Using image recognition technology to realize coal gangue recognition is one of the development directions of intelligent fully mechanized caving mining. Aiming at the problem of low accuracy of coal gangue recognition in fully mechanized caving mining, the extraction method of Coal and gangue images features is proposed, and the corresponding coal gangue recognition model is constructed. The illuminance value is an important factor affecting the imaging quality. Therefore, a multi-light source image acquisition system is designed, and the optimal illuminance value suitable for coal and gangue images acquisition is determined to be 17,130 Lux. There is a large amount of image noise in the gray-sc5ale image, so Gaussian filtering is used to eliminate the noise in the gray-scale image of coal and gangue. Then, six gray-scale features and four texture features are extracted from 900 coal and gangue images respectively. It is concluded that the three kinds of features of gray skewness, gray variance and texture contrast have the highest discrimination on coal and gangue images. Least squares vector machine has a strong ability to classify, so the use of least squares vector machine to achieve coal gangue identification, and build coal gangue identification model. The results show that the recognition accuracy of the model for coal gangue images is 92.2% and 91.5%, respectively, with gray skewness and texture contrast as indicators. This study provides a reliable theoretical support for solving the problem of low recognition rate of coal gangue in fully mechanized caving mining.

To address the technical limitations of automatic coal and gangue detection technology in fully mechanized top coal caving mining operations, the low radiation level radioactivity measurement method is utilized to assess the degree of coal-gangue mixture in top coal caving process. This approach is based on the distinguishing radiation characteristics of natural γ-rays between coal and gangue. This study analyzed the distribution characteristics of natural γ-rays in coal and rock layers of thick coal seams and the applicability of this method, introduced the basic principle of coal-gangue detection technology based on natural γ-ray, developed the test system about automatic coal-gangue detection, studied the radiation characteristics of coal and gangue, proposed determination model of the coal-gangue mixed degree, combined with the time sequence characteristics of the top coal’s releasing flow and the energy spectrum characteristics of different layers of rock, realized the precise coal-gangue detection technology in complex structure thick coal seam with multiple gangue. Field tests were conducted in Lilou, Xiaoyu and Tashan Coal Mine. The test results were well corroborated with the research results and achieved the expected results, which laid the foundation for the field application of intelligent coal mining.

The large mining height fully mechanized top-coal caving mining technique has developed rapidly and become the most extensively used mining method for ultra-thick coal seams. The arching of coal/gangue and the drawing out of gangue are peculiar phenomena in the process of fully mechanized top-coal caving mining, which not only affects the recovery of top-coal, but also affects the quality of the coal. This paper studies the arching phenomenon in top-coal caving mining process of ultra-thick coal seam. A series of laboratory granular material simulation experiments were performing and a top-coal arching model in the framework of mechanics was established to explore the formation characteristic of arches and their effects to top-coal loss. Then the countermeasures against the arches and technology path of intelligent mining based on improving top-coal recovery were put forward and performed in practice. The results show that the recovery ratio of top-coal has increased nearly 6%, and increased the production efficiency at the same time. The research on arching mechanism and removing strategies of dynamic random arches effectively improves the efficiency of fully mechanized top-coal caving mining in ultra-thick coal seams, and provides the foundation for the realization of intelligent top-coal caving mining technology.

A coal mine in Datong is an integrated mine. At present, there is goaf in the upper and lower part of the mining coal seam. There is a lot of ponding in the goaf, which has great potential safety hazards for production. In order to find out the scope and location of ponding in goaf, the comprehensive geophysical exploration method combining transient electromagnetic method and high‐density resistivity method is used to carry out the research. Firstly, the time‐base, turn‐off time, receiving delay, current, superposition times, and other parameters of the instrument are tested on the surface of known goaf to obtain the best instrument parameters, and the parameters are used to verify the feasibility of the research scheme; then, the transient electromagnetic method is used for large‐area exploration on the surface of the mine, the suspected goaf ponding area is found through comprehensive analysis, and the high‐density resistivity exploration is arranged in the suspected goaf ponding area. According to the obtained results, the scope and location of the goaf ponding area are accurately located through comprehensive analysis. The results show that there are two goaf ponding areas in the exploration area, which are located above the 8# coal seam currently mined; the range and location of goaf ponding area can be accurately obtained by using the comprehensive geophysical method of high‐density electrical method and transient electromagnetic method. This method can provide reference for mine water prevention and control in Datong area and has great practical significance to ensure coal mine safety production.

Due to the existence of a hard and massive roof (HMR), severe ground pressure behaviors have been observed at the working face, resulting in safety issues and the degradation of production effectiveness. Based on the HMR conditions of the Datong Mining Area, the fracture-related instability of the HMR and its effects on the support selection were investigated by analyzing the interaction between support and overlying strata. Advancefixed-distance presplitting blasting (AFPB) technology was proposed to control the caving interval of HMR, and the influence of the controlled interval on the working load of supports was also analyzed. The working load of the support and the caving interval of the HMR were determined based on the controlled HMR fracture technology, and these were verified by field application tests. The working resistance of the support and the step distance were determined based on controlled roof fracture and were verified by on-site application experiments. The results revealed that cracks emerged after the presplitting blasting, resulting in significantly reduced strata behaviors. Furthermore, the support exhibited good adaptability.

Coal is the main energy source in China, but coal mining is a high-risk industry, making the prevention and control of coal mining hazards an important topic. Constrained by the complexity and unpredictability of underground spaces, current research on coal mining disaster prevention and control technologies mainly focuses on the characteristics of overlying strata and the laws of mine pressure, resulting in significant deficiencies in accuracy. Given this, a data-driven pressure prediction method is proposed, which uses deep learning models to learn the patterns in existing data and generate the required predictions. This approach avoids the challenges of accurately extracting rock mass physical and mechanical parameters and geological structure modeling, thereby improving the accuracy of disaster prevention and control. The stage of working face pressure exertion is a period prone to disasters during coal mining. To achieve accurate prediction of working face pressure, the task is divided into three steps: the first step is to predict support resistance data ahead of the working face, the second step is to classify the pressure labels of coordinate units, and the third step is to predict the characteristic parameters of pressure exertion. Deep learning models were designed and trained separately for each of the three steps: For the first step, a deep Spatiotemporal sequence model was selected, and the trained model achieved a mean absolute error of 4.65 kN in prediction. For the second step, an image segmentation-based classification model was chosen, with the trained model reaching a classification accuracy of 97.77%. For the third step, a fusion model consisting of three LSTM (Long Short-Term Memory) networks was designed. The trained model achieved a mean absolute error of 0.17 for the dynamic pressure coefficient, a maximum resistance error of 810.93 kN during the pressure period, an error of 9.96 cycles for the pressure duration, and a classification accuracy of 92.35% for the pressure type. Simulating the actual situation of application scenarios, the input data for the second and third steps were set as the output data from the previous step, and the model was evaluated. The model achieved a mean absolute error of 1035.21 kN for the prediction of support resistance and classification accuracy of 82.90% for the pressure labels of coordinate units. In the simulated scenario, there were 9922 instances of pressure exertion, and the model predicted 10,336 instances, with 9046 of them matching the actual instances. The prediction of characteristic parameters was evaluated for 4946 instances of pressure exertion, which included three complete pressure exertion cycles. The mean absolute error for the dynamic pressure coefficient was 0.21, the maximum resistance error during the pressure period was 1218.31 kN, the error for the duration of the pressure cycle was 11.03 cycles, and the classification accuracy for the pressure exertion type was 91.75%.