In traditional in situ leaching (ISL) uranium mining, the injection volume depends on technicians’ on-site experience. Therefore, applying artificial intelligence technologies such as machine learning to analyze the relationship between injection volume and leaching rate in ISL uranium mining, thereby reducing human factor interference, holds significant guiding importance for production process control. This study proposes a novel uranium leaching rate prediction method based on a CNN–LSTM–LightGBM fusion model integrated with an attention mechanism. Ablation experiments demonstrate that the proposed fusion model outperforms its component models across three key metrics: Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE). Furthermore, comparative experiments reveal that this fusion model achieves superior performance on MAE, MAPE, and RMSE metrics compared to six extensively utilized machine learning methods, including Multi-Layer Perceptron, Support Vector Regression, and K-Nearest Neighbors. Specifically, the model achieves an MAE of 0.085%, an MAPE of 0.833%, and an RMSE of 0.201%. This attention-enhanced fusion model provides technical support for production control in ISL uranium mining and offers valuable references for informatization and intelligentization research in uranium mining operations.

The drilling core sampling and chemical analysis method for the quantitative determination of solid mineral deposits has several drawbacks, including a low core drilling efficiency, a high core sampling cost, and a long chemical analysis cycle. In current uranium quantification practices, advanced techniques have been developed to preliminarily determine the formation of uranium content based on the interpretation results of natural $\gamma$-ray total logging. However, such methods still require supplementary core chemical analysis to derive the uranium–radium–radon balance coefficient, which is then used for equilibrium correction to obtain the true uranium content within the uranium-bearing layer. Furthermore, conventional prompt neutron time spectrum logging is constrained by low count rates, resulting in slow logging speeds that fail to meet the demands of practical engineering applications. To address this, this study proposes a uranium quantification method that corrects the natural $\gamma$-ray total logging using prompt neutron time spectrum logging. Additionally, a calibration parameter determination method necessary for quantitative interpretation is constructed. Experimental results from standardized model wells indicate that, in sandstone-type uranium deposits, the absolute error of uranium content is within $\pm 0.002\%\mathrm{eU}$, and the relative error is within $\pm 2.5\%$. These findings validate the feasibility of deriving the uranium–radium–radon balance coefficient without relying on core chemical analysis. Compared with the prompt neutron time spectrum logging method, the proposed approach significantly improves the logging speed while producing results that are essentially consistent with those of natural $\gamma$-ray total logging. It provides an efficient and accurate solution for uranium quantitative interpretation.

Effective protection of high-value assets against electro-optical guided missiles requires precise spatiotemporal coordination of smoke-screen deployment. This paper develops a UAV-assisted deployment framework that integrates a multi-stage kinematic model, a time-varying line-of-sight (LOS) occlusion metric, and a hierarchical genetic algorithm (H-GA) for joint optimization of UAV trajectories, release schedules, and task assignments. Across five scenarios, cooperative five-UAV deployment against three simultaneous missiles achieves 14.44 s effective shielding, compared with 1.42 s under fixed-parameter deployment. Sensitivity analysis indicates strict operational constraints: heading errors of 2–3° reduce shielding by over 50%, and the feasible release window is limited to 0.5–0.8 s. These results provide quantitative guidance for navigation accuracy and timing synchronization in practical smoke-screen defense systems.