During the smelting process of high-grade pipeline steel X80, the control of large inclusions is one of the key factors to ensure its performance stability and engineering applicability. Through intensive sampling and systematic analysis in the LF（ladle furnace）-VD（vacuum degasser） refining process, it was found that large inclusions of pipeline steel X80 concentrated after VD calcium treatment, and the main types were C₁₂A₇（12CaO·7Al₂O₃） and CA₂（CaO·3Al₂O₃）. In the soft blowing stage, the large inclusions were not effectively removed, and the number of large inclusions instead showed an increasing trend. Therefore, it is judged that the current VD soft blowing process control is unreasonable, resulting in the aggregation and growth of inclusions due to insufficient stirring dynamics to form large inclusions. Combined with the parameters of ladle equipment and process control standards, the three-dimensional multi-physical field numerical simulation of the VD soft blowing stage of pipeline steel was carried out. A comprehensive analysis was carried out from aspects such as flow field distribution, inclusion migration paths, and the stirring intensity of molten steel. It is pointed out that the original soft blowing process cannot form an effective inclusion migration channel under the current metallurgical reaction conditions. Combined with the simulation results, a new VD soft blowing control scheme was proposed. The soft blowing flow rate was increased from 50 L/min to 60 L/min. While ensuring the uniformity of the molten steel temperature, the internal convection intensity of the molten steel and the floating efficiency of inclusions were significantly enhanced. After implementing the optimized soft blowing process, verified by industrial tests, the large inclusions number density in the molten steel at the end of soft blowing decreased from 0.074/mm² to 0.020/mm², and the large inclusions number density in the pipeline steel decreased from 0.071/mm² to 0.040/mm², significantly improving the quality of the molten steel and the qualified rate of flaw detection for pipeline steel. This research achievement provides an effective technical path for achieving fine control of inclusions in the LF-VD refining process of high-grade pipeline steel X80, and also offers a reference for the subsequent optimization of clean smelting processes for other high-end steel grades.

Longitudinal off-corner cracks on the wide face of stainless steel slabs were found in a domestic steel plant. The formation mechanism of longitudinal off-corner cracks on the wide edge of continuous casting slabs was revealed through laboratory testing and analysis, and stress simulation. The grains near the longitudinal off-corner cracks were coarse, and there was a layer of pre-precipitated ferrite with a thickness ranging from tens to hundreds of microns at the original austenite grain boundary, and the cracks mainly occurred along the filmy pre-precipitated ferrite. The characteristic element Na of the mold powder was detected inside the longitudinal off-corner cracks on the wide face of slabs, therefore it was determined that the longitudinal off-corner cracks were generated in the mold. When the distance from meniscus was greater than 0.2 m, the stress on wide surface of continuous casting slab increased with the distance from corner. The maximum value was reached near the corner, and then the stress value gradually decreased and tended to be stable. The corners of continuous casting slab were subjected to two-dimensional cooling, resulting in a large solidification shrinkage and led to the appearance of gaps, which weakened the cooling intensity of the corners and caused the thinnest point of the solidified shell to appear near corner on the wide face of slab. Due to the solidification shrinkage, there was maximum stress near the edge of continuous casting slab. When the taper of mold could not completely match the solidification shrinkage, depressions or cracks would appear at the thinnest point of solidified shell at the wide face edge of continuous casting slab. The formation mechanism of longitudinal off-corner cracks on the wide face of stainless steel slabs is revealed, and a theoretical basis for controlling surface defects in continuous casting slabs is provided.

The efficient utilization of semi-coke is of great significance in realizing the hierarchical conversion and stepwise utilization of coal resources. The blast furnace injection performance of five semi-cokes was analyzed in depth, including proximate analysis, elemental analysis, ash composition and other key indexes, as well as the combustion characteristic kinetics, XRD and industrial application of the coke. The results show that, relative to the remaining 4 semi-cokes, semi-coke E has the lowest ash and volatile matter content and the highest fixed carbon content. The order of high and low content of alkaline metal oxides in semi-coke ash is consistent with the order of high and low content of CaO. The excessive alkaline metal oxides in the ash may affect the integrated combustion characteristic index S value. The abradability of semi-coke is negatively correlated with its ash content. The ash content of semi-coke E is the lowest and the abradability is the best with 80.21. The ignition point of semi-coke is in the temperature range of 573-673 K. It is non-explosive, which meets the requirements of safe production in the blast furnace of the iron and steel enterprises. The combustion performance of semi-coke is between that of bituminous coal and anthracite. Semi-coke B has the highest S value of and the best combustion performance. The kinetic curves of different samples has good linear relationship, in which the lowest combustion activation energy of bituminous coal is 22.86 kJ/mol, and the highest combustion activation energy of semi-coke E is 69.65 kJ/mol. The combustion performance is affected by the activation energy and the pre-finger factor together. A small amount of semi-coke B is introduced into the blast furnace injection process for the test. The results show that the fuel ratio of the blast furnace is reduced and the utilization efficiency of the blast furnace is improved after adding semi-coke B. However, after further elevating the proportion of semi-coke injection, it is found that the fuel ratio of the blast furnace gradually increases and the gas utilization efficiency gradually decreases.

During steady casting, the flow and temperature field inside the tundish and mold are in dynamic equilibrium. While unsteady processes such as casting start, ladle change, and casting end are unavoidable, disturbances in the flow field can have adverse effects on the removal of inclusions. With the increasing cleanliness requirements for high added value steel, the cleanliness of molten steel and the migration behavior of inclusions in unsteady casting processes have become key factors restricting product quality. The unsteady process causes severe fluctuations in the flow and temperature fields, leading to steel reoxidation, slag entrainment, and obstruction of inclusions floating up, significantly increasing the inclusion content in the first slab, transition slab, and last slab. A large number of unsteady process slab are downgraded or even scrapped, significantly increasing production costs. A systematic review is conducted on the migration behavior of non-metallic inclusion particles in molten steel in typical metallurgical reactors during the unsteady process of continuous casting. The sources and migration behavior of inclusions in the unsteady process are summarized, as well as their impact on the cleanliness of castings. By integrating numerical simulations, physical experiments, and industrial testing data, the three-dimensional distribution pattern of inclusions are systematically revealed and key control technologies are extracted, which provides theoretical support for accurately defining the range of unsteady slab and reducing degradation and scrap rates. The existing control technology can improve the impact of unsteady processes on the cleanliness of molten steel, but there are still problems such as insufficient model accuracy, poor universality of technical solutions, and scarce real-time monitoring and dynamic control capabilities. The combination of multi-physics mechanism and intelligent control technology will be the future trend.

Digital and intelligent technologies are driving the intelligent transformation of blast furnace ironmaking in China, emerging as a kind of new quality productive forces. Nowadays, using the universal large language models （U-LLMs） as a base framework to build industrial vertical large language models （V-LLMs） for guiding industrial production via secondary training with domain-specific corpora has become a new trend. Despite the emergence of V-LLMs for the entire steelmaking process, research on V-LLMs specifically for blast-furnace operations is still in its infancy. By reviewing the recent evolution of blast furnace ironmaking intelligence, a new idea of its paradigm reconstruction and fusion driven by V-LLMs was proposed. Classifying the blast furnace V-LLM tasks scenario into scheduling and decision making, the "Data-Application-Perception" penetration and application path is presented for the first time. Also, 5-dimensional evaluation system for future performance assessment and optimization is puts forward, including process understanding, safety and reliability, knowledge transfer, real-time performance, and continuous learning. Then, 3 kinds of new paradigms for intelligent upgrade driven by blast furnace V-LLMs are discussed, including blast furnace condition characterization, blast furnace condition metaverse, and multi-scene fusion. A 3-dimensional collaborative deep representation architecture of "Physical↔Virtual↔Perception" with blast furnace V-LLMs as the core and a new concept of “blast furnace portrait” are proposed. The construction route of blast furnace condition metaverse and the policy of multi-scene fusion are sorted out and discussed. Finally, the key issues and possible solutions in the future development of blast furnace V-LLMs are analyzed. Focused on the feasibility of blast furnace V-LLMs in construction, application, and evaluation, the paradigm reconstruction of intelligence update driven by V-LLMs in the context of industry development is discussed. It aims to offer theoretical guidance for the deep application of V-LLMs in China's blast furnace ironmaking and further promote its intelligent transformation and development.

The catalytic purification of carbon monoxide （CO） from sintering flue gas in the steel industry has attracted increasing attention in recent years due to its unique advantages in energy conservation and environmental protection. However, challenges such as catalyst poisoning and barriers to large-scale implementation have so far prevented the realization of practical engineering applications both domestically and internationally. An industrial-scale CO catalytic purification project implemented on a 435 m² sintering machine at Handan Steel was reported, handling the full flue gas volume of 1.6×10⁶ m³/h （standard condition, wet basis）. The project employed noble metal honeycomb catalysts loaded into the spare layer of the original denitrification （DeNO_x） tower, achieving CO reduction without the need for additional external equipment. Moreover, the heat released from CO oxidation significantly reduced the consumption of coke oven gas required for flue gas reheating during catalytic denitrification. Results from 4 months of operation show stable overall performance, with CO catalytic conversion efficiency ranging from 76% to 85%, CO emission concentrations between 1 070 and 2 365 mg/m³ （well below the current environmental limit of 2 800 mg/m³）, flue gas temperature increase of 33-55 ℃, and a coke oven gas saving rate of 63% to 100%. Based on continuous monitoring data, it analyzed how flue gas temperature, flow rate, and the presence of other pollutants affect the CO catalytic performance. Special attention was given to system performance during key operational phases, including initial start-up, temporary shutdowns, and switching of flue gas circulation. The findings indicate that the current system is resilient to operational fluctuations and is capable of achieving both compliant CO emissions and complete coke oven gas savings. This translates to an energy consumption reduction of 3.4 kg of standard coal per ton of sintered product. An engineering demonstration system for catalytic purification of CO in sintering flue gas was established, providing an industrial-scale practical paradigm for synergistic multi-pollutant control in the steel industry. Furthermore, it offers a technical foundation for advancing the coordinated optimization of energy conservation, emission reduction, and cleaner production in sintering processes.

After the cutting process of 20MnCrS5 gear steel produced by a steel plant, obvious bright line defects are found on the surfaces of several gear blanks, which are mainly caused by inclusions. SEM （scanning electron microscopy） and EDS （energy-dispersive spectroscopy） were used to identify the basic characteristics of inclusions at the defect sites, such as their morphology and composition. By sampling during the steelmaking process of 20MnCrS5 steel, analyzing the slag composition at different stages and the morphology and composition of inclusions in steel samples, and combining thermodynamic calculations, the evolution of inclusions during steelmaking was analyzed to determine the formation mechanism of defect-causing inclusions. The results show that the defects on gear blanks were primarily large-sized CaO-MgO-Al₂O₃ inclusions, with the maximum size exceeding 200 μm. These inclusions contain no SiO₂, exhibit a uniform distribution of CaO and Al₂O₃, and have locally concentrated MgO, this type of large-sized inclusions is formed by the agglomeration of numerous small-sized CaO-MgO-Al₂O₃ inclusions. The inclusions originate from the reaction of abundant Al₂O₃ inclusions originally present in the molten steel with Mg and Ca during steelmaking. Such inclusions mainly exist in liquid state at high temperatures, making them difficult to float and remove, thus remaining in the molten steel. Small-sized CaO-MgO-Al₂O₃ inclusions, after passing through the nozzle along with the molten steel, gradually precipitate solid phases during the subsequent temperature drop and solidification process. These solid phases adhere to each other due to the effect of cavity bridge force and complete sintering within a short time after contact. Such small-sized inclusions collide and agglomerate in this way, eventually forming large-sized aggregated inclusions. To reduce the formation of such large-sized inclusions, the cleanliness of molten steel needs to be further improved. The slag composition should be adjusted to enhance its ability to absorb Al₂O₃ inclusions in the molten steel,and the feeding quality of calcium wire during Ca treatment should be reasonably controlled to minimize the formation of liquid CaO-MgO-Al₂O₃ inclusions. These research results are of great significance for addressing the surface defect issue of 20MnCrS5 gear blanks and further enhancing the product quality of 20MnCrS5 steel.

CO₂ emissions rapidly growth have lead to the deterioration of the global environment. The steel production industry has been one of the major CO₂ emitters, accounting for 7% of global CO₂ emissions, of which 70% is emitted in the iron-making process. Currently, as a technology that has attracted much attention and development, electrochemical reduction has advantages such as easy control of the reaction process and high energy efficiency, which provids a potential low-carbon production approach for the steel industry. The research progress of iron making by electrochemical reduction, and the influence parameters of electroreduction reaction were discussed. According to the different properties of electrolytes, the electrochemical reduction of iron-compound can be divided into molten salt systems, acid-alkaline aqueous solution systems and ionic liquid systems, and each system has its advantages and disadvantages. Due to the strong compatibility characteristics of electrolytes, molten salt systems can directly use iron ores as raw materials, which undoubtedly has a huge advantage in reducing costs, but higher reaction temperature and the corrosive of electrolytes are not conducive to long-term operation of equipments. Alkaline solution systems is under development stage, which has the advantages of mild electrolytic conditions and smaller hydrogen evolution side reaction. Electrolytic iron production in acidic solution system has been used commercially, but the competitive hydrogen evolution side reaction promoted by high concentration of hydrogen ions will cause the decrease of current efficiency, which is the main problem at present. Compared with electrolysis technologies in other systems, acidic solution system has greater development prospects. The ionic liquid has the advantages of high ionic conductivity, high thermal stability etc, and the electrolyte composed with iron compounds can overcome the limitations of aqueous solution system, but the high cost limits its prospect for large-scale application. Additionally, the current main challenges and possible solutions were summarized, and the future development directions were prospected.

In steel manufacturing, the quality of the hot rolling plan directly impacts production efficiency, costs, and delivery schedules. To address the limitations of existing models, such as incomplete objective coverage and ineffective resolution of multi-objective conflicts, the multi-objective hot rolling batch planning problem was formulated as a prize-collecting vehicle routing problem （PCVRP）. An improved non-dominated sorting genetic algorithm III （NSGA-III）, incorporating constraints and path optimization, was proposed to solve this problem. The model utilized virtual slabs from continuous casting and physical inventory slabs as inputs, considering key factors including attribute changes between adjacent slabs, rolling unit length, total batch plan length, and the proportion of hot slabs and constructs optimization evalution system combing three core evalution values with comprehesive evaluation score. The algorithm initialized the population using a hybrid strategy of constraint satisfaction and a path nearest-neighbor pool to enhance initial solution quality while maintaining diversity. Novel crossover and mutation operators, integrating model constraints with path optimization, were designed to accelerate convergence and avoid local optima. Through the synergistic design of the model and algorithm, an effective balance among conflicting objectives and efficient problem-solving were achieved. Experimental results based on real-world production data from a steel plant demonstrate that the proposed algorithm outperforms MOEA/D（multi-objective evolutionary algorithm based on decomposition）, NSGA-II, and GA（genetic algorithm）, with improvements in the comprehensive evaluation value of 2.3%, 5.1%, and 35.4%, respectively. Furthermore, the comprehensive evaluation value of the initial solution is optimized by 57.5% during the iterative process, confirming that the proposed model and algorithm significantly enhance the efficiency and quality of hot rolling batch planning.

The 700L high-strength large beam steel, which is a key component for bearing the weight of vehicle body and external loads, must has excellent product performance and requires a high level of cleanliness for molten steel. The hard inclusions such as Al₂O₃, Ca-（Mg）-Al-O present in the steel are difficult to deform during the rolling process and their improper control can cause cracking during stamping and bending of the steel plate. The rare earth element, represented by Ce, has high reactivity and can transform the inclusions in the steel into rare earth inclusions, reducing the harm caused by large-sized spherical inclusions. Moreover, the rare earth inclusions in the steel have a lower degree of misfit and can serve as hetero nucleation cores to refine the solidification structure of casting billet. The results show that when the mass fraction of rare earth Ce in the steel is 0.001 3%, the spherical inclusions are refined, with the number per unit area decreasing from 22 to 11 and the maximum size reducing from 11 μm to 6 μm. The number and size of TiN-like inclusions remain unchanged. The rare earth Ce transforms Al₂O₃, Ca-（Mg）-Al-O inclusions in the steel into Ce-Al-O inclusions, and the degree of transformation depends on the local Ce content. The transformed Ce-Al-O inherits the morphological characteristics of original inclusions in the steel. The morphology of TiN changes, presenting a brittle structure, which is conducive to alleviating stress concentration. Based on FactSage thermodynamic calculations, when no rare earth Ce is added, the steel liquid composition is in the non-ideal phase region （liquid+slag+CaAl₄O₇）, and after adding rare earth Ce, the steel liquid enters the liquid+slag+AlCeO₃ phase region. The rare earth transforms the deoxidation products and calcium-aluminate inclusions into Ce-Al-O inclusions, which is consistent with the SEM（scanning electron microscopy） observation results. The mismatch degree of Ce-Al-O in ferrite steel is less than 8%, which promotes the solidification nucleation of the steel. The equiaxed grain zone of slab is expanded, the dendrite is refined, and the primary dendrite spacing is reduced from 426 μm to 280 μm.

Given the problems of complex image noise, low separation between coke powder and background, and blurred images collected on the spot in the detection of fuel coke powder in metallurgical industry, it is difficult to accurately detect the particle size. A coke powder detection model ESGE-RTDETR （edge-sparse graph-enhanced efficient real-time detection transformer） based on the improved RTDETR algorithm was proposed, which can efficiently and accurately detect the coke powder particle size in coke powder detection scenarios. The MutiScaleEdge multi-scale edge convolution module combined with the ConvEdgeFusion edge feature fusion module was proposed to obtain feature maps with different dimensions of information. Windowed attention and dynamic adaptive sparse attention were used to optimize the calculation complexity. The neck feature was fused through the CSP-MSF（cross stage partial multi-scale fusion） fusion module, and finally the coke powder detection results were obtained by inputting the detection head. In order to improve the training accuracy, combining the characteristics of PowerIoU and FocalerIoU, FocalPowerIoU was proposed to replace the original GIoU, which made the training faster convergent and stable, and improved the accuracy of the model. Enhanced the interpretability of the model detection process by visualizing the feature extraction points and regions focused by the model. In the actual production process, adaptive histogram equalization （contrast limited adaptive histogram equalization,CLAHE） was selected to enhance the image after experimental comparison in the preprocessing stage of detection, highlighting the edge characteristics of coke powder, providing stable input for model detection, and improving the comprehensiveness and accuracy of model reasoning results through CLAHE image enhancement. The experimental results on the coke powder image data set of a steel plant show that the ESGE-RTDETR model has a better effect on improving the recognition accuracy of coke powder multi-scale particle size compared with the mainstream target detection model. Compared with the original RTDETR model, the mean average precision（P_mA50） has increased by 20.6 percentage points, and the recall rate has increased by 14.1 percentage points. Compared with the mainstream detection model YOLOv8, the accuracy rate of P_mA50 has increased by 8.9 percentage points, and the recall rate has increased by 8.5 percentage points. It can provide technical support for on-site production and industrial closed-loop control of coke powder particle size. The actual production verification of a steel plant meets the requirements of production detection accuracy and speed.

As a key demonstration project for the transformation and upgrading of Hebei Iron and Steel Group, Tangsteel New District aims to achieve the construction goals of "greening, intelligence and branding", and is committed to building a production network architecture that integrates material flow, energy flow and information flow. The iron-steel interface, as a crucial link in the steel production process, encompasses multiple technological elements such as production organization, scheduling management, and logistics transportation. It plays a pivotal role in connecting the upper and lower levels and is of great significance for enhancing the production efficiency and economic benefits of enterprises. Based on the theory of metallurgical process engineering, Tangsteel New District had carried out a series of innovations and applications of iron-steel interface technology. An intelligent management system for the steel process had been established, a software platform for intelligent processes had been built, and core technologies such as laminar flow operation of processes, interface collaborative optimization, digital simulation of processes, and five-dimensional dynamic Gantt charts had been innovatively developed. Through the innovative application of models such as the blast furnace iron tapping prediction model, the iron water temperature drop prediction model, the full-process covering technology of the iron water ladle, and the optimization technology of the ladle allocation mode, the iron water ladle turnover rate and the iron water ladle accuracy rate in the new area of Tangsteel have been significantly improved, and the iron water temperature drop has been greatly reduced. The application of intelligent technologies such as the Gantt chart of the iron-steel interface, the iron separation decision, the intelligent ladle allocation decision, the tail ladle transfer decision, the molten iron transportation task scheduling, and the KPI（key performance indicator） statistical analysis have achieved real-time dynamic tracking and scheduling of the iron-steel interface, which not only improves the operational efficiency of the iron-steel interface, at the same time, it has also enhanced the ability of refined management. The innovation and application of iron-steel interface technology in Tangsteel New District have enriched the theoretical connotation of metallurgical process engineering and set a model for the practical application of metallurgical process engineering in the steel industry.

Driven by China's "dual carbon" strategy, the metallurgical industry is rapidly advancing toward green and intelligent transformation. As a core process in steel production, the level of intelligence in continuous casting significantly influences the overall efficiency, energy utilization, and product quality of the steel manufacturing chain. Recent advances and innovative applications of artificial intelligence （AI） in the continuous casting process are systematically reviewed. Firstly, regarding breakout prediction, the causes and consequences of breakouts are analyzed, and the effectiveness and limitations of AI-based prediction models in enhancing warning accuracy and reducing false alarms are discussed. Secondly, in the domain of secondary cooling dynamic water control, a hybrid approach is proposed, combining genetic algorithms for parameter optimization with deep neural networks to construct a multi-variable control model. This enables intelligent adjustment of water flow and precise local thermal field regulation, effectively reducing thermal stress and crack formation in the plate. To address challenges in plate surface defect detection, the integration of deep learning and machine vision is explored. Convolutional neural networks （CNN） are used to automate defect recognition, classification, and statistical analysis, significantly improving detection precision and efficiency. Moreover, the latest developments in the deployment of casting operation robots are reviewed, focusing on tasks such as mold replacement, temperature sampling, and automatic slag addition in high-risk, labor-intensive scenarios, demonstrating strong potential for intelligent operation. Despite these advances, key challenges hindering intelligent transformation in continuous casting are identified, including the lack of standardized data acquisition protocols, limited generalization capability of AI models under complex boundary conditions, and poor adaptability to extreme casting environments. Therefore, it is imperative to accelerate the development of large-scale industrial data platforms and multimodal sensing technologies to realize intelligent perception, prediction, and control throughout the continuous casting process. The result of study aims to provide strong technical support for the steel industry in achieving smart manufacturing goals characterized by zero defects, self-adaptation, and ultra-low carbon emissions.

As the wind power industry rapidly develops, the installed capacity of wind turbines continues to increase, with individual turbine units becoming progressively larger, placing more stringent demands on the scalability and high performance of wind power equipment. To meet the critical requirements for high strength, low-temperature fracture toughness, and fatigue crack arrest ability of steel used in wind tower structures, a 500 MPa grade high crack resistance toughness wind tower steel was developed through low-carbon microalloying and the application of TMCP （thermo-mechanical controlled processing）. The microstructure of the steel primarily consists of uniformly fine ferrite and granular bainite, with an average grain size of 3.57 μm. The fraction of low-angle grain boundaries （3°-15°） is 43.5%, while the fraction of high-angle grain boundaries （>15°） is 56.5%. The material exhibits excellent mechanical properties, with a yield strength of 580 MPa, a tensile strength of 689 MPa, and an elongation of 19.36%. At -80 ℃, the impact energy reaches 257 J, and at -40 ℃, the CTOD （crack tip opening displacement） value is 0.604 mm, demonstrating outstanding low-temperature toughness and crack arrest performance. Microstructural analysis reveals that the presence of polygonal ferrite significantly enhances the material's low-temperature toughness, while a small amount of granular bainite contributes to the strength. The refined grain size improves the material's strength, and the high fraction of high-angle grain boundaries effectively inhibits crack propagation, further enhancing the low-temperature toughness and crack arrest capabilities of the material. Provide a theoretical basis for the practical application of high-performance steel for wind power.

As a high-carbon emission field, the green low-carbon transformation of the steel industry is of great strategic significance to achieve the goal of "double carbon". The deep integration of hydrogen energy and steel processes, especially hydrogen metallurgy technology, fundamentally reduces the dependence on carbon by replacing "hydrogen" with "carbon" reduction, and has become an important direction of future ironmaking technological innovation. The current mainstream multi-hydrogen metallurgy process technology and its implementation path are systematically described, in-depth analysis is carried out from multiple dimensions such as raw material acquisition and preparation, core reaction mechanism, key equipment application to process flow construction, and is the characteristics, advantages and limitations of each process technology comprehensively evaluated. The coke oven gas zero reforming hydrogen metallurgy demonstration project （HyMEX） of HBIS Group is focused on, and its process technology innovation and operation practice is introduced in detail. The HyMEX project successfully applied the "coke oven gas zero reforming direct reduction technology" engineering for the first time, breaking through the international conventional means of using natural gas to produce reduction process gas, and becoming the direct reduction process of gas-based shaft furnace with the highest proportion of hydrogen in industrial production, setting a benchmark for the industrial application of hydrogen metallurgy technology. Combining with China's industrial policy orientation and the characteristics of resource endowment, the development prospect and sustainable development technology path of shaft furnace hydrogen metallurgy in China is deeply discussed. As the world's largest steel producer, China has rich coke oven gas resources, which provides a unique resource advantage for the large-scale promotion of hydrogen metallurgy technology. In the future, with the continuous improvement of the hydrogen energy industry chain and the reduction of costs, hydrogen metallurgy technology is expected to achieve large-scale commercial application in China, so as to promote the steel industry to accelerate the transformation to the green and low-carbon direction, and provide strong technical support for the realization of the "double carbon" goal.

The residual ferrite in austenitic stainless steel significantly impacts its service performance, with its characteristics primarily influenced by composition, cooling rate, and solidification mode. The distribution characteristics of residual ferrite and precipitated phases along the width direction of a continuous-cast 12.5% nickel 316L austenitic stainless steel billet was investigated. Characterization was performed using metallographic analysis, electron back-scatter diffraction （EBSD）, and electron probe microanalysis （EPMA）. Solidification mode was determined based on residual ferrite morphology, Thermo-Calc thermodynamic calculations, chromium/nickel equivalent empirical formulas, and high-temperature laser scanning confocal microscopy （HT-CLSM） experiments. The results indicate that the ferrite volume fraction along the centerline thickness direction of the billet width exhibits an "A"-shaped distribution. The surface region shows the lowest residual ferrite volume fraction （4.14%）, while the center region has the highest （8.99%）. The calculated cooling rate decreases from 7.60 °C/s at the billet surface to 0.38 °C/s at the center. Regarding morphology, in the surface fine-grained zone （≤20 mm from surface）, granular and parallel short-rod ferrite are formed. In the columnar grain zone （20-60 mm）, ferrite morphology evolves from skeletal→lath-like→vermicular. In the central equiaxed grain zone （>60 mm from the surface）, dense lath-like and clustered vermicular structures are formed. EBSD and EPMA results reveal that partial transformation of δ-ferrite to σ/chi phases occurs at subsurface of the billet. In the billet center region, the δ→σ+γ₂ eutectoid reaction leads to σ-phase precipitation within the austenite grains. Concerning solidification mode, both results of Thermo-Calc thermodynamic calculations and empirical chromium/nickel equivalent formulas predict an AF （austenitic-ferritic） mode. However, in-situ HT-CLSM observation shows skeletal δ-ferrite preferentially precipitating at 1 392.6 ℃, followed by austenite growing interdendritically at 1 386.5 °C. This sequence （L→L+δ→L+δ+γ→δ+γ） aligns with an FA （ferritic-austenitic） solidification mode. Features such as ferrite enveloping austenite in the billet and the δ→σ+γ₂ eutectoid decomposition are characteristic of FA mode. A discrepancy thus exists between the solidification mode predicted by thermodynamic calculations and empirical formulas and the mode observed experimentally. It aims to provide theoretical guidance for the control of ferrite in the production of 12.5% nickel 316L stainless steel continuous casting billet.

Ausforming, as a classical thermomechanical treatment process, can effectively refine the bainitic microstructure, providing a promising solution for the production of low carbon CFB steels with a combination of high strength and excellent ductility. However, systematic studies of the relationships among ausforming temperature, bainitic transformation kinetics, microstructure characteristics and their effects on mechanical properties in low carbon CFB steels are still limited. Therefore, a low-carbon CFB steel was subjected to ausforming treatment at different temperatures （600, 500, 400 ℃） prior to isothermal bainitic transformation at 350 ℃ and compared with conventional isothermal heat treatment. The effects of ausforming temperature on bainitic transformation kinetics, microstructural evolution, and mechanical properties were studied using a combination of DIL805A/D dilatometry, SEM, XRD, EBSD, TEM, and hot rolling experiments. The results show that, compared to the conventional isothermal heat treatment, high-temperature （600 ℃） ausforming has little effect on the bainitic transformation kinetics, while ausforming at medium-temperature （500 ℃） and low-temperature （400 ℃） significantly accelerates the bainitic transformation kinetics and the maximum bainite volume fraction increases. The ausforming temperature also influences the morphology of the CFB steel microstructure. When ausforming at 600 ℃, the volume fraction of granular features of bainitic ferrite （BF） increased, whereas the morphology of BF was almost lath-like when ausforming at 400 ℃. Additionally, the content of martensite/austenite （M/A） islands produced during secondary cooling in the ausformed samples was reduced and its size was smaller. While maintaining the same strength level as the conventional process, the uniform elongation increased from 6.8% to 17.6% when ausforming at 500 ℃. This is primarily attributed to the large amount of RA in the microstructure, which undergoes a progressive transformation-induced plasticity （TRIP） effect over a larger strain range during tensile deformation, thus enhancing the work hardening capability. The result provides a new method for the development of low-carbon CFB steels with high strength and plasticity, which has great potential for industrial applications.

Vanadium-titanium magnetite is widely distributed globally with abundant reserves. Vanadium and titanium are key metal elements in aerospace, electronic information, high-end manufacturing, new energy, new materials and other fields, and their development value in metallurgy and material industry is extremely important. However, in the traditional process smelting process based on the blast furnace-converter method, the blast furnace slag is often accompanied by a high content of titanium nitride and titanium carbide. These high melting point compounds not only easily lead to the increase of blast furnace slag viscosity in the furnace, but also lead to bad phenomena such as iron content in the slag and accumulation of the hearth, which affect the normal production. Vanadium-titanium magnetite provided by Heilongjiang Jianlong Company was used as raw material. Firstly, its physical and chemical properties and composition were characterized and analyzed, and relevant experiments were designed to explore the evolution law of melting performance in slag. At the same time, the specific effects of alkalinity and w（MgO）/w（Al₂O₃） on melting performance were further studied. The results show that when the mass fraction of TiO₂ increases from 11% to 14%, the softening temperature, hemispherical temperature and flow temperature of slag reach 1 245, 1 255 and 1 267 ℃, respectively. With the increase of alkalinity from 1.1 to 1.4, the three temperature indexes also show an increasing trend, which are 1 240, 1 247 and 1 259 ℃ respectively when the alkalinity is 1.4. With the increase of w（MgO）/w（Al₂O₃） from 0.5 to 0.9, the slag softening, hemispherical and flow temperatures decrease significantly, with the lowest values of 1 208,1 222 and 1 238 ℃, respectively. In terms of viscosity, the change of TiO₂ content has little effect on it. The increase of alkalinity can effectively reduce the viscosity. The w（MgO）/w（Al₂O₃） is the most sensitive to the viscosity, the viscosity decreases significantly from 0.5 to 0.9, and the downward trend slows down when it continues to rise to 1.1. It provides a theoretical basis for optimizing the vanadium-titanium ore smelting process and inhibiting the adverse effects of TiN and TiC.

Under the current "dual-carbon" framework, the high-value utilization of typical large-scale solid wastes from metallurgical processes and other major industrial sources has emerged as a critical issue demanding urgent resolution within the industry. Large-scale solid wastes from metallurgical processes, including blast furnace slag and dust ash, as well as other major industrial solid wastes such as fly ash and coal gangue, are rich in valuable resources like silicon and aluminum. Through appropriate conditioning and proportioning treatments, these solid wastes can be transformed into inorganic fiber materials exhibiting excellent thermal insulation and refractory properties, thereby achieving high-value utilization of both metallurgical and other major industrial solid wastes. The research progress in preparing mineral wool fibers is reviewed, using conditioned blast furnace slag as the raw material and silicon-aluminum-based ceramic fiber materials using dust ash in combination with fly ash or coal gangue. It analyzes the advancements in the production of inorganic fibers via the spray-blowing method and the centrifugal method from three perspectives, fundamental principles, experimental studies, and industrial practices. Pilot-scale experimental studies are conducted to produce mineral wool fibers from conditioned blast furnace slag using both spray-blowing and centrifugal methods. The results indicate that spray-blowing pressure significantly affects the content of fiber slag balls, while having minimal impact on the average fiber diameter. Specifically, increasing the spray-blowing pressure from 0.20 MPa to 0.38 MPa reduces the fiber slag ball mass fraction from 25% to 16%, with negligible changes in fiber diameter. Conversely, roller speed has little effect on fiber slag ball content but significantly influences fiber diameter; as roller speed increases, the average fiber diameter decreases from 3.17 μm to 2.73 μm. The study further compares the differences between mineral wool fibers produced by the two methods in terms of fiber diameter, fiber slag ball content, and fiber compressive strength. Additionally, it outlines the application prospects of silicon-aluminum-based inorganic fiber materials, derived from metallurgical large-scale solid wastes in conjunction with other major industrial solid wastes, in sectors such as construction, industry, aerogels, and photocatalytic materials. Building upon the existing research foundation, future research directions for the production of inorganic fiber materials from metallurgical large-scale solid wastes is proposed, aiming to further advance waste valorization, energy conservation, and carbon reduction.

GH4151 alloy is a difficult-to-deform nickel-based superalloy capable of serving at temperatures up to 800 ℃. This alloy features a high degree of alloying, with the γ′ strengthening phase accounting for 50%-60% of its mass fraction. Not only are its manufacturing processes, such as smelting and cogging, highly challenging, but its properties are also notably sensitive to heat treatment parameters. Current research on GH4151 alloy primarily focuses on smelting process control, cogging forging optimization, and the relationship between hot-working parameters and microstructural homogeneity. In contrast, studies on the heat treatment of GH4151 are relatively scarce, particularly regarding the influence mechanisms of solution and aging treatments on microstructure and properties.It investigated the nickel-based superalloy GH4151, which could operate at temperatures up to 800 ℃. Using double aging heat treatment （850 ℃×6 h+760 ℃×16 h） under both sub-solvus （1 130 ℃） and super-solvus （1 170 ℃） solution conditions, combined with optical microscopy, scanning electron microscopy, and selected area electron diffraction analysis via transmission electron microscopy, the grain size distribution, evolution of primary and secondary γ′ phases, and precipitation behavior of grain boundary phases were systematically examined under different heat treatment parameters. Furthermore, through a series of tensile tests at room and elevated temperatures, the effect of heat treatment on the mechanical properties of GH4151 alloy was thoroughly investigated.The results indicate that the solution temperature significantly influences the dissolution behavior of the primary γ′ phase and the grain size distribution. After sub-solvus treatment, a certain amount of undissolved large-sized primary γ′ phase remains, which pins the grain boundaries and effectively inhibits grain growth, resulting in fine-grained microstructure. In contrast, super-solvus treatment leads to complete dissolution of the primary γ′ phase, resulting in rapid grain growth and coarse-grained structure.The double aging heat treatment promotes further precipitation of the secondary γ′ phase, thereby enhancing the tensile strength of the alloy at room temperature, 650 ℃ and 750 ℃, as well as the room-temperature hardness. However, the size and distribution of the secondary γ′ phase vary with different solution treatments. Moreover, since the aging temperature falls within the precipitation range of grain boundary μ phase and M₂₃C₆ carbides, extensive co-precipitation of these phases occurs along the grain boundaries. In particular, the precipitation of the brittle and hard μ phase leads to a reduction in tensile strength at 800 ℃.

The cold strength of carbon-containing pellets plays a critical role in transportation efficiency and smelting process permeability,directly impacting industrial production stability and economic viability. Zinc extraction tailings,which are secondary solid wastes generated after the pyrometallurgical recovery of zinc from zinc-containing dusts,have garnered increasing attention in industrial research due to their potential for resource recovery and utilization. A method was devised to produce carbon-containing pellets by utilizing zinc tailings,iron concentrate powder,and semi-coke,aiming to lower smelting expenses and improve the efficiency of solid waste resource utilization. The compressive strength of green pellets was selected as the response variable,with moisture mass fraction,binder ratio,and forming pressure identified as key process parameters. Response surface methodology（RSM）was employed to optimize the cold consolidation process for zinc tailings-based carbon-containing pellets. The results demonstrate that moisture mass fraction（A）and forming pressure （C）exert statistically significant positive effects on compressive strength. Optimal moisture enhances feedstock plasticity and promotes particle adhesion. Meanwhile,higher forming pressure increases particle contact density,thereby improving structural stability. Although the effect of binder ratio（B）alone did not reach a significant level,its interaction with moisture content exhibited a significant synergistic effect. The experimental analysis results led to the development of a binomial model to predict the compressive strength of carbon-containing pellets. The optimal process parameters for the cold consolidation of these pellets were determined to be a moisture mass fraction of 4.67%,a binder ratio of 5.51%,and a forming pressure of 26.98 MPa. The results confirmed the accuracy and reliability of the model,showing deviations in compressive strength of less than 2% from the predicted values. This systematic optimization offers reliable theoretical basis and guidance of high-performance metallurgical raw materials derived from solid waste resources.

As one of the key components in the field of high-end intelligent equipment, precision bearings have higher requirements for the mechanical properties, high-temperature oxidation performance, and corrosion resistance of bearing steel. The high-temperature oxidation behavior of the heating process for bearing steel is the key factor affecting the dimensional accuracy and performance of bearing parts, and controlling and optimizing it is key. The influence of rare earth elements on the oxidation properties of GCr15 bearing steel was tested and characterized through oxidation experiments in the process of high-temperature heating and heat preservation, and the related mechanisms were discussed and analyzed. The structure and phase composition of the oxide layer on the surface of bearing steel with different rare earth element contents were characterized by XRD, SEM, EDS, and other methods. The thermodynamic phase diagram of the composition and evolution of the oxide coating layer is calculated and analyzed, and an oxidation kinetic model is constructed. The experimental results show that the oxidation rate constants of bearing steel with rare earth addition （mass fraction） of 0, 0.003% and 0.015% were 3.44, 3.28 and 1.55, respectively, during the oxidation process at 900 ℃. Compared with the bearing steel without the addition of rare earth elements, the thickness of the oxide layer is reduced from 99.00 μm to 76.67 μm after the addition of 0.015% rare earth elements, which is 22.56% thinner. The oxide film is composed of inner oxide layer, middle oxide layer, and outer oxide layer, which is mainly composed of dense spinel oxide, and the formation of the middle Cr-rich layer and the inner layer of Si-rich oxide affects the high-temperature oxidation resistance of rare earth steel. The oxide layer of bearing steel without rare earth elements is thick, and there are numerous cracks and voids. The oxide layer of bearing steel with the addition of rare earth elements is relatively dense and continuous. In the initial oxidation stage at high temperatures, rare earth elements increases the diffusion rate of Si and Cr in the bearing steel, promotes the formation of dense oxide layer with high Fe₂SiO₄ and FeCr₂O₄ content on the surface, and effectively slow down the oxidation rate. During the growth process of the oxide film, rare earth elements hinder the outward diffusion of Cr³⁺, leading to a shift in the dominant growth of the oxide film towards inward diffusion of O^2-, and significantly improves the high temperature oxidation resistance of GCr15 bearing steel.

With the gradual expansion of marine resource development into deep-sea and extremely cold regions, marine engineering steels are required to possess high strength, high toughness and excellent corrosion resistance in extreme environments. Non-metallic inclusions, as inherent defects in steel, their composition, size, morphology and distribution characteristics have a decisive influence on the resistance to hydrogen-induced cracking, pitting corrosion resistance and mechanical properties of marine engineering steels. Through Ca treatment, Mg treatment, RE modification and oxide metallurgy technologies, the morphology, size and distribution of inclusions can be optimized to balance their "defects" and "functions". Therefore, the influence law and mechanisms of inclusion characteristics on different properties of marine engineering steel are systematically reviewed. The key strategies of inclusion control to improve hydrogen-induced cracking resistance, pitting corrosion resistance, and mechanical properties are comprehensively summarized. Furthermore, the future development trend of inclusion control technology in marine engineering steel is put forward. It aims to provide a theoretical basis and technical route for the development of new generation of marine engineering steels with the characteristics of "high strength, high toughness, and corrosion resistance".

Aiming at the bottlenecks of traditional vanadium-titanium-magnetite smelting process, such as high carbon emission and low resource utilization, a green metallurgical process was proposed utilizing H₂ reduction of internally mated biomass vanadium-titanium magnetite pellets. A test of H₂ reduction of vanadium-titanium magnetite pellets with internal biomass at 1 100 ℃ was carried out to systematically investigate the effects of H₂ volume fraction （30%,40%,60%） and reduction time （20,30,40,50,60 min） on the reduction effect of the pellets as well as the changes in the microscopic morphology of pellet, and compared with that of the original ore pellets. The results show that the composite pellets containing biomass improve gas diffusion efficiency by altering their pore structure. Under 30% H₂（volume fraction） conditions, they achieve the same reduction effect as the original ore pellets under 40% H₂ volume fraction conditions. When the H₂ volume fraction is increased to 60%, the metallization rate of composite pellets reaches 94.58%, and titanium ore residues are eliminated, confirming that 40%Ar-60%H₂ is the optimized process condition for pellet deep metallization. The internally mated biomass significantly optimizes the reduction kinetics of composite pellets, continuously accelerating the metallization process during the critical reduction period of 20,30,40,50,60 min. By enhancing the reduction atmosphere, reducing activation energy, and improving diffusion pathways, composite pellets achieve the same high metallization rate （approximately 96%） at least 10 min earlier than original ore pellets, significantly shortening the reduction cycle and reducing reduction atmosphere consumption. The composite pellets form a porous honeycomb-like iron matrix, exhibiting 22.68% increase in BET（Brunner-Emmet-Teller）specific surface area compared to the original ore pellets, with rough surface structure expanding the gas-solid contact area. The increased contribution of medium/large pores doubles the average pore size, providing diffusion pathways for the directed migration of iron atoms and promoting the deep reduction and aggregation of titanium iron oxides （disappearance of the FeTiO₃ phase）, synergistically achieving diffusion-reaction kinetic optimization.

To improve the surface quality of low-carbon steel and address defects caused by incomplete removal of iron oxide scale, the microstructure and formation mechanism of oxide scales on low-carbon steels with different Si contents during high-temperature oxidation were investigated by using FE-SEM （field emission scanning electron microscopy） and EDS （energy dispersive spectroscopy）. It focused on elucidating the effects of temperature and Si content on oxidation behavior. The results show that the oxide scales mainly consist of Fe₂SiO₃, Fe₃SiO₄, FeO and an internal oxidation layer containing SiO₂2SiO₄ phase. At heating temperature of 1 050 ℃, the phase of Fe₂SiO₄ appears as finely dispersed particles, and the oxide-scale/substrate interface remain relatively flat. However, when the heating temperature is 1 170 ℃, the Fe₂SiO₄ phase transforms into continuous dendritic or network-like structures, significantly deteriorating interface flatness. Moreover, the influence of Si content on oxidation weight gain exhibits temperature dependence. At heating temperature of 1 050 ℃, the steel with high Si content promotes the formation of solid Fe₂SiO₄, which could effectively hinder the diffusion of iron and oxygen ions, thus inhibiting the oxidation process. In contrast, at 1 170 ℃, the formation of Fe₂SiO₄-FeO eutectic liquid phase provides a faster channel for atomic and ionic diffusion in oxidation reaction, thus accelerating the oxidation process. Additionally, the eutectic liquid phase tends to infiltrate into the matrix and FeO, forming anchor and/or grid shape with it, pinning the grain boundary of the matrix, strengthening the bonding between the oxide scale and the matrix, and making the peeling of the iron oxide scale more difficult. Based on this, process and chemical composition optimization are proposed, when the content of Si in steels is relative high, the heating temperature should be controlled below 1 170 ℃ to suppress the formation of Fe₂SiO₄-FeO eutectic liquid phase, thereby improving surface quality and reducing iron loss. The intrinsic relationship among temperature, Si content and oxidation behavior is clarified, providing theoretical foundation and process guidance for surface quality control in hot-rolled low-carbon steel production.

During the electric furnace smelting process of high titanium slag, the excessively fine particle size of fine-grained titanium concentrate leads to significant generation of furnace dust. To enhance the recycling utilization of this high titanium slag smelting dust, experimental investigations were conducted on its cold-bonded briquetting performance and high-temperature bursting characteristics. The orthogonal experimental method was employed to study the influence of cold-bonding parameters,including binder ratio, mass fraction of binder ,water addition（mass fraction） and forming pressure on the drop strength and compressive strength of both green balls and dry balls of furnace dust. The orthogonal test results demonstrated that forming pressure exerted the most significant effect on the tumble strength of green balls, as well as the compressive strength and tumble strength of dry balls. Meanwhile, water addition（mass fraction） shows the most pronounced influence on the compressive strength of green balls. The drying process substantially improves both the tumble strength and compressive strength of the pellets. Considering both cost efficiency and pellet strength, the optimal parameter combination is determined as follows,8% binder ratio, 8% binder mass fraction, 0 water addition, and 6 MPa forming pressure. Under these conditions, the produced pellets exhibit stable performance, the tumble strength and compressive strength of green balls reach 3 times/（0.5 m） and 241.5 N/pellet, respectively, while those of dry balls reach 125 times/（0.5 m） and 477.3 N/pellet. High-temperature bursting tests reveal that the bursting ratio of pellets increases progressively with rising temperature, and the bursting temperature range for cold-bonded pellets make from high titanium slag dust is 700-800 ℃ . The cold-bonded briquetting process enables effective recycling of high titanium slag dust, thereby improving the resource utilization efficiency of fine-grained titanium concentrate in electric furnace slag smelting.

Based on the process characteristics of China's steel industry and the progress of hydrogen metallurgy technology, hydrogen-rich ironmaking in blast furnace is an important path for reducing carbon emissions in the current scale of China's steel industry. The basic principles of reducing carbon emissions, lowering energy consumption per ton of iron, and improving production efficiency by enriching hydrogen in blast furnaces was explained. It analyzed the characteristics and problems of different smelting process modes based on hydrogen rich coupling furnace top gas circulation, high oxygen enrichment, and preheating. The carbon reduction under modes such as hydrogen rich coupling with top gas circulation and electric heating in blast furnaces can reach over 50%, and the development trend of high-efficiency, low-carbon, and green blast furnace ironmaking technology is high efficiency. The experiment and industrial demonstration of pure hydrogen injection into blast furnaces, as well as the freezing dissection study of hydrogen rich blast furnaces, provide foundation for the construction of prototype of low-carbon smelting process for hydrogen rich blast furnaces. The lack of hydrogen sources in large-scale economies, the failure to connect downstream low-carbon product consumption market chains, and the failure to implement carbon taxes have resulted in a lack of economic viability for hydrogen metallurgy technology, which is bottleneck hindering its development. Building a globally recognized high-quality standard system and low-carbon industry ecosystem based on top-level design, accelerating core technology research and breaking through scale bottlenecks, the development of hydrogen metallurgy technology is promoted by incorporating monitoring and trading systems, and utilizing carbon trading mechanisms to encourage steel enterprises to accelerate the adoption of low-carbon technologies. With the transition from traditional carbon metallurgy to hydrogen metallurgy, hydrogen flow as a reducing agent and fuel has become a new variable that needs to be considered in the upgrading and transformation of traditional metallurgical processes. It poses new challenges for the overall process optimization and functional optimization of metallurgical processes, and has become an important research topic in metallurgical process engineering.

Based on the first and second laws of thermodynamics, classical energy analysis models and grey-box exergy analysis models for the BF（blast furnace） and SF（shaft furnace） smelting processes were established respectively. Employing evaluation indicators such as thermal efficiency and exergy efficiency, it systematically analyzed the differences in energy utilization between BF and SF smelting processes and compared their carbon emissions and gas utilization rates. The results show that, the material input of BF is dominated by iron ore and blast air. The material input of SF is primarily comprised of feed ore and reducing gas. In terms of energy efficiency, the main heat sources for BF ironmaking are carbon combustion before the tuyere and the sensible heat of blast air, with a thermal efficiency of 80.70%. The BF exergy efficiency is 56.57%, and the exergy loss during ironmaking is 4.25 GJ/t, of which external exergy loss accounts for a relatively high proportion （approximately 27.71% of total exergy input） and constitutes 63.80% of the total exergy loss. For SF ironmaking, the main heat source is reducing gas, with a thermal efficiency of 74.55%. The exergy efficiency of the SF smelting process is 41.61%, and the exergy loss during smelting is 5.462 GJ/t, which accounts for 58.39% of the total exergy expenditure. Comparison reveals that both the thermal efficiency and exergy efficiency of the BF are higher than those of the SF, indicating the BF's advantage in thermal energy conversion and quality maintenance. The shaft furnace process uses hydrogen-rich gas as a reducing agent, and its carbon emission per unit product is 322.42 kg/t,52.5% lower than that of the blast furnace （679.69 kg/t）. At the same time,the gas utilization rate （57.87%） and hydrogen utilization rate （59.77%） of the shaft furnace are higher than those of the blast furnace （52.00% and 41.35%, respectively）.