The early warning models for coal and gas outburst have become increasingly more important and have gained more attention in the mining industry in an effort to further improve mine safety. In the warning process, however, the theoretical models do not always work in a timely manner largely due to the delayed capture of the real time parameters. Based on the evolving mechanism of gas outburst, the gas emission is considered a dominant factor in this work because its data is attainable in real time and clearly characterizes the entire outburst process. In order to characterize and distinguish the variation of the gas emission during an outburst and normal mining activity, a total of four statistical methods were employed to quantify the variation of gas emission: the moving average, the deviation ratio, the dispersion ratio, and the fluctuation ratio. Also, the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are also included to demonstrate the accuracy of the deep learning model for predicting the variation of gas emission. Developed from these six indicators, the multi-factor fuzzy comprehensive evaluation model forms the outburst early warning system by calculating the combined index of the difference among the indicators. The accuracy of the early warning system is examined in the case study of the “3.25” gas outburst hazard in Shigang Coal Mine. The results show advantages of the comprehensive evaluation model established from the six characteristic indicators when predicting an outburst.

Coal is a kind of rock with the characteristics of soft structure, developed joints, cleats, cracks, and pores, resulting in that its mechanical behaviors are highly sensitive to stress, pore coalbed methane (CBM, gas) pressure and temperature. Thus, due to the variations in stress, pore pressure and temperature caused by the drilling operation, the coal surrounding wellbore can be easily damaged, which would cause serious wellbore instability problems. In this presented work, a stress path of loading axial stress and unloading confining pressure (LAS-UCP) was first determined based on the stress redistribution of the coal surrounding horizontal wellbore in CBM reservoir during drilling process. A series of triaxial compression tests with the LAS-UCP stress path was then conducted to study the effects of axial loading rate, pore pressure and temperature on the mechanical behaviors of coal. The results show that: (1) Under the LAS-UCP stress path, the deformation of coal can be divided into elastic deformation stage, plastic deformation stage, and stress reduction stage. With the decrease in axial loading rate and the increases in pore pressure and temperature, the elastic deformation stage becomes shorter, the plastic deformation stage becomes more significant, stress reduction rate in the stress reduction stage becomes slower, and the coal shows more features of plasticity and ductility. (2) With the increasing axial loading rate, the compressive strength and apparent elastic modulus increase linearly, the absolute values of axial strain, radial strain and volumetric strain at peak stress grow gradually, but the apparent Poisson’s ratio changes irregular. (3) With the increase in pore pressure, the compressive strength, axial strain at peak stress and apparent elastic modulus decrease linearly, the radial strain and volumetric strain at peak stress have no change rule, and the apparent Poisson’s ratio increases gradually. (4) With the increasing temperature, the compressive strength, axial strain at peak stress and apparent elastic modulus reduce gradually, but the absolute values of radial strain and volumetric strain at peak stress, and the apparent Poisson’s ratio increase linearly. The results can not only provide a guidance for safety drilling operation of the horizontal wellbore in CBM reservoir, but also have important significance for other engineering constructions related to coal seam.

The great threat and destructiveness brought by a rock burst make its prediction and prevention crucial in engineering. The rock burst hazard evaluation at project locations is an effective way of preventing rock burst since currently real-time prediction is not available. Since different control factors and discrimination conditions of rock burst were accepted by conventional risk determination methods, the rock burst risk determination in the same area may produce conflicting results. In this study, Naive Bayes statistical learning models based on different model prior distributions representing highly complicated nonlinear relationship between rock burst hazard and impact factors were built to evaluate the rock burst hazards. The results suggested that the Bayes statistical learning model based on a Gaussian prior has the strongest performance over four preset prior distributions. Combining the rock mechanics parameters measured in the laboratory and the stress data collected on the project sites, the proposed model was successfully employed to evaluate the kimberlite rock burst risk of a diamond mine in Canada. The Bayes statistical learning model exhibits its robustness and generalization in rock burst hazard evaluation, which can be generalized for similar engineering cases with enough supported data.

Energy evolution process of rock deformation is conducive to essentially reveal the rock failure mechanism and is of great significance to uncover the breeding of dynamic disasters in rock engineering. To characterize the damage evolution of energy dissipation during rock failure, the digital image correlation (DIC) technique is proposed to describe the rock failure mechanics and its energy evolution process. The uniaxial compression experiment of sandstone specimen was carried out, and the whole field deformation and failure characteristics of the rock had been captured by the DIC system. Measurement accuracy was verified by the fiber Bragg grating (FBG) sensor, the elastic region of the specimen was divided according to the location of strain localization band (SLB), and the evolution process of elastic strain energy of the rock was analyzed. The results show that the time history development of rock strain obtained by the FBG and DIC system matches identically, and the deviation of peak axial strain of both means is less than 5%, which verifies the applicability of DIC system. The uncoordinated evolution of rock deformation displacement field is discussed to reveal the crack development and failure form of the sandstone specimen under uniaxial compression. The energy evolution of the elastic region of the specimen is revealed, and the development of releasable elastic strain energy would be divided into three stages, which correspond to the stress–strain characteristics of rock failure mechanics. This study could provide an alternative analytical method for the experimental rock mechanics research studies.

The expected extraction efficiency of coalbed methane (CBM) depends significantly on the laws considered to govern its gas flow. This study applies a non-Darcy gas flow model to describe the CBM migration in mine gobs; by mine gob, we mean that it is a fractured zone along with massive cracks and the primary place where gas flows after mining activities. A permeation experiment involving crushed sandstone is first conducted to prove the CBM undergoes Forchheimer-type non-Darcy flow. Subsequently, the three-dimensional continuous distribution functions of the permeability parameters are determined. The non-Darcy flow model includes the influence of inertial force on the gas flow, which is neglected in the Darcy model. A coupling model is established based on the experimental results and the gas flow characteristics in different regions. Thereafter, the model and distribution functions are applied to a series of numerical simulations of CBM extraction at the Sihe coal mine in China, to ascertain the most appropriate location for a ground borehole. These simulations involve boreholes placed in three different zones: the natural accumulation zone (NAZ), the load affected zone (LAZ), and the compaction stable zone (CSZ). The simulation results show that the total extraction quantity expected from the borehole in the NAZ is 2.4 and 13.5 times that from the boreholes in the LAZ and CSZ, respectively. This confirms that the NAZ is the most suitable zone for a borehole. This research ultimately provides a realistic gas flow model for CBM extraction from mine gobs.