In-situ leaching of uranium, as a green uranium mining technology, generates massive data in production operation, which are available for the big data analysis and trend prediction to improve the reliability of technicians in making production plans. In the current prediction algorithms, the attention mechanism in the temporal prediction model based on the encoder-decoder structure has the problems of computational complexity and high memory consumption. In this paper, we proposed a depthwise separable convolutional model, in which the semantic damage caused by fixed segmentation was reduced by the dynamic sequence segmentation module, and the depthwise separable convolutional mixer module was used to reduce the model running time and capture local features as well as global features. The results show that the Mean Square Error (MSE) and Mean Absolute Error (MAE) of the depthwise separable convolutional hybrid network model are reduced by 1.04% and 4.13% respectively, compared with Patch Time Series Transformer (PatchTST), and the proposed dynamic sequence segmentation module MSE and MAE are reduced by 7.32% and 5.03% respectively, compared to the original model; in the comparative performance analysis, the training speed of this model is 59.91% faster than the Trend Seasonal Decomposition Linear (Decomposition Linear, DLinear) model. The depthwise separable convolutional model can accurately predict the future trend of sulfuric acid injection volume in the production operation of the mining area, improve the existing prediction model for in-situ leaching data by solving the problem of long running time, large running memory, poor data fitting problems, which provide a theoretical and practical reference for the decision-making of in-situ leaching production.

The design of injection and extraction fluid volume is closely related to both the effective leaching range of the lixiviant and the operational cost of the well field. To facilitate economically efficient decision-making regarding injection and extraction fluid volume in the "horizontal well-vertical well" system within the in-situ leaching uranium mining domain, a simulation-based optimization decision method for the "horizontal well-vertical well" system is proposed, leveraging multi-objective optimization algorithms. The research indicate that, compared to the initial scheme of uniformly distributing injection and extraction fluid volume, the decision scheme R2 achieves an 8.84% increase in the leaching range while reducing operational costs by 51.14%. This method has identified four "key wells" in the LK mining area, emphasizing the importance of closely monitoring the flow rates of such wells during scheme decision-making and adjustments. Based on the Pareto solution set derived from the multi-objective optimization, six combinations of injection and extraction operational schemes with varying effective leaching ranges and cost-weight ratios are selected to guide decision-makers in their scheme selection.

For a multi-layer sandstone-type uranium ore in an in-situ leaching uranium mine in Inner Mongolia, four experimental methods of layered injection and layered extraction, layered injection and mixed extraction, mixed injection and layered extraction, mixed injection and mixed extraction were designed. The amount of injected liquid, leaching situation, advantages and disadvantages of each schemes were analyzed. The feasibility of single well layered injection was explored, and the key technology of single well layered extraction was solved. The results show that the layered injection and layered extraction method has the fastest increase in uranium concentration of the leachate, the best leaching effect, and lower construction cost, which has good application and promotion prospects.

In order to study the influence of heterogeneous strata on blasting, the stratigraphic factors and coupling methods were analyzed, and the thickness, density and dip angle of the rock strata were numerically studied by different coupling methods. The results show that the increase of the thickness of the middle sandstone and the increase of the density of the upper and lower layers will inhibit the crack propagation, and the inclined sandstone rock layer will produce horizontal and oblique staggered cracks, and the water-coupled charge blasting can protect the blast hole and increase the effective stress of the blast stress wave, and the selection of appropriate uncoupling coefficient can increase the energy transmitted by the blasting stress wave. The blasting force of the cracker is greater, and the addition of a warhead can form an initial crack and guide further crack propagation. Designing a reasonable scheme according to the formation condition can improve the blasting and permeability efficiency.

To address the wastewater treatment issue in electrode uranium plating process, an evaporation concentration device suitable for the special requirements of this process has been developed, which can effectively reduce the discharge of radioactive process wastewater. The principle of this device is to vaporize the water in the waste liquid through micro-vacuum distillation, condense it back into liquid water, and then discharge it externally. The remaining concentrated water and solid in the evaporator are treated as radioactive solid waste. Considering the actual application scenarios, leak-proof foundation pits are set up to address equipment failure or emergency shower problems. The device uses neutralizing tanks to handle the acidity and alkalinity of the waste liquid, and is equipped with a detection port to ensure that the waste liquid after condensation meets the emission standards.

Rare earth elements are extensively utilized in defense technology and high-tech industries, leading to a surge in global demand. However, conventional recoverable rare earth resources are limited, necessitating the development and utilization of associated rare earth resources. Taking a sandstone-type uranium deposit in Ili Basin, Xinjiang as the research subject, using inductively coupled plasma mass spectrometry (ICP-MS) to test the content of rare earth elements and analyze their occurrence states. The differences in rare earth element occurrence under various particle sizes of uranium minerals are systematically investigated. The results show that associated rare earth element content within the Ili Basin's sandstone-type uranium deposit ranged from 85.48×10^-6 to 221.17×10^-6. The overall REE curve using average values derived from Australian Shale demonstrate a right-upper inclination trend, indicating fractionation between heavy and light rare earth elements during uranium mineralization processes. Fine ores with particle sizes <0.425 mm exhibite a higher propensity for REE occurrence. The correlation between rare elements such as Mo and Sc and rare earth elements is significant, and both have similar ore-forming environments. The mass concentration of rare earth elements in the production liquids of in-situ leaching site for uranium range from 23.92 to 26.03 mg/L, which reached the standard of recycling and use. The content of rare earth elements in the production liquid is medium rare earth elements, light rare earth elements and heavy rare earth elements from high to low. It is feasible to recover rare earth elements from sandstone uranium deposits using acid in-situ leaching process.

To investigate the influence of factors such as slope ratio, rainfall intensity, and protection form on the erosion characteristics of dam slopes, a full-scale artificial rainfall simulation test platform was built to simulate the runoff and sediment production processes on slopes with different slope ratios, rainfall intensities, and protection forms. The differences in runoff and sediment production on slopes under different operating conditions were analyzed. The results show that the starting time of runoff decreases with the increase of slope ratio and rainfall intensity, and is greatly affected by rainfall intensity. The trench drainage protection, due to the formation of slope runoff channels, advances the time of runoff generation. The grass planting and gravel slope protection on the slope reduce the response of the slope to rainfall to varying degrees, resulting in a significant delay in the start time of runoff generation. The rainfall intensity has the most significant impact on the runoff yield. The rainfall intensity increases from light rain to rainstorm, with the runoff yield increasing by 7.6 times, the sediment yield increasing by 18.5 times, and the time for the sediment yield to reach its peak reduced by about 75%. The ditch drainage slope protection forms a runoff channel, resulting in the maximum flow rate on the slope surface. The grass planting slope protection and gravel slope protection can significantly reduce the sediment yield on the slope surface. The grass planting slope protection improves the physical and chemical properties of the soil to a certain extent, increases the porosity of the soil, increases the infiltration rate of the slope surface, makes the runoff process smoother, and has better erosion resistance.

The data related to mine resource reserves is extensive and subject to frequent changes. Traditional resource reserve management system typically handles singular data types, which leads to low operational efficiency among personnel and increases the risk of data loss. In response to national policies advocating for the use of three-dimensional digital methods in resource reserve management, and to address the challenges in controlling resource reserve data caused by weak informatization equipment and infrastructure, the Shitoumei No. 1 open-pit coal mine in the Santanghu mining area of Hami, Xinjiang, has actively promoted intelligent construction. The mine has implemented a three-dimensional dynamic management system for mineral resource reserves, designed using a combination of B/S and C/S architectures. This system is based on digital three-dimensional models that calculate and display dynamic changes in data during production processes in real time. The resource reserve management system operates on a logical architecture of data storage, platform services, and application services, enabling refined management across mining ledgers, data management and reporting, mining rights, exploration activities, and coal quality. Since its deployment, the system has been performing well, reducing the discrepancy rate in coal output design by 16%, achieving a 100% rate of centralized data control, and significantly enhancing the accuracy of resource reserve evaluations. The system has improved the overall efficiency of resource reserve management and advanced the informatization of open-pit coal mine reserve management within the company.

The in-situ leaching wastewater of uranium mining and metallurgy is characterized by large volume, acidity, and low radioactivity, etc. The evaporation ponds of some uranium mining and metallurgical enterprises cannot meet the demand of expanding production. The advantages and disadvantages of forced evaporation technology such as vacuum evaporation, three-effect evaporation and MVR(Mechanical Vapor Recompression) evaporation were compared and analyzed. It was found that, under long-term use, MVR technology has higher efficiency, lower exhaust emission, and lower energy consumption, making it relatively more suitable for in-situ leaching wastewater. Based on MVR technology, a fully integrated control forced evaporation system was designed and constructed. The heating temperature and material of the equipment were determined according to the waste liquid composition. The on-site device achieved automatic control of temperature, pressure, and liquid level, as well as continuous cyclic evaporation. The actual test shows that the evaporation capacity and efficiency coefficient of the device are positively correlated with the evaporation temperature.

Traditional natural uranium storage relies on manual forklifts or cranes for unloading and storage, which suffers from drawbacks such as high labor intensity, low efficiency, high radiation exposure for personnel, high risk factors, low warehouse utilization, inefficient inventory management, and unreliable accuracy. Aiming to address these issues, an intelligent vertical storage system is designed to achieve efficient vertical storage, automatic retrieval, real-time querying, and rapid inventory.Compared to the stacker crane + shuttle scheme, the dual extended stacker crane scheme offers superior performance in inventory efficiency, network transmission, equipment reliability, maintenance methods, and shelf structure requirements. Based on this analysis of intelligent natural uranium storage processes and optimized equipment selection, a smart vertical storage solution tailored for natural uranium storage has been researched and designed.