Unmanned aerial vehicle (UAV) tilt photography technology has gradually become a new technical means of disaster risk identification. This technology combines UAVs, satellite remote sensing, and ground online monitoring systems to establish an integrated space–sky–Earth system that can be used for tailings pond risk identification. With the use of this system for visual interpretation, water body identification, and monitoring data analysis, multiple types of monitoring parameters of a typical tailings pond in China, such as the seepage line and surface deformation, were obtained. Moreover, intelligent fusion analysis was performed of multisource data to outline the problems affecting tailings safety in the process of elevation expansion and irregular ore discharge of the tailings pond. Warning values of different levels were obtained to assess the overall safety condition of the tailings pond, and the proposed technology was verified. The research results could provide a new basis for accurate evaluation of the running state of tailings ponds and offer an effective remote monitoring means for tailings pond enterprises and supervisory departments.

With the large-scale exploitation of mineral resources, the safety problems of open-pit slopes and dumps become increasingly prominent. In this work, the safety demonstration of the interaction between the eastside slope of an open-pit quarry and the dump was carried out. Firstly, the influence of blasting vibration on the stability of the eastside slope and the dump was analyzed, and the propagation law of blasting vibration wave and the attenuation law of velocity were obtained. Secondly, the finite element analysis of the seepage field of the slope was carried out, and the basic seepage parameters of the rock and soil were determined. Then, the limit equilibrium analysis method was used to quantitatively calculate the stability of the eastside slope of the stope under the load of the dump to assess landslide risk. The results show that no landslides would occur at the dump or affect the performance of the slope, and that the slope was able to reach the required safety reserve factor. Then, through the finite element analysis of the 3-3 profile, it was found that the deformation caused by the excavation unloading of the eastside slope was mainly concentrated in the lower part of the slope, and the displacement variation caused by the top area of the slope was about 0~10 mm. The research verified the stability of slope and dump, which provides technical support for slope project design and blast operation in open-pit quarries.

Because an open-pit mine is an open operating environment, mining and stripping equipment inevitably pollutes the environment to some extent during the operation process. Therefore, the current dust concentration monitoring methods and technologies are relatively simple, and the dust distribution characteristics and diffusion laws of each production link in open-pit mines are not clear. Taking the transportation link of the Anjialing open-pit mine as the research object, a set of integrated monitoring methods combining ground and space and fixed and mobile technology is proposed. First, aiming at the complex transportation system of an open-pit mine, a physical model of dump truck movement in an open pit mine was constructed. On the basis of the principle of gas‒solid two-phase flow, a starting model of dust particles under impact and a dynamic model of dust particles under wind pressure were constructed, and the dusting, movement and diffusion mechanisms of dust particles were defined. Second, in view of the missing data and noise caused by the stability of the acquisition system and the coverage of the mining area signal, a method of dust concentration prediction based on time series and sample data correction processing of background noise was proposed. To fully restore and characterize the characteristics of the induction and influence of mining trucks on dust particles under different environmental conditions, a simulation method based on fluid mechanics, which effectively reveals the characteristics of air flow field migration and dust diffusion during the driving of mining trucks, was . Finally, on the basis of field monitoring data and numerical simulation data, the law of dust diffusion and migration of mining trucks under different speed conditions was revealed, and the characteristics of dust diffusion and migration prediction models were constructed. The relative error of the model prediction accuracy was 1.2% ~ 10.6%. The mean square error in the z-axis direction was 13.8%, whereas in the x-axis direction, it was 52.0%. The research results provide data support and a theoretical basis for follow-up dust concentration prediction in open-pit mines and the formulation of dust control programs.

Because an open-pit mine is an open operating environment, mining and stripping equipment inevitably pollutes the environment to some extent during the operation process. Therefore, the current dust concentration monitoring methods and technologies are relatively simple, and the dust distribution characteristics and diffusion laws of each production link in open-pit mines are not clear. Taking the transportation link of the Anjialing open-pit mine as the research object, a set of integrated monitoring methods combining ground and space and fixed and mobile technology is proposed. First, aiming at the complex transportation system of an open-pit mine, a physical model of dump truck movement in an open pit mine was constructed. On the basis of the principle of gas‒solid two-phase flow, a starting model of dust particles under impact and a dynamic model of dust particles under wind pressure were constructed, and the dusting, movement and diffusion mechanisms of dust particles were defined. Second, in view of the missing data and noise caused by the stability of the acquisition system and the coverage of the mining area signal, a method of dust concentration prediction based on time series and sample data correction processing of background noise was proposed. To fully restore and characterize the characteristics of the induction and influence of mining trucks on dust particles under different environmental conditions, a simulation method based on fluid mechanics, which effectively reveals the characteristics of air flow field migration and dust diffusion during the driving of mining trucks, was. Finally, on the basis of field monitoring data and numerical simulation data, the law of dust diffusion and migration of mining trucks under different speed conditions was revealed, and the characteristics of dust diffusion and migration prediction models were constructed. The relative error of the model prediction accuracy was 1.2%~10.6%. The mean square error in the z-axis direction was 13.8%, whereas in the x-axis direction, it was 52.0%. The research results provide data support and a theoretical basis for follow-up dust concentration prediction in open-pit mines and the formulation of dust control programs.

With the expansion of urban underground space, the structural safety of underground infrastructure has become increasingly critical. The urban underground utility tunnel is a typical deeply buried lifeline project. Its internal environment is humid and confined, and the structure is subjected to long-term heavy loads and earth pressure. Under such conditions, micro-cracks readily propagate into leakage channels and eventually cause structural damage. However, the slender morphology, low contrast, and complex background of cracks make it difficult for traditional inspection methods and general-purpose models to achieve reliable identification and accurate segmentation. This study proposes a two-stage framework that combines YOLOv8 with SAM2 and incorporates a coordinate attention module for high-quality crack segmentation. For image processing, a topology-aware post-processing strategy is introduced, together with a scoring function based on crack morphological features and a post-processing constraint mechanism, to ensure crack continuity and geometric consistency and to mitigate the over-segmentation that may occur during segmentation. YOLOv8-SAM2 achieves 85.2%, 90.5%, 77.2%, and 71.6% in mIoU, mDice, Recall, and Precision, respectively. Compared with YOLOv8-seg, mIoU and Precision are improved by 46.3 and 17.2 percentage points, respectively; compared with the baseline SAM2, mIoU and Precision are improved by 23.0 and 16.9 percentage points, respectively. In summary, on the self-built underground utility tunnel crack dataset, the proposed model significantly outperforms standalone YOLOv8 segmentation and the direct application of SAM2 in terms of intersection-over-union and precision, demonstrating its high-quality segmentation capability.