Precursory characteristics of coal-rock failure using acoustic emission and computed tomography under uniaxial monotonic and cyclic compression

Predicting coal-rock instability under cyclic loading is a significant challenge. However, there is a lack of studies that have systematically investigated damage evolution and failure precursor in the cyclic loading of coal. This research has combined the scanning electron microscope, acoustic emission (AE) monitoring, and X-ray microcomputed tomography (CT) scanning techniques for uniaxial monotonic and cyclic load-unload experiments on coal specimens. The damage evolution and failure precursors of coal specimens under monotonic and cyclic loading were investigated using the cusp catastrophe theory and the load and unload response ratio (LURR) theory, respectively. The onset of AE quiet period before peak stress was considered the predicting point of coal failure, around 90.5 % of coal strength. The cusp point of accelerated release of AE energy was determined the key predicting point based on the cusp catastrophe theory, about 95.5 % of coal strength. In cyclic load-unload tests, LURR values based on multiple AE parameters approaching 1 before peak stress was viewed a failure precursor. The CT observations show that the crack area and aperture extent of coal specimens increased as the cyclic loading progressed and peaked at failure, with 110 mm2 and 5.6 %, respectively. The fractal dimension, derived from microcrack distributions using the box counting method, increased continually in the cyclic loading and peaked at failure. The SEM and CT observations at the microscale and AE characteristics at the mesoscale were consistent under staged cyclic loading. Results can provide insights into prediction of coal-rock instability in underground mining.