The rapid advancement of artificial intelligence has introduced new vitality into cemented paste backfill (CPB) technology. However, current machine learning models for CPB-strength prediction are generally forward-only and single-output, and lack clarity on multi-feature interactions and an integrated full-process design framework. To this end, this study proposes a light gradient boosting machine (LightGBM) model, optimized by Optuna, for predicting the CPB strength at multiple curing ages (3, 7, and 28 d). The model dataset comprised the unconfined compressive strength (UCS) results of 738 CPB specimens prepared with various types of tailings and mix proportions. SHapley Additive exPlanations (SHAP) were employed to elucidate the influence patterns and relative importance of input features while inherently considering their complex, multi-feature interaction effects on the output of the model. Additionally, a simulated annealing (SA) algorithm was integrated with the predictive model to enable the inverse process from the target UCS value to the optimal material-mix proportions. The results demonstrated the effectiveness of Optuna in hyperparameter tuning, leading to an optimized LightGBM model that accurately predicted the multi-age CPB UCS (R² > 0.98). SHAP analysis identified key features, notably the high correlation of the water–cement ratio and CaO content in the tailings with the CPB strength. The SA algorithm effectively provided optimal CPB mix proportions that met the target 28-d UCS value, balancing multiple conflicting objectives such as the solid content and cement dosage. Finally, a user-friendly graphical user interface was developed to integrate these models and provide an accessible, visual, machine-learning-based CPB-strength design tool for mining engineers.

Detection of hazardous arsenic contamination in geological formations is a critical challenge in construction and environmental monitoring. Traditional methods for identifying arsenic-containing zones in boring cores rely on time-consuming leaching tests and expert analyses, which lead to delays and increased tunnel construction costs. This study proposes an advanced approach that combines hyperspectral imaging (HSI) and convolutional neural networks (CNNs) to identify arsenic-containing areas in boring cores rapidly and accurately. Boring cores consisting of mudstone and tuff were analyzed, and arsenic concentrations were measured using a handheld X-ray fluorescence (XRF) analyzer. These XRF measurements served as viable proxies for leaching potential. Hyperspectral images were captured under controlled conditions and preprocessed for CNN training. Initial models using spectral data alone achieved low prediction accuracy (40%–70%) owing to the subtle spectral variations caused by arsenic concentrations. By integrating geological classification with spectral data, the model demonstrated significant improvement, achieving a prediction accuracy exceeding 80%. This study highlights the potential of combining HSI with CNNs for the efficient detection of arsenic-containing areas, thereby offering a scalable, cost-effective, and rapid solution for identifying hazardous zones in boring cores. These findings pave the way for improved environmental safety and reduced delays in construction and resource extraction projects.

This comprehensive review explores the integration of artificial intelligence (AI) in shale gas and oil resource management, with a focus on the application of machine learning and deep learning techniques. Shale gas and oil extraction, once constrained by complex geological and petrophysical challenges, has benefited significantly from AI methodologies, which enhance reservoir characterization, fracture prediction, production forecasting, and optimization of hydraulic fracturing. The review examines the current state of AI applications, identifying key advances in AI-driven models that improve predictive accuracy, address data heterogeneity, and integrate diverse data sources. Special attention is given to the combination of AI with traditional physical simulation models, such as physics-informed neural networks, and the potential for hybrid modeling approaches. Despite these advancements, the review highlights several challenges, including data sparsity, model interpretability, and the need for continuous updates with real-time data. Moreover, the review discusses emerging trends, such as explainable AI and real-time monitoring systems, which offer the promise of improving model transparency and adaptability. Future research directions are proposed to enhance the scalability and robustness of AI models, particularly through the integration of advanced hybrid models, uncertainty quantification techniques, and collaborative efforts across academia and industry, ultimately aiming to foster sustainable and efficient shale resource management.

With the rapid development of industries such as fifth generation (5G), semiconductors, and electric vehicles, the exploitation and utilization of tin resources in China are advancing towards a greater scale, intensification, and standardization. The tin industry is undergoing a critical period of transformation, upgrading, and high-quality development, which will significantly drive the demand for tin. Therefore, ensuring the security of tin resources has become increasingly important. Based on introducing the distribution, current status, production capacity, and supply-demand situation of global and Chinese tin resources, this paper analyzes the future trends in the supply and demand of tin resources and addresses key challenges in the tin industry. Finally, recommendations are proposed to enhance the security of China’s tin resources and offer valuable insights into mining management, development, utilization, and investment.

Fluidized bioleaching is an efficient, environmentally friendly, and cost-effective mining method that has been widely explored and utilized for recovering low-grade copper sulfide minerals, such as chalcopyrite. However, the proliferation and apoptosis of dominant leaching bacteria, such as Acidithiobacillus ferrooxidans, within complex pore, void, and fracture structures in deep-earth environments commonly results in a dynamic bacterial community that evolves continuously. This unclear genetic-scale microbial succession often leads to low leaching reaction efficiency, undesirable reaction passivation, and poor bioleaching operations. This review integrates genetic-scale insights with industrial challenges in chalcopyrite bioleaching, proposing novel strategies for regulating microbial communities. A systematic analysis of five critical dimensions is conducted, focusing on: 1) The adaptations of Acidithiobacillus spp. to high Ag⁺ stress. 2) The direct, indirect, and cooperative bioleaching pathways are linked to bacterial extracellular polymer substance (EPS) and Fe/S oxidation genes. 3) The passivation dynamics governed by bacterial genomics, including thiosulfate, polysulfide, and biofilm mechanisms. 4) The microbial succession patterns under genetic control Hi-C sequencing-guided consortia design. 5) Molecular detection methods (16S rDNA, Hi-C) for optimizing leaching efficiency. The following innovations have been identified as being of key significance: A genomic-environmental interaction model has been developed to bridge the gap between bacterial genetics and passivation dynamics. A comprehensive analysis of Ag⁺ catalysis has been conducted, resulting in a 40% reduction in jarosite formation through jar gene suppression. Practical strategies, such as thermophilic consortia engineering, have been validated in pilot trials, achieving a 32% increase in copper recovery. Additionally, this study meticulously reviews and summarizes typical potential stimulations and enhanced bioleaching methods. The genetic sequencing methods, such as 16S rDNA and Hi-C, have been shown to hold promising potential for improving bioleaching reactions and delaying the formation of passivation substances like jarosite.

The recent resurgence of pneumoconiosis among coal miners in the United States has been linked to their exposure to excessive levels of coal dust. PDM3700 monitors are used in the mining industry to measure each miner’s coal dust exposure levels and control overexposure. However, the high cost of the PDM3700 hinders its use in measuring the exposure levels of all miners. Plantower PMS5003 low-cost particulate matter (PM) sensors can measure coal dust concentrations with high spatial resolution in real-time owing to their low cost and small size. However, these sensors require extensive calibration to ensure a high accuracy over long deployment periods. Because they have only been calibrated for mining-induced PM monitoring using linear regression models, the objective of this study was to leverage machine learning algorithms for calibration of coal-dust-monitoring sensors. Laboratory collocation tests were performed using the PDM3700 and aerodynamic particle sizer as reference monitors in a wind tunnel at a wide range of concentrations (0–3 mg/m³), temperatures (20–32°C), and relative humidities (23%–43%). The results revealed that nonlinear machine learning techniques significantly outperformed traditional linear regression models for low-cost sensor calibration. With the artificial neural network (ANN) being the strongest calibration model, Pearson’s correlation of the PMS5003 sensors reached 0.98 and 0.97, those of the Airtrek sensors reached of 0.89 and 0.91, and those of the GasLab sensors reached 0.93 and 0.92. This shows a 2%–11% improvement in model performance over the linear regression model using ANN calibration. The success of the machine learning algorithms used in this study demonstrates the feasibility of deploying low-cost PM sensors for coal dust monitoring in mines.

Coal spontaneous combustion fires threaten personal safety, increase carbon emissions, release toxic and harmful gases, and cause serious environmental pollution. The study of intelligent early warnings for coal spontaneous combustion can advance fire prevention and control measures, making a meaningful contribution to the ecological and environmental protection in mining areas. To address the limitations in selecting characteristic index gases for coal spontaneous combustion and the low accuracy of traditional temperature prediction and discrimination models, an intelligent identification system was developed. The system integrates laboratory research and analysis, intelligent algorithm optimization, index rationality verification, and field measurement and application, all based on characteristic index gases. By constructing a dynamic discriminant model of coal self-gas temperature, the composite index of coal spontaneous combustion characteristics is further optimized and verified. Model performance was evaluated using root mean square error (RMSE), decision coefficient (R²), mean absolute error (MAE), and mean absolute percentage error (MAPE). The prediction results for three, four, and five parameters were obtained. The results indicate that the R² value was 0.9975 under the conditions of O₂, CO, C₂H₄, and CH₄/C₂H₆, demonstrating the best model performance. The MAE was 1.9272, the RMSE was 2.5114, and the MAPE was 2.0830%. These findings enable optimal selection of self-ignition warning indicators for coal. A comparative analysis of the improved whale optimization (MSWOA-BP), gray wolf optimization (GWO-BP), standard whale optimization (WOA-BP), and particle swarm optimization (PSO-BP) models was performed to verify the universality of preferred feature indicators and the accuracy of prediction models. A comparative analysis between on-site measured temperatures and model-predicted temperatures demonstrated that the model exhibited high accuracy. This research provides a valuable reference for developing on-site coal spontaneous combustion warning systems, enabling efficient prediction and early warning, which are crucial for coal resource safety, efficient mining, and fire prevention.

Backfilling mining goafs with municipal solid waste incineration fly ash (MSWI FA) and industrial solid waste offers novel possibilities for advancing green mining engineering. However, the recycling of MSWI FA using Si–Al-based solid wastes to prepare MSWI FA-based backfill materials (MBM) remains at the laboratory stage, with limited studies addressing the in-depth mechanisms and sustainability of MBM in actual goafs. This study investigated the leaching safety, chemical bonding, and valence states of Si–Al products and microstructure, carbon emissions, and economic feasibility of a multi-component backfill material (MLBM) composed of MSWI FA, blast furnace slag (BFS), steel slag (SS), coal fly ash (CFA), flue gas desulfurized gypsum (FGDG), and lead–zinc tailings. The binder weight ratio was MSWI FA : BFS : SS : CFA : FGDG = 10:33.6:33.6:20:12.80 under engineering application conditions. The toxicity characteristic leaching procedure test showed that the anion and heavy metal leaching concentrations from MLBM were below the Class II thresholds specified in the Standard for Groundwater Quality (GB/T 14848–2017). The hydration mechanism involves: (i) Cl^– competition that accelerates and promotes the generation of ettringite and Friedel’s salt; (ii) MSWI FA-induced depolymerization and repolymerization of the Si–Al vitreous network, facilitating C–(A)–S–H gel formation with high average connectivity degree/mean chain length (D_c/L_MC) ratio and a low Ca/Si atomic ratio (1.27), resulting in an increased content of Si–O–Si (Q² p and Q2 b) and Si–O–Si (Q3) (bridging oxygen) in the C–(A)–S–H gel; (iii) pore-filling effect by various functional hydration products that increases the proportion of gel pores (3.5–100 nm) in MLBM to 47.31vol% and reduces the porosity to 40.20vol%. Compared with that of traditional cement materials, the production of MLBM led to reduced material costs by 80% and CO₂ emissions by more than 98%. These findings contribute to the sustainable development of green mining, enhance the efficiency of solid waste management, reduce environmental pollution, and support global carbon reduction targets.

The treatment of post-mining goafs presents a significant challenge for the coal industry, grappling with environmental safety and carbon neutrality goals. To address issues such as abandoned goafs, high carbon emissions, and conflicts in backfill operations during coal mining, this study proposes a CO₂ mineralization grouting method that targets voids in post-mining goafs. A two-stage mineralization process was designed with a focus on in-situ goaf grouting using open-pore materials. Grouting materials were prepared from carbide slag and fly ash, with cetyltrimethylammonium bromide (CTAB) as a surfactant and oleic acid acts as a foam stabilizer to regulate the pore structure alongside hydrogen peroxide (H₂O₂) used to regulate the pore structure. Scanning electron microscopy, nuclear magnetic resonance, unconfined compressive strength tests, and random forest regression identified the optimal mix: CTAB = 1.2 g, oleic acid = 5.0 g, H₂O₂ = 1.0 g, water-to-solid ratio = 0.7, Si/Al and Na/Al molar ratios are controlled at 3.0 and 2.7, respectively. Results from custom-built testing devices showed that open-pore materials doubled the mineralization depth compared with closed-pore materials. The method demonstrates technical feasibility and economic viability, while emphasizing the importance of rigorous risk management to ensure environmental and operational sustainability. Compatible with existing green mining systems, it offers economic and environmental benefits, providing a novel approach for low-carbon coal mining.

Lepidolites are important carriers of Li ores. Because of the strong positive correlation between the presence of Li and F, Li extracted from lepidolite must be defluorinated. Moreover, synergistically extracting Li, Rb, and Cs is important for the comprehensive utilization of lepidolite. Steam roasting converts F to HF without introducing impurities. In this study, the effects of steam roasting on Li activation and the synergistic extraction of Rb and Cs were investigated during the defluorination process. Notably, the defluorination mechanism and migration behavior of F during steam roasting were revealed based on qualitative and quantitative analyses of the liquid and corresponding solid phases. Furthermore, density functional theory calculations were used to investigate the pathways and behaviors of defluorination and the removal behavior of F atoms connected to Al and Li sites, revealing that the F atoms on the (010) crystal surface were easily replaced by water molecules, thereby releasing surface F atoms. Under the optimal roasting conditions, the defluorination efficiency of lepidolite was 91.08% at 880°C for 1 h. Subsequently, the Li encapsulated in the crystal lattice, as well as Rb and Cs, was activated and leached using optimal sulfuric acid leaching. The leaching efficiencies of Li, Rb, and Cs were 97.37%, 97.77%, and 97.19%, respectively at 120°C for 2 h. This study provides a comprehensive and systematic perspective on the defluorination and comprehensive utilization of lepidolite, providing a comprehensive, clean, efficient, green, and low-C lepidolite Li extraction process.