is unique to this method. Thus, there are not many studies that analyze the stability of the filling material and of the gob-side entry wall under these circumstances, which is a vital element for the safety of the mining operations. This paper studies the stability of the gob-side retained entry of the Hongjingta underground coal mine in China. The authors perform the analysis using a discrete fracture network (DFN) model developed by the Massachusetts Institute of Technology (MIT), GEOFRAC, in combination with the discrete-element method (DEM) software UDEC. GEOFRAC estimates the rock blocks (fracture networks) in the filling body, from fracture traces measured along the gob-side wall of the entry. Statistical methods are used to estimate the fracture intensity and the mean fracture areas, which are inputs to the DFN. The generated fracture networks are then inputted into the DEM code, UDEC, to evaluate the stability of the mine-retained entry. For this study, we developed two types of models in UDEC, one considering the fractures generated by the DFN model and another considering the gob-filling material as a continuum. This methodology—combined DFN–DEM—is unique and provides a more realistic representation of the gob-filling material by considering the fractures in the filling body and by estimating those fractures (interfaces between rock blocks) using field data. In addition to this, we also considered the effects of uncertainties that are associated with estimating three-dimensional (3-D) fractures’ networks/rock blocks from two-dimensional (2-D) field data. Finally, the results of the simulations were compared with the measurements from the mine and are in good agreement.