This paper summarizes the application and award of research funding in the Mining and Metallurgical Engineering Discipline at the Department of Engineering and Materials Science of the National Natural Science Foundation of China in 2026. A detailed analysis is conducted from the following four research fields: (1) petroleum engineering; (2) mining engineering; (3) safety science and engineering; (4) metallurgy and material processing engineering. This paper also introduces the development strategies of the discipline and proposes the next steps for continuous improvement.
As pavements age,sustained traffic,environmental degradation,and material aging lead to increasing occurrences of cracks,potholes,and other distresses,threatening safety and infrastructure longevity.Textured road networks,combined with varying illumination,narrow geometries,and low-contrast cracks,make reliable detection challenging,leading to missed detections and false positives.Although deep learning has advanced automated pavement assessment,common lightweight single-stage detectors struggle to balance accuracy and speed on edge devices such as vehicle-mounted systems and unmanned aerial vehicles(UAVs).This is due to the f act that repeated downsampling,fixed interpolation,and coarse cross-scale fusion blur crack boundaries and weaken texture cues,thereby reducing recall and degrading localization.To address these challenges,this study introduces YOLO-RDD,a lightweight,task-specific pavement distress detection algorithm derived from YOLOv11n architecture.Designed explicitly for the intrinsic characteristics of pavement distress,including small spatial extent,high aspect ratio,low-intensity contrast relative to the surrounding pavement,and susceptibility to masking by heterogeneous background textures,YOLO-RDD achieves a principled balance between detection accuracy and real-time deployability on edge platforms.This method systematically strengthens the feature fidelity,cross-scale alignment,and context-aware prediction via targeted architectural refinement across the backbone,neck,and detection head.First,in the backbone network,the original C3k2 modules are replaced with a structurally reparametrized RepGELAN module.RepGELAN integrates progressive intralayer feature aggregation with cross-stage information feedback,thereby expanding the effective receptive fields while preserving high-frequency edge responses and weak texture discriminability,which are critical for detecting narrow,low-contrast cracks.Structural reparameterization enables the equivalent conversion of its multibranch training-time architecture into a single standard convolution at inference time,ensuring minimal computational overhead without sacrificing representational expressiveness.Second,for the neck network,we propose DySlim-Neck,a lightweight,semantic-aware fusion architecture that synergistically combines a dynamic upsampling unit(DSU)with depthwise-separable convolutions.Before cross-layer fusion,the DSU performs content-adaptive alignment,explicitly correcting geometric misalignment and suppressing over-smoothing artifacts commonly induced by fixed-resolution interpolation and naïve concatenation.Coupled with mixed convolutional aggregation,DySlim-Neck significantly reduces the fusion-induced latency and memory footprint while maintaining high-fidelity propagation of small-object features across scales,thereby safeguarding fine crack morphology and continuity.Third,the detection head adopts DynamicHead,which dynamically reweighs and reorganizes multiscale features along scale,spatial,and channel dimensions based on target-specific semantic cues.This adaptive coupling explicitly addresses the classification-regression decoupling problem prevalent in small,weakly contrasted targets,enhancing localization confidence and scale-invariant prediction robustness under heterogeneous pavement backgrounds.Furthermore,an embedded dynamic activation mechanism selectively suppresses background clutter responses while amplifying discriminative crack signatures,yielding higher precision in ambiguous regions.The proposed algorithm is rigorously evaluated on the RDD2022 benchmark dataset using four complementary metrics:F1-score(harmonic mean of precision and recall),mean Average Precision(mAP)at IoU thresholds of 0.5 and 0.5∶0.95,parameter count(in millions),and computational cost measured as GFLOPs and inference latency(ms)on an embedded Jetson AGX Orin platform.Experimental results demonstrate that,relative to the YOLOv11n baseline,YOLO-RDD achieves a 19.7%reduction in parameter count while delivering consistent gains across detection robustness metrics—improving F1-score by 1.0 percentage point and mAP@0.5 by 2.3 percentage points—without compromising mAP@0.5∶0.95.indicating enhanced localization consistency under varying overlap criteria..Notably,ablation-guided analysis confirms that these improvements are particularly pronounced for fine-scale distress:the detection recall for narrow cracks(<2 pixels wide)increases by 8.6%,and the pothole localization accuracy(measured by the bounding-box IoU)improves by 5.4%,especially under low-illumination conditions and highly textured asphalt surfaces.YOLO-RDD excels in pavement defect detection,particularly in narrow cracks.However,the stability decreases under real-world challenges such as uneven lighting,dirt,varying camera angles,and limited training diversity.The next step involves:(1)expanding real-world road data across times,road types,and capture conditions,as well as augmenting the generative crack/damage samples to boost generalization;and(2)adopting transfer and continual learning for cross-region adaptation and fusing multi-source data into a pavement health assessment framework to accelerate engineering deployment.
With the rapid advancement of artificial intelligence, large language models (LLMs) have emerged as a promising technological paradigm for accelerating the intelligent transformation of the steel industry. Owing to their capabilities in semantic understanding, logical reasoning, knowledge organization, and cross-task generalization, LLMs offer a promising new pathway for addressing the limitations of traditional industrial artificial intelligence approaches, which are often restricted to isolated processes, narrowly defined tasks, and weak cross-process coordination. Against this background, the architectural construction and application logic of LLM-driven industrial intelligent agents for the steel industry have attracted increasing attention, particularly in relation to the development context, core capabilities, adaptation strategies, and representative application paths of industrial large models across the full steel production process. Starting from the evolution of mainstream LLM architectures and their industrial adaptation mechanisms, current progress shows how general-purpose models can be transformed into domain-oriented industrial large models through prompt engineering, retrieval-augmented generation, fine-tuning, and domain pretraining. In contrast to general LLMs designed for open-domain language tasks, industrial large models are fundamentally oriented toward constrained, verifiable, and operationally grounded decision support. On this basis, the system construction framework of steel-industry large models can be understood as a layered architecture composed of foundation models, domain enhancement modules, and scenario application modules. In particular, three major fusion modes for integrating process knowledge, industrial data, and model reasoning in steel production can be identified: serial fusion, parallel fusion, and embedded fusion, thereby clarifying the structural route through which LLMs may be incorporated into highly coupled metallurgical processes. Considering the characteristics of the steel industry, including dense multisource heterogeneous data, strong process constraints, complex operating conditions, and long decision chains, current research and industrial practice can be organized around three interrelated dimensions: data processing and dataset construction, fault diagnosis and root-cause analysis, and process scheduling optimization. In the first dimension, emphasis is placed on converting structured sensor streams, equipment logs, technical manuals, standards, and literature into domain-specific corpora and multimodal datasets suitable for supervised fine-tuning, reasoning tasks, and downstream evaluation. In the second dimension, LLMs—often combined with knowledge graphs, causal analysis, and multiagent reasoning—support the interpretable diagnosis of static defects, dynamic deviations, and systemic failures across the steelmaking, casting, and rolling processes. In the third dimension, LLM-related techniques help bridge perception, analysis, and execution for static and dynamic scheduling problems under multiple objectives, such as energy consumption, production rhythm, quality stability, and resource allocation. The role of LLMs in the steel industry has evolved from auxiliary analytical tools toward system-level intelligent decision-support hubs. Nevertheless, broader deployment still faces critical challenges, including multimodal data alignment under physical constraints, hallucination risks in high-stakes industrial settings, limited interpretability of model reasoning, and the difficulty of integrating multidisciplinary knowledge from metallurgy, materials science, control engineering, and artificial intelligence. Future development is expected to focus on high-quality industrial corpus construction, stronger physical consistency and explainability, multi-agent collaboration, industrial internet integration, and lightweight real-time deployment. Overall, this line of inquiry provides a structured reference for the systematic research, engineering implementation, and paradigm evolution of LLM-driven intelligent agents in the steel industry.
Metal-organic frameworks(MOFs)hold great promise for hydrogen storage owing to their high specific surface areas,tunable pore structures,and chemical modifiability.Mg-MOF-74 has attracted considerable attention because of its unique structural features:Mg ions coordinate with five oxygen atoms and one solvent molecule,and open metal sites(OMSs)are generated upon activation to remove the solvent.This material possesses a highly polarized pore environment,offering inherent potential for hydrogen storage.However,existing studies have shown that the hydrogen uptake of Mg-MOF-74 remains below 0.15%weight present at 298 K and 7 MPa,and its practical application is severely limited by sluggish H2 diffusion kinetics.Introducing noble metal components can trigger the hydrogen spillover effect,in which H2 molecules dissociate into H atoms and migrate to the support,significantly enhancing the storage capacity.To address the insufficient adsorption sites and stability issues of MOF-based hydrogen storage materials,a series of Pt-Mg-MOF-74-R composites with varying Pt loadings and defect levels were fabricated via in situ Pt introduction using K2PtCl4 as t he Pt source,followed by thermal reduction in an Ar/H2 atmosphere with Mg-MOF-74 as the host.Systematic characterization by XRD,FT-IR spectroscopy,SEM,TGA,TEM,and XPS revealed the effects of the reduction temperature and Pt loading on the framework defect formation and metal valence states.The thermal reduction process effectively reduced the Pt ions to highly active Pt0 nanoparticles while removing solvent molecules and inducing partial ligand loss,leading to the construction of abundant MgO5 open metal sites.The FT-IR results showed that the Pt-loaded materials exhibited peak positions similar to those of pristine Mg-MOF-74,with no new functional groups detected,indicating that Pt was predominantly present in a physically loaded state within the n-Pt-Mg-MOF-74 framework.Hydrogen storage tests showed that 4-Pt-Mg-MOF-74-R200 achieved a gravimetric uptake of 0.35%weight present at 298 K and 10 MPa,which was 2.5 times that of pristine Mg-MOF-74 under identical conditions.The adsorption enthalpy reached-22.35 kJ·mol-1,3.8 times higher than that of the original Mg-MOF-74,with the thermal reduction process playing a key role.Mechanistic analyses indicated that thermal reduction increased the content of well-dispersed Pt0,facilitating hydrogen spillover by dissociating H2 molecules into H atoms.Simultaneously,thermal reduction removed linkers without collapsing the framework,increasing the number of MgO5 open metal sites available for hydrogen binding and optimizing the mass transfer pathways and coordination environment,thereby providing more transport routes and adsorption sites for the H atoms generated via spillover.The two mechanisms reinforce one another and collectively enhance the room-temperature hydrogen storage capacity of 4-Pt-Mg-MOF-74-R200.Notably,increasing the reduction temperature from 200℃to 300℃caused partial pore collapse,a decrease in the specific surface area,and a significant increase in the mesoporosity.These structural changes adversely affected the hydrogen storage performance,with 4-Pt-Mg-MOF-74-R200 clearly outperforming 4-Pt-Mg-MOF-74-R300.Although thermal reduction generates highly active defect sites that substantially enhance the hydrogen adsorption capacity,excessive temperatures compromise the pore architecture.The balance between defect engineering and structural integrity indicates that an optimal reduction temperature of 200℃maximizes the hydrogen storage performance by preserving an adequate pore structure and surface area while optimizing the quantity and distribution of defect sites.This study primarily reveals the synergistic mechanism of"hydrogen spillover+defect sites"in MOF-based hydrogen storage.
In constructional backfill mining,variations in the heights of the roof and coal seam lead to the formation of roof-backfill composite load-bearing structures with different height ratios between the backfill columns and roof at different backfilling positions.Such dimensional variations lead to performance differences between the roof and backfill,which in turn alter the load-bearing and failure characteristics of the composite structures,ultimately impacting the effectiveness of constructional backfill mining.To address this issue,rock-backfill composite specimens with height ratios ranging from 1∶3 to 3∶1 were prepared.Uniaxial compression tests were conducted,and the failure characteristics were monitored by combining digital image correlation(DIC)technology with acoustic emission(AE)monitoring.A mechanical model was established to analyze the synergistic-bearing characteristics and their underlying mechanisms.The results showed that as the height ratio increased from 1∶3 to 3∶1,the strength of the rock-backfill composite increased from 5.60 to 11.71 MPa.When the height ratio was greater than or equal to 1∶1,the strength of the composite was higher than that of the single cemented gangue backfill specimen(8.96 MPa),exhibiting a synergistic strength enhancement effect.The failure mode transformed from splitting failure to tensile-shear mixed failure,and the failure region propagated from the local backfill region to the entire composite,thereby achieving synergistic failure.Analysis of AE data revealed that the failure mode transitioned from the sole failure of the backfill to the combined failure of the sandstone and backfill owing to the variation in the height ratio of the specimens.When the height ratio was less than 1∶1,no synergistic failure occurred,and the cumulative AE ringing count increased gradually;when the height ratio was greater than or equal to 1∶1,synergistic failure occurred,and the failure of the sandstone caused a sudden surge in the cumulative AE ringing count.The average failure intensity of the specimens increased with the height ratio;however,the growth rate decreased when the height ratio exceeded 2∶1.From the AE location points and DIC strain contours,it was found that stress concentration occurred easily at the interface when the height ratio was less than 1∶1,leading to premature failure of the composite.Moreover,the failure position of the specimens shifted from the backfill to the sandstone part as the height ratio increased.Based on the interface bonding constraint force coefficient and Drucker-Prager strength criterion,a synergistic failure criterion was established.The interface constraint effect reduces the strength of the sandstone while increasing that of the backfill,thereby enhancing the overall strength of the composite.The critical height ratio for synergistic failure was calculated to be 1∶1.07.When the height ratio exceeded this value,the strength of the composite was higher than that of sandstone at the interface,and the failure in the backfill propagated into the sandstone.The bearing capacity of the rock-backfill combined body increased with the height ratio,and all specimens exhibited collaborative failure.This behavior enables full utilization of the surrounding rock's bearing capacity,thereby satisfying the requirements of constructional backfill mining.These results support the selection of reasonable backfilling positions for constructional backfill mining.In specific engineering practices,it is necessary to comprehensively consider potential influencing factors such as interface parameters and geological conditions.The critical value should be appropriately modified and optimized according to specific engineering conditions,and the rationality of the layout of key positions of the roof-backfill composite load-bearing structure should be evaluated for constructional backfill mining.This enables the structure to adapt precisely to engineering practices and enhances the practical applicability of the research findings.
