With the continuous development of deep learning, the application of object detection based on deep neural networks in the coal mine has been expanding. Simultaneously, as the production applications demand higher recognition accuracy, most research chooses to enlarge the depth and parameters of the network to improve accuracy. However, due to the limited computing resources in the coal mining face, it is challenging to meet the computation demands of a large number of hardware resources. Therefore, this paper proposes a lightweight object detection algorithm designed specifically for the coal mining face, referred to as CM-YOLOv8. The algorithm introduces adaptive predefined anchor boxes tailored to the coal mining face dataset to enhance the detection performance of various targets. Simultaneously, a pruning method based on the L1 norm is designed, significantly compressing the model’s computation and parameter volume without compromising accuracy. The proposed algorithm is validated on the coal mining dataset DsLMF+, achieving a compression rate of 40% on the model volume with less than a 1% drop in accuracy. Comparative analysis with other existing algorithms demonstrates its efficiency and practicality in coal mining scenarios. The experiments confirm that CM-YOLOv8 significantly reduces the model’s computational requirements and volume while maintaining high accuracy.

In recent years, the deformation detection technology for underground tunnels has played a crucial role in coal mine safety management. Currently, traditional methods such as the cross method and those employing the roof abscission layer monitoring instrument are primarily used for tunnel deformation detection in coal mines. With the advancement of photogrammetric methods, three-dimensional laser scanners have gradually become the primary method for deformation detection of coal mine tunnels. However, due to the high-risk confined spaces and distant distribution of coal mine tunnels, stationary three-dimensional laser scanning technology requires a significant amount of labor and time, posing certain operational risks. Currently, mobile laser scanning has become a popular method for coal mine tunnel deformation detection. This paper proposes a method for detecting point cloud deformation of underground coal mine tunnels based on a handheld three-dimensional laser scanner. This method utilizes SLAM laser radar to obtain complete point cloud information of the entire tunnel, while projecting the three-dimensional point cloud onto different planes to obtain the coordinates of the tunnel centerline. By using the calculated tunnel centerline, the three-dimensional point cloud data collected at different times are matched to the same coordinate system, and then the tunnel deformation parameters are analyzed separately from the global and cross-sectional perspectives. Through on-site collection of tunnel data, this paper verifies the feasibility of the algorithm and compares it with other centerline fitting and point cloud registration algorithms, demonstrating higher accuracy and meeting practical needs.

Background: To optimize the layout position of the residual coal pillar return roadway when mining a close coal seam group and to clarify the optimization mechanism, a roadway optimization layout analysis was conducted on the Tashan coal mine. Methods: Surface displacement monitoring was conducted using field tests, and the main stress magnitude, plastic zone morphology, deformation variables, and connectivity between the plastic zone of the roadway and the plastic zone of the residual coal pillar were analyzed at different locations with the help of FLAC3D numerical simulation software. Results: It was found that, in the process of close coal seam group mining, the residual coal pillar of the overlying coal seam seriously affects the stress state and plastic zone distribution of the lower coal seam roadway. The roadway is arranged in a position that is relatively far away from the residual coal pillar, which could reduce the stress influence of the residual coal pillar on the roadway and guarantee the stability of the roadway. Conclusion: Since the Tashan Mine uses the top release method for mining, the stability of the roadway can be better ensured by placing the roadway in the middle and lower regions of the coal seam and using the layout method to retain small coal pillars.

With the continuous development of deep learning, the application of object detection based on deep neural network in the coal mine has been expanding. Simultaneously, as the production applications demand higher recognition accuracy, most research chooses to enlarge the depth and parameters of the network to improve accuracy. However, due to the limited computing resources in the coal mining face, it is challenging to meet the computation demands of a large number of hardware resources. Therefore, this paper proposes a lightweight object detection algorithm designed specifically for coal mining face, referred to as CM-YOLOv8. The algorithm introduces adaptive predefined anchor boxes tailored to the coal mining face dataset to enhance the detection performance of various targets. Simultaneously, a pruning method based on the L1 norm is designed, significantly compressing the model's computation and parameter volume without compromising accuracy. The proposed algorithm is validated on the coal mining dataset DsLMF+, achieving a compression rate of 40% on the model volume with less than a 1% drop in accuracy. Comparative analysis with other existing algorithms demonstrates its efficiency and practicality in coal mining scenarios. The experiments confirm that CM-YOLOv8 significantly reduces the model's computational requirements and volume while maintaining high accuracy.

Aiming at the surrounding rock failure of heterogeneous and unequal pressure circular roadways, the approximate analytical algorithm of the plastic zone is proposed. Through the comparison of analytical calculations and numerical simulation, the shape characteristics and evolution law of the surrounding rock’s plastic zone are studied, and the relative error of the two calculation methods is analyzed. The results show the following: (1) With an increase or decrease in the lateral pressure coefficient, the shape of the plastic zone of the roadway-surrounding rock presents a circular-ellipse-butterfly shape change law, and the outer edge of the plastic zone presents a scattered shape, without a continuous boundary. (2) In the butterfly-shaped plastic zone, the butterfly leaf always lies between the maximum confining pressure and the minimum confining pressure and rotates with the pressure. (3) There is a certain amount of error between the maximum value of the plastic zone solved by an analytical algorithm and numerical simulation, and the relative error is positively related to the burial depth and lateral pressure coefficient of the roadway and negatively related to the strength of the surrounding rock. (4) In addition, the relative error of the plastic zone calculation of the multi-layer heterogeneous combination roadway is larger than that of the single heterogeneous roadway.