- 百种中国杰出学术期刊
- 中国精品科技期刊
- “中国科技期刊卓越行动计划二期”领军期刊
- 入选“中国精品科技期刊顶尖学术论文(F5000)”
- 入选《科技期刊世界影响力指数(WJCI) 报告》
- EI、CA、JST收录期刊
- 中国金属学会会刊
- 中文核心期刊
- 中国科技核心期刊
- 中国科学引文数据库(CSCD)来源期刊
The iron and steel industry is one of the major source of carbon emissions, and advancing pellet technology is one of the effective measures to achieve the "dual-carbon" goals. Excessive reduction swelling of pellets can degrade reactor permeability and even leads to production accidents. This paper systematically reviews the primary mechanisms of reduction swelling in iron ore pellets, including lattice expansion from phase transformations, iron layer cracking due to gas pressure, structural damage induced by carbon deposition, cracking resulting from uneven reduction stresses, and the precipitation morphology of nascent iron. Studies have shown that regulating the formation of iron whiskers is a key breakthrough in inhibiting malignant expansion, while reduction swelling is significantly influenced by preheating/roasting parameters, porosity, gangue composition, and reduction conditions. Key measures to suppress pellet swelling involve optimizing ore blending, rational control of basicity, refining preheating/roasting processes, and restricting harmful element intake. It provides a theoretical foundations and technical pathways for optimizing pellet performance and advancing low-carbon ironmaking technologies.
In the modern production of high-quality steel materials, precise control of the metallurgical behavior inside the continuous casting mold is crucial for ensuring slab quality and process stability. Asymmetric flow in the mold, a typical and highly detrimental form of flow field distortion, has become a key technical bottleneck limiting production efficiency and yield of high-grade steels under high-speed continuous casting conditions. This paper systematically reviews recent publicly reported domestic and international research on the topic. It traces and analyzes both intrinsic and extrinsic factors leading to asymmetric flow, and clarifies its evolution mechanism from microscopic disturbances to macroscopic biased flow. From an evolving perspective, it organizes the technical development of control strategies from passive defense to active intervention. Finally, based on a comprehensive analysis of existing studies, the paper identifies current research gaps in the field and proposes forward-looking perspectives, aiming to provide insights for process improvement in the fields of metallurgical physical chemistry, multi-physics coupling simulation, and intelligent metallurgy.
The quenching and partitioning (Q&P) process is a new type of heat treatment technology proposed for martensitic steels. In the subsequent processing and deformation process, due to the mechanism of the dislocation absorption of retained austenite (DARA), transformation-induced plasticity effect (TRIP), and crack propagation barrier (BCP) effect of retained austenite in the steel, the materials exhibited the good strength and ductility matching, effectively solving the unity of high strength and high ductility. Stainless steels have been used in a wide range of applications in the fields of petrochemicals, household appliances, and products due to its excellent corrosion resistance, good weldability, and processability. However, due to the low strength of stainless steel, its application in engineering structures, transportation construction and other fields is greatly limited. The Q&P process can retain a certain volume fraction of retained austenite in low alloy steel, which can greatly improve the strength and plasticity of the material, which provides a reference for obtaining stainless steel materials with high strength and high plasticity. In this paper, the development process of Q&P process is reviewed, and the strengthening and toughening mechanism of microstructure and properties of martensitic stainless steel, ferritic stainless steel and austenitic stainless steel under Q&P process is introduced. The influence of austenitizing process, quenching termination process and partitioning process on the microstructure and properties of stainless steel in Q&P process is emphatically expounded. Then, the role and influence mechanism of N, Si, Al, Cu and other alloying elements in stainless steel Q&P process are introduced. Finally, the application prospect of Q&P in stainless steel is prospected, and the future development idea of high strength and toughness stainless steel is put forward, which provides some reference for the industrialization development of Q&P stainless steel.
As a new generation of high-strength and high-toughness structural steel, air-cooled bainitic steel has received extensive attention in recent years, because its bainite structure can be directly formed through air cooling without isothermal treatment. Its excellent ability to regulate microstructure and properties, as well as the potential for multi-performance coupling, has shown broad prospects in the field of structural steel application. This paper systematically reviews the development history of air-cooled bainitic steel and elaborates on the characteristics of its microstructure evolution and the mechanism of solid-state phase transformation. The effects of alloying element regulation, controlled cooling paths and non-isothermal heat treatment processes (such as Q&P, B&P) on the formation of its microstructure are analyzed emphatically. Furthermore, at the microscopic level, the performance differences among granular bainite, lower bainite and acicular ferrite structures are compared, and the stability of residual austenite and its important effect mechanisms in enhancing strength and toughness, reducing hydrogen embrittlement susceptibility and improving fatigue performance are summarized. For the new generation of air-cooled bainitic steel, the synergistic strengthening of dislocation density and subgranular structure is regarded as a key approach to improving yield strength and workability.
Electric arc furnace (EAF) steelmaking technology, featuring advantages such as a short process flow, low energy consumption, and reduced carbon emissions, serves as a critical pathway for the low-carbon transition in the steel industry. However, during the smelting process, this technology faces challenges including multivariable coupling and strong nonlinear dynamic characteristics. Leveraging its robust data processing and nonlinear mapping capabilities, machine learning technology exhibits significant application potential in EAF steelmaking production. This paper systematically summarizes the research progress of machine learning applications in the core links of electric arc furnace steelmaking. In the process of electric arc furnace steelmaking, within the field of molten steel property regulation, machine learning technologies have significantly improved the prediction accuracy of end-point carbon content and temperature through multi-algorithm fusion, but still need to overcome challenges such as data dependence and dynamic working condition adaptability. In terms of precise energy consumption management and control, semi-supervised learning technology effectively taps the value of unlabeled data to optimize energy input efficiency, and urgently needs to break through the core problems of model interpretability and cross-scenario generalization ability. Regarding the optimization of slag behavior, although time-series modeling methods can successfully predict the foaming state and guide process adjustment, they are confronted with issues of insufficient data robustness and weak mechanism correlation. In the links of process closed-loop and equipment operation and maintenance, the "data collection-real-time prediction-process adjustment-quality feedback" closed-loop system constructed by machine learning has realized fault early warning and parameter optimization, but is limited by constraints on high-dimensional data processing efficiency and algorithm real-time performance. For the future, the application of machine learning in EAF steelmaking should focus on directions including integrating mechanism modeling, advancing semi-supervised learning techniques, enhancing model interpretability, optimizing algorithmic tools, and addressing complex data processing challenges, thereby driving the intelligent and green-oriented deep transformation of EAF steelmaking.
Based on the resource availability and blast furnace smelting characteristics in Yunnan Province, this study systematically analyzes the impact of blended pulverized coal on combustion performance and the effects of mixed injection on the gas volume in the hearth and bosh. An innovative process route for high-intensity, ultra-low silicon hot metal smelting with "medium-titanium slag + lignite mixed injection" was proposed for medium-sized blast furnaces in the Yunnan region. On this basis, an appropriate blending ratio of lignite for mixed injection under current conditions was determined, and industrial trials were conducted. Research on the resource utilization of vanadium-titanium magnetite and lignite and its impact on blast furnace intensification, along with industrial practice results, show that when the mass fraction of bituminous coal in the blended coal is 10%, increasing the mass fraction of lignite leads to a rising trend in the maximum combustion rate, while the overall burnout characteristic index begins to decrease, and the combustion characteristic index first increases and then decreases. Considering these factors comprehensively, the suitable mass fraction of lignite is determined to be 10%-20%. As the mass fraction of lignite with high hydrogen content increases, the volume of hearth gas and the hydrogen content in the gas gradually increase. When the mass fraction of hydrogen in the mixed coal injection exceeds 4.0%, the volume fraction of hydrogen in the hearth gas exceeds 5.0%. Within the suitable lignite blending ratio range, for every 0.4-percentage-point increase in the mass fraction of hydrogen in the injected coal, the theoretical combustion temperature decreases by approximately 18 ℃, and the bosh gas volume increases by approximately 0.42%. To maintain the theoretical combustion temperature balance before lignite blending, the oxygen enrichment rate (oxygen volume fraction) needs to be increased from 4.00% during the baseline period to 5.80%. After increasing the oxygen enrichment rate, the hearth gas volume decreases by 4.82%. The reduction in hearth gas volume due to increased oxygen enrichment benefits blast furnace smooth operation and expands the operating window for intensifying smelting with medium-titanium slag characterized by high theoretical combustion temperature. Under the condition that the furnace burden grade (iron mass fraction) is less than 54.0%, the "medium-titanium slag + lignite mixed injection" high-intensity, ultra-low silicon smelting process achieves a blast furnace productivity coefficient exceeding 3.65 t/(m3·d).
Calcium carbide (CaC2), as a high-melting-point industrial raw material, remains solid at steelmaking temperatures and exerts the strong deoxidizing power of calcium. With its relatively low cost, it can partially substitute for aluminum in steel deoxidation, thereby reducing production expenses. This study analyzes the deoxidation capability of calcium carbide through thermodynamic and kinetic principles, explaining the theoretical basis of its deoxidizing behavior. The equilibrium oxygen mass fraction after deoxidation by calcium carbide alone is about 0.008 0%; therefore, other deoxidizers must be combined during steelmaking to reduce the oxygen content to a sufficiently low level. During converter tapping, when the oxygen content in the steel is 0.050 0%, the addition of 1.70 kg/t of calcium carbide and 0.82 kg/t of aluminum is required for deoxidation, whereas using aluminum alone would require 1.77 kg/t of aluminum. The combined use of calcium carbide and aluminum reduces aluminum consumption by 0.95 kg/t compared to aluminum-only deoxidation. The amount of calcium carbide per ton of steel and the corresponding aluminum reduction can be calculated based on the total oxygen content in the steel. Under the conditions considered in this study, adding 0.98 kg/t of calcium carbide and 1.11 kg/t of aluminum achieves effective deoxidation and meets the required composition standards. The cost per ton of steel is reduced by 4.03 RMB compared to the current cost of 28.67 RMB. Thus, the addition of calcium carbide effectively reduces aluminum consumption and lowers cost. Kinetic calculations show that the deoxidation rate of calcium carbide depends on its particle size. For the same addition amount, smaller particles provide greater contact area with the molten steel, leading to a faster deoxidation rate. Considering both safety and efficiency in the use of calcium carbide, a particle size of 5-10 mm is recommended for the deoxidation process.
To address the issue of macroscopic segregation in GCr15 bearing steel ingots and their corresponding finished products, this study innovatively employs the point tracking function of the Deform rolling simulation software. By reverse-tracking the initial position of the ingot based on the measured segregation points in the rolled material, a direct comparison and validation between the solidification simulation results obtained using ProCAST and the post-rolling experimental data are achieved. The carbon segregation distributions at the riser/spout end exhibit a high degree of consistency. Based on this, a systematic investigation was conducted using ProCAST solidification simulation software to further explore the influence of riser design parameters (including adiabatic plate thickness, pouring height, and thermal insulation performance) on the final macrosegregation results of the steel ingot. The findings indicate that increasing the insulation board thickness significantly improves segregation behavior. Specifically, when the thickness is increased from 42.5 mm to 62.5 mm, the feeding efficiency of the riser increases by 23%, while the carbon segregation index decreases from 2.75 to 1.82, representing a reduction of 34%. The application of hollow float bead insulation boards with low thermal conductivity (0.16 W/(m·K)) contributes to mitigating positive segregation near the riser line, although the effect remains relatively limited. Conversely, increasing the pouring height exacerbates top segregation defects. When the pouring height increases from 310 mm to 350 mm, the maximum segregation index in the riser transition zone rises from 2.18 to 2.67, an increase of 22.5%, primarily due to prolonged solidification sequence and solidification time. Additionally, research on the hereditary characteristics of segregation reveals that macroscopic segregation formed during the solidification of steel ingots is difficult to eliminate through subsequent processing stages. These findings provide a theoretical foundation for optimizing riser design in large steel ingots and for mitigating the hereditary effects of segregation.
In slab continuous casting processes, continuous casting rolls are subjected to alternating thermal-mechanical loads, which frequently lead to outer roll surface cracks, spalling, and other failures, posing a critical technical bottleneck to efficient and stable continuous forming. Composite rolls, combining high-temperature resistance and superior strength, offer a potential solution. However, the impact of composite structures on service performance remains unclear. A transient finite element model of 414N/42CrMo continuous casting composite rolls under simulated service conditions was established using Abaqus software to analyze the effects of varying slab surface temperatures and cladding thickness on temperature field distribution. The evolution of surface temperature gradients and the range of high-temperature sensitive zones are investigated, followed by structural optimization of the composite roll and post-service microstructural and mechanical property analysis of wire arc additive repaired continuous casting composite rolls. The results demonstrate that no significant temperature gradient exists between the 414N cladding and the 42CrMo substrate due to their similar thermophysical properties and compatibility. Variations in slab temperature significantly impact the overall temperature field of the composite roll, leading to an average surface temperature increase of approximately 35 ℃ per 100 ℃ rise in slab surface temperature. Conversely, changes in cladding thickness mainly influence the temperature within the surface layer of the continuous casting roll. The high-temperature sensitive zone has a radial thickness of 0-6 mm below the outer roll surface, where both the peak temperature and the temperature variation amplitude decrease with increasing distance from the roller surface. Controlling the remanufactured cladding thickness within the range of 4-8 mm is crucial to ensure excellent high-temperature oxidation corrosion resistance and thermal fatigue resistance of the outer roll surface. Post-service metallographic analysis revealed a progressive transition from boundary toward interior of the 414N cladding, followed by a distinct alloying element transition zone at the composite interface. Tensile-shear testing results showed fracture occurred within the 42CrMo substrate, indicating that the interfacial bonding strength exceeds the shear strength of the substrate material, confirming the achievement of complete metallurgical bonding and maintenance of good bonding integrity throughout service. Consequently, these findings provide theoretical support for the structural optimization design during the manufacturing and remanufacturing of continuous casting composite rolls, facilitating the recycling and reuse of iron and steel resources.
To investigate the influence of rare earth yttrium on the rolling contact fatigue (RCF) performance of GCr15 bearing steel, RCF tests were conducted to study the effects of varying yttrium mass fractions on the contact fatigue life of the steel. The mechanism by which yttrium treatment enhances the RCF life was elucidated through analysis of RCF crack initiation and propagation, austenitic grains, and inclusions in the steel. Under experimental conditions, the RCF life of the bearing steel samples exhibited an initial increase followed by a decrease with the increase of yttrium mass fraction. The bearing steels with a yttrium mass fraction of 0.002 8% exhibited the longest fatigue life, with a rating life L10 1.74 times that of yttrium-free bearing steels and an average life L50 2.75 times higher. The high-carbon chromium bearing steel with a yttrium mass fraction of 0.002 8% also had the highest grain size grade (9.98) and the smallest austenite grain size (7.99 μm). Reasonable addition of yttrium can significantly improve the size and morphology of inclusions in bearing steel. The size and quantity of non-metallic inclusions in bearing steel exhibit a trend of initial decrease followed by increase with rising yttrium mass fraction. At a yttrium mass fraction of 0.002 8%, the total oxygen cmass fraction reached its minimum value of 0.000 48%, leading to the smallest inclusion size and the lowest quantity, thereby improving the RCF life of the bearing steel. Cross-sectional observation of spalling pits in fatigue-fractured specimens and statistical analysis of inclusions demonstrate that when the total oxygen mass fraction reaches its minimum after rare earth yttrium treatment, the reduction in inclusion size, morphological transformation, modification of physical properties, and decreased number density collectively serve as the primary reasons for the improved RCF life of bearing steel.
The impact toughness of HRB500E steel bars with and without rare earth elements (REE) was systematically investigated through Charpy impact tests. The results show that the addition of REE significantly improves the impact toughness of the steel bars, as evidenced by the higher crack initiation and propagation energies of the REE-alloyed steel bars(HRB500ERE)compared to the non-REE-alloyed ones(HRB500E). This improvement is mainly attributed to the refining effect of REE on the inclusions in the steel and the optimization of the microstructure. Advanced characterization techniques, including optical microscopy (OM), electron backscattered diffraction(EBSD), and scanning electron microscopy(SEM), were employed to analyze the microstructures of the two types of steel bars in detail. It was found that the addition of REE caused most of the sulfide inclusions in the HRB500ERE steel to take on a spherical shape, which effectively reduces stress concentration and thus increases the crack initiation energy. Moreover, the addition of REE also significantly altered the microstructure of the HRB500ERE steel. The ferrite content increased from 45% to 57%, the ferrite grain size improved from grade 10.5 to 11.5, and the proportion of high-angle grain boundaries increased by 10 percent points. These microstructural changes contribute to enhanced plastic deformation capacity. EBSD analysis revealed a uniform distribution of Kernel Average Misorientation (KAM) values near the crack tip in the HRB500ERE steel, indicating significant orientation rotation and adequate plastic deformation capacity in this region. This enhanced plastic deformation capacity results in higher crack propagation energy for the HRB500ERE steel compared to the HRB500E steel. In conclusion, the addition of REE significantly improves the impact toughness of HRB500E steel bars by optimizing the morphology of inclusions and refining the microstructure. These research findings provide important theoretical and experimental support for the application of REE in high-strength steel bars.
The application of 310S stainless steel in hydrogen storage devices is constrained by the high alloy cost resulting from its elevated nickel content. A strategy involving partial substitution of nickel with Mn-N composite alloying has been proposed, leading to the development of a low-cost austenitic stainless steel designated as 06Cr22Ni10Mn8N. This approach provides a replicable alloy design paradigm for developing cost-effective metallic materials with enhanced high-temperature corrosion resistance and hydrogen embrittlement immunity for the hydrogen energy industry chain. Solution treatment temperature gradients of 1 000, 1 050, and 1 100 ℃ were set. Analytical techniques such as scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and electron backscatter diffraction (EBSD) were used to systematically compare its microstructure with that of 310S steel. Slow strain rate tensile tests were conducted to evaluate its mechanical properties and hydrogen embrittlement susceptibility. The results show that with increasing solution temperature, both steels exhibit grain coarsening and a simultaneous increase in twin volume fraction. While achieving a 40% reduction in material cost, the 06Cr22Ni10Mn8N steel demonstrates comprehensive improvement in mechanical properties over 310S steel: hardness increases by 3.0%, yield strength by 59.4%, tensile strength by 28.7%, and elongation after fracture by 13.2%. Meanwhile, the hydrogen embrittlement sensitivity index shows no significant decrease, which is mainly attributed to the synergistic alloying effect of Mn-N and the dispersion of hydrogen segregation by Σ3 twin boundaries.
In order to domesticate the production of wire rods for ultra-high strength ultra-fine steel wire, addressing the long-standing reliance on imports, ϕ5.5 mm high carbon steel wire rods were successfully developed by composition design, inclusion purification and segregation control. Furthermore, in collaboration with downstream partners and utilizing a complete, proprietary production technology, 36 μm diameter ultra-fine pearlitic steel wires with 5 000 MPa tensile strength were successfully fabricated. The developed SGC100Cr wire rod had a chemical composition (mass fraction) of 1.01%C, 0.21%Si, 0.39%Mn and 0.23%Cr, with strictly controlled levels of Al, Ti, V, O and N elements. To effectively control inclusions and segregation, the SGC100Cr wire rod was produced via a manufacturing process which consists of using high-quality raw materials, vacuum induction melting, vacuum consumable remelting for primary blooming, and controlled cooling and rolling in a high-speed wire mill. After initial drawing, two intermediate drawings, salt bath heat treatment, and brass plating, the resulting steel wire achieved a tensile strength of (1 530±20) MPa and an average pearlitic lamellar spacing of approximately 67 nm prior to the final drawing step. The final ultra-high strength steel wires with a diameter of 36 μm produced from SGC100Cr wire rods exhibited an average wire breaking rate of less than 1.7 occurrences per 10 000 km. This performance meets the demands for large-scale industrial production, and the drawing characteristics of the wire rods are comparable to those of imported counterparts.
In the traditional blast furnace smelting process, a significant amount of carbon emissions are usually associated, including carbon combustion and carbon loss. Ensuring stable and low-carbon production in blast furnaces is crucial for reducing the carbon footprint of the steel industry and mitigating adverse environmental impacts. Timely monitoring and predicting the thermal state of blast furnaces can optimize furnace operation, reduce energy consumption, minimize carbon losses, and enhance production stability. However, due to the complexity of blast furnace production processes and the lagging nature of detection methods, traditional means are inadequate for timely monitoring and prediction of blast furnace thermal states. Therefore, employing machine learning models to monitor blast furnace thermal states has become a new trend in blast furnace ironmaking development. In this paper, causality analysis and integrated learning were combined to mine the strong causal characteristic parameters of blast furnace thermal state by convergence cross mapping (CCM) method. Real-time monitoring of thermal state was realized based on Stacking ensemble learning model, and future state prediction was completed by combining variational mode decomposition-attention mechanism-bidirectional long short-term memory network (VMD-AM-Bi-LSTM) time series prediction model, forming a complete technical chain of "causal feature screening-real-time monitoring-time series prediction". The results show that the goodness of fit of thermal state monitoring of blast furnace is higher than 0.92, and the prediction accuracy (error within ±5%) of silicon content and temperature of molten iron reaches 82% and 88%, respectively, which meets the accuracy requirements for guiding production. Based on the internet platform of blast furnace ironmaking industry, this study realizes the monitoring and prediction of blast furnace thermal state, and promotes the transformation and upgrading of blast furnace ironmaking process to intelligence and low carbonization.
The hot metal temperature at the blast furnace taphole is a core process parameter that characterizes both the quality of the iron product and the thermal state of the furnace. Its accurate prediction is of great significance for process control in ironmaking. However, due to the nonlinear, highly dynamic, strongly time-delayed, and spatiotemporally multi-scale coupled nature of the blast furnace smelting process, traditional mechanism-based metallurgical modeling methods are inadequate for real-time temperature prediction and furnace condition diagnosis. To address this issue, a CNN-GRU-Attention deep learning prediction model based on heterogeneous big data was constructed. This model integrates the local feature extraction capability of convolutional neural networks (CNN), the advantage of gated recurrent units (GRU) in temporal dynamic modeling, and the adaptive self-attention mechanism to achieve deep mapping of the complex relationships between multidimensional process parameters and hot metal temperature. Emphasis was placed on discussing data preprocessing methods based on metallurgical mechanisms and network architecture optimization enhance model performance, leading to the establishment of an optimized architecture composed of a CNN-GRU cooperative feature extraction layer and an adaptive attention weight allocation layer. The results show that data cleaning guided by metallurgical mechanisms significantly improves data quality. The optimized model achieved an accuracy of 86% within a ±5 ℃ error range for hot metal temperature prediction on the test set. During industrial application, the model attained a hit rate of 88% within a ±10 ℃ range between predicted and actual values of continuous hot metal temperature. The developed model markedly enhances the digital representation capability of the blast furnace ironmaking process. The prediction system has been validated in industrial field trials, demonstrating considerable engineering application value and promising potential for broad adoption.
In ladle furnace (LF) refining process, accurately predicting alloying element yield is of great importance for controlling the chemical composition of molten steel, improving alloy utilization efficiency, and reducing smelting costs. In recent years, machine learning methods have been widely applied to metallurgical process modeling. However, most machine learning models typically rely on complex hyperparameter tuning and often require re-tuning hyperparameter when new data are introduced, limiting modeling efficiency. To address these challenges, a tabular prior-data fitted network (TabPFN)-based prediction model for Si element yield was established using actual production data. And this model′s performance was evaluated using multiple metrics and compared with the reference heat method, multiple linear regression model, and various machine learning models reported in previous studies. Shapley additive explanations (SHAP) was then employed to conduct both global and local interpretability analysis. The results show that the TabPFN model is superior to the existing models in key performance indicators such as R2, EMA, ERMS, hit rate and model reasoning time without a lot of hyperparameter tuning. The indicators reach 0.83, 1.59, 2.03, 98.4% and 0.430 s, respectively. Meanwhile, the SHAP analysis reveals the influence of each input feature on the Si element yield at the global level, and quantifies the influence of each input feature on the predicted Si element yield at the local level, so as to realize the efficient, high-precision and interpretable prediction of alloying element yield, offering new research ideas and technical pathways for metallurgical process modeling in the context of intelligent manufacturing in the steel industry.
XML
Related Articles
File
