Carbon segregation is a persistent defect in continuous casting of large-diameter steel billets, leading to deteriorated mechanical properties and compromised service reliability. Conventional empirical or machine learning models generally estimate segregation indices but cannot resolve local variations of carbon distribution across billet sections. In this work, a microstructure-informed convolutional neural network (CNN) framework is proposed to predict and map carbon segregation in 600 mm round 42CrMo steel billets. A comprehensive dataset comprising microstructural images and corresponding carbon content measurements was established. The customized CNN achieved a testing accuracy of 81.3% with a mean absolute error of 0.012 wt.% and showed good robustness in out-of-sample validation. Compared with transfer learning models (VGG16, VGG19, etc.), the customized architecture exhibited superior generalization on this domain-specific dataset. Contrast-enhanced imaging significantly improved predictive performance, while Gradient-weighted Class Activation Mapping visualizations highlighted key microstructural regions correlated with carbon distribution, providing interpretability. This study demonstrates a proof-of-concept methodology to achieve quantitative mapping of segregation patterns in large-diameter 42CrMo billets, offering a complementary tool to traditional metallurgical analysis and providing a workflow that may support future data-driven research on segregation formation mechanisms and process optimization in steel casting when extended to additional steels and casting conditions.

To overcome the industrial limitation of conventional deformation-induced ferrite transformation (DIFT), which is typically restricted to temperatures below 810 degrees C, this study employed Al alloying in an Fe-0.06C-1.5Mn steel. The addition of 1.5 wt % Al reduced the Gibbs free energy difference for the austenite-to-ferrite transformation by approximately 140 J/mol. This reduction effectively lowered the activation barrier for DIFT, significantly destabilized austenite, and successfully induced DIFT within an elevated temperature range of 800-1100 degrees C. Rolling at 1100 degrees C yielded DIFT ferrite grains of approximately 5 mu m with a maximum ferrite fraction of 83 %, while rolling at 800 degrees C with low strain rates produced ultrafine DIFT ferrite of similar to 2.36 mu m. The enrichment of C and Al elements at prior austenite grain boundaries enhances local stability, suppressing boundary nucleation and forcing intragranular nucleation. This new phenomenon can be termed the "Continuous DIFT". This work provides a novel pathway for achieving DIFT at high temperature through Al alloying and developing highstrength-toughness steels.

Utilizing internal carbon (C) vacancies in vanadium carbide (VC) precipitates is crucial for mitigating hydrogen embrittlement in high-strength steels. This study employs first-principles calculations to establish a comprehensive evaluation framework for alloying elements (Ti, Cr, Nb, Mo) influencing hydrogen trapping within VC, based on four atomic-level indicators: trap density (vacancy formation), trap stability (segregation energy), trap accessibility (diffusion barrier), and trap depth (escape barrier). Results indicate that Cr and Mo doping reduces C vacancy formation energy, increasing the density of irreversible traps. Uniquely, Cr significantly lowers the hydrogen diffusion barrier between vacancies, facilitating hydrogen migration into the carbide interior while maintaining trap irreversibility. Consequently, Cr doping appears to be a promising alloy design strategy for increasing the number of accessible irreversible hydrogen traps among the investigated elements, thereby potentially enhancing the hydrogen resistance of high-strength steels.

Based on a limited experimental dataset, a machine learning-based model has been developed for the prediction of properties and design of composition-process of NiAl-based nanoprecipitates strengthened non-oriented silicon steel. Validation experiments were conducted to evaluate the model accuracy and understand the coordination between magnetic and mechanical properties. Among different machine learning algorithms, the random forest algorithm exhibited superior accuracy with the determination coefficient over 90%. Based on the objective function set in the alloy development framework, three new alloy composition and processing routes were rapidly designed. The explored non-oriented silicon steels exhibited recrystallization microstructure with average grain size of 22 - 31.8 mu m and medium gamma-fiber texture. After aging, a high density of dispersed B2-type NiAl-based precipitates with average size of similar to 5 nm and low lattice misfit of less than 0.4% was formed. High yield strengths of 826 and 854 MPa were achieved, with elongations of 9.8% and 4% for the two candidate steels, respectively, showing high reliability in strength prediction. The high strength increment of similar to 250 MPa was attributed to the NiAl-based nanoprecipitates, which likely resided within the critical size range for cutting-to-bypass transition. High electrical resistivity and combined effects of grain size and texture contributed to high magnetic induction B-50 of 1.63 - 1.64 T and low iron loss P-1.0/400 of similar to 24 W/kg. Paramagnetic NiAl-based nanoprecipitates with low lattice misfit had a negligible effect on magnetic properties. This approach offers a feasible route for prediction and integrated design of high-strength non-oriented silicon steels with tunable composition-process combinations.

Medium-Mn steel typically exhibits continuous yielding after hot-rolling and intercritical annealing (HRA), whereas cold-rolled and annealed (CRA) specimens show discontinuous yielding. This distinct difference is revealed by this study to depend critically on the coherence of ferrite/austenite interfaces. We found that HRA specimens with lath-shaped microstructure are dominated by semi-coherent interfaces, while CRA specimens with equiaxed microstructure are primarily dominated by incoherent interfaces. The formation mechanisms of both types of interfaces were clarified. Austenite memory effect was found in HRA specimens which allows reversed austenite and ferrite to inherit the specific orientation relationships between displacive-formed martensite and retained austenite, thereby forming semi-coherent interfaces. In contrast, cold rolling eliminates the austenite memory effect, leading to the formation of incoherent interfaces. We subsequently revealed the mechanism by which dislocations, through their interactions with both types of interfaces, govern the yielding behavior of medium-Mn steel. In CRA specimens dominated by the incoherent interfaces, the initial number of dislocations within individual grains is insufficient. Mobile dislocations must nucleate at grain boundaries by overcoming an activation energy barrier during deformation, causing an abrupt stress drop and resulting in discontinuous yielding. Interestingly, most dislocations can transmit across the semi-coherent interfaces in HRA specimens. An individual prior austenite grain contains sufficient dislocations to achieve macroscopic plastic deformation, and dislocation interactions within prior austenite grains enable dislocation multiplication and work hardening, leading to continuous yielding. These findings propose a universal strategy to suppress discontinuous yielding in ultrafine-grained materials by introducing coherent or semi-coherent interfaces enabling dislocation transmission.

Medium manganese steels often suffer from a strength-ductility-toughness trade-off, particularly regarding transverse impact performance, making it challenging to unify these properties in a low-cost system without expensive alloying. In this study, we report a low-cost 1.3 GPa grade V-Nb microalloyed steel achieving 29.7% elongation and 36.3 J/cm2 transverse toughness at -50 degrees C, realized through V-Nb synergistic microalloying and precise intercritical annealing. The optimized IA620 sample possesses a refined duplex microstructure containing 37.0 vol.% highly stable retained austenite (RA). Unlike unstable RA, this mechanically stable RA enables a sustained transformation-induced plasticity effect, ensuring continuous work hardening. Furthermore, the stable RA triggers crack tip blunting, which, combined with crack deflection by high-density high-angle grain boundaries, simultaneously improves crack initiation and propagation energies. Consequently, the optimized steel achieves an outstanding synergy of properties compared to microstructures with unstable austenite.

The Al-Cu-Y-Zr alloy was prepared by low-pressure spray forming (LPSF), successfully constructing a multiscale heterogeneous structure of θ, θ', and Al3 (Zr, Y) phases, and Al-Cu-(Zr, Y) nanoclusters containing nanoscale stacking faults. This structure increased the tensile strength of the alloy from ∼194 MPa in the deposited state to 431 MPa, while maintaining an elongation of ∼8.3%. For the first time, a Nanoscale stacking fault(NSF) was discovered inside the θ' phase. This defect not only provides reinforcement but also serves as a dislocation source to coordinate local strain, endowing the precipitated phase with a self-toughening function. The deformation process drives the dynamic recrystallization path of discontinuous dynamic recrystallization (DDRX), continuous dynamic recrystallization (CDRX), and then back to DDRX. During the stretching process, the strong texture acts as a hard skeleton to constrain the strain distribution, inducing the grains to rotate towards the<111>//RD soft orientation and initiate multiple slips in the critical fracture zone, thereby transforming from intergranular brittleness to a high-energy absorbing ductile fracture.This proposed model of nano-cluster-stabilized matrix-coordinated deformation with θ'-strong texture and optimized strain distribution provides the theoretical basis for designing high-performance aluminum alloys.

Evaluation of corrosion’s effect on the development length of prestressing strand is crucial for pretensioned prestressed concrete structures serving in the aggressive environment. This paper presents an analytical model for predicting the development length of corroded prestressing strand, in which the effects of strand corrosion on the transfer length and flexural bond length are considered separately. The effect of corrosion on the initial prestressing transfer length is analyzed by considering the combined effects of initial releasing cracking and subsequent corrosion-induced cracking on the evolution of confinement around the prestressing strand. Thereafter, beyond the transfer stage, the effect of strand corrosion on the additional flexural bond length is confirmed based on the calculation of maximum bond strength for various corrosion degrees, in which the effect of strand rotation along its axis is considered for the pull-out failure mode in a well-confined condition. Comparison of analytical and experimental results shows that this model can reasonably predict the development length of corroded prestressing strand.

To overcome the strength-toughness trade-off in aircraft landing gear steels, a novel low-cost 4 wt% Mn medium-Mn steel was designed, focusing on the synergistic regulation of retained austenite (RA) stability and martensitic variant collaboration via deep cryogenic treatment (DCT). It was found that DCT at -60 degrees C induces a "thermodynamic selection" effect, preferentially eliminating unstable blocky austenite (>1.5 mu m) and retaining 15.1 vol% of ultra-refined, highly carbon-enriched (0.70 wt%) RA. This optimized RA exhibits superior mechanical stability, providing sustained work-hardening through a steady transformation-induced plasticity (TRIP) effect. Consequently, the steel achieves a yield strength of 1832 MPa, ultimate tensile strength of 2226 MPa, and an excellent total elongation of 16.3%. Simultaneously, the intense phase transformation driving force during -60 degrees C DCT reshapes the crystallographic features, promoting a "platform-like" near-equiprobable distribution of the 24 martensitic variants. This variant self-accommodation effectively alleviates local strain concentration and enhances crack initiation energy (26.7 J). Furthermore, DCT refines the microstructure and increases the density of high-angle grain boundaries and toughening variant pairs (V1/V2), significantly improving impact toughness (40.8 J/cm(2) at room temperature and 36.8 J/cm(2) at -60 degrees C). This study reveals the physical essence of improving damage tolerance through cryogenic-induced microstructural refinement and variant self-accommodation, offering a cost-effective solution for high-performance aerospace applications.

Further enhancing the performance of oxide dispersion strengthening (ODS) IN718 superalloy and overcoming the stress concentration induced by heterogeneous oxide particles remain key challenges. In this study, IN718 superalloys with uniformly dispersed oxides were prepared by arc consumable remelting. These oxides form coherent interfaces with both the gamma matrix and the gamma '' precipitates, thereby achieving simultaneous improvements in both strength and ductility. The yield strength of the ODS IN718 increased by 7.6% at 650 degrees C and 17.2% at 700 degrees C, while the elongation after fracture was enhanced by approximately 100% and 101%, respectively. These improvements stem from the heterogeneous nucleation of gamma ''/gamma ' at the dispersed yttrium oxides during aging, which refines their distribution and spacing, thereby strengthening dislocation pinning. Moreover, yttrium oxides inhibit grain boundary migration and stabilize grain size during hot deformation. The composite precipitates of gamma '' and yttrium oxides also suppress dislocation climb and reduce dynamic recovery, limiting sub-grain formation. The increased stored deformation energy and dislocation reactions promote a deformation mechanism dominated by primary and secondary deformation twinning, ultimately enhancing uniform deformation capability.

A hot-rolled medium-Mn steel achieves an exceptional strength—ductility synergy: yield strength of 1.8 GPa, ultimate tensile strength of 2.2 GPa, and total elongation of 18%. This performance results from an architectural design in which banded prior austenite promotes V1/V4 martensite variants that activate intragranular slip, while nano-sized retained austenite, stabilized via carbon partitioning, supplies sustained transformation-induced plasticity. The work establishes an intrinsic link among prior austenite structure, martensite variant selection, and slip system activation, providing a new theoretical basis for overcoming the strength—ductility trade-off.

The pursuit of high-quality steel production necessitates precise control of operational parameters, particularly in the pretreatment of hot metal (PHM), where traditional methods may lead to inefficiencies. Consequently, the hierarchical fusion learning architecture (HFLA), an innovative system leveraging statistical theory, machine learning, and intelligent optimization techniques, was presented for accurate sulfur content prediction in PHM. By employing a strategy-driven fusion approach, HFLA enhances feature extraction via stacked kernels, incorporating Tikhonov regularization to guide the meta-learner predictions. Validation results from steel mill production demonstrate an impressive 4.3

The multi-stage manufacturing process of non-oriented silicon steel (NOSS) involves complex interactions among composition, processing parameters and properties. When quality deviations arise, pinpointing critical process variables from extensive production data poses significant technical challenges. To enable rapid root-cause diagnosis of property anomalies, this study proposes a data-driven machine learning framework for decoding composition–process–property (CPP) correlations in NOSS. A data preprocessing strategy is developed, and a Bayesian-optimized categorical gradient boosting (BO-CatBoost) algorithm is applied to construct predictive models for CPP relationships, achieving high-precision performance prediction. Additionally, an intelligent optimization framework combining data analytics and particle swarm optimization (PSO) is implemented to identify dominant process parameters influencing property deviations and formulate corresponding optimization strategies. Results show titanium (Ti) content exhibits the strongest correlation with core loss under specified processing conditions. A controlled reduction of Ti content by 0.0007 mass

Continuous casting is a critical and complex process in steel manufacturing, in which product quality is highly sensitive to process parameter variations. Early identification of anomalous parameters and timely process optimization are, therefore, essential for reducing quality defects and production losses. However, existing data-driven approaches often face limitations in practical industrial applications, particularly in achieving high accuracy under strict response time requirements. In this study, a data-driven framework is proposed to address the above limitations by identifying key anomalous parameters and inferring feasible parameter adjustments under operational constraints, with interpretability analysis incorporated to enhance consistency with metallurgical knowledge. The proposed framework was trained and evaluated on a real-world continuous casting dataset comprising 191,254 records, where the data were temporally split, and the final 25 pct was reserved as an independent test set. Experimental results demonstrate strong performance in both anomaly identification and process optimization, achieving a Top-10 anomaly identification rate of 96.64 pct and an optimization hit rate within ±5 pct of 91.54 pct, demonstrating its effectiveness and practical applicability.

This study fabricated composite magnetic powder cores by mixing low-loss FeSiBCuNb amorphous/nanocrystalline powder with high-permeability FeNi powder. The effects of different FeSiBCuNb/FeNi mass ratios on the microstructural features and high-frequency magnetic properties of the composite cores were systematically investigated. With increasing FeNi content, the porosity in the powder cores decreased and the density increased, resulting in higher effective permeability and lower loss. When the FeNi content was 50-60 wt%, the best overall magnetic performance was achieved: compared with the cores without FeNi addition, the effective permeability increased by 66.8-76.9%, and under 20 mT at 2 MHz, the total loss was 842.2-872.6 kW/m3, representing a reduction of 19.4-22.2%. However, when the FeNi content reached 80 wt%, the loss increased rapidly and the DC bias performance decreased to 68%. These results indicate that introducing an appropriate amount of smallsize FeNi powder into FeSiBCuNb powder cores enables composite cores with high permeability and low loss at high frequency. Owing to these characteristics, this composite system shows potential for power-electronic magnetic components that require miniaturization and high efficiency, such as high-frequency power inductors and filter chokes.

In the twin-roll strip casting process, high-temperature molten metal solidifies upon contact with casting rolls, leading to complex heat transfer during solidification. However, few studies have simultaneously addressed the solidification heat transfer and wettability of copper alloys. This study experimentally investigates the interfacial heat transfer behavior of the Cu-9Ni-6Sn alloy under varying coating thickness and surface roughness. Temperature evolution profiles were obtained to characterize heat transfer performance under different substrate conditions. Numerical simulations were further employed to validate the experimental temperature curves and to further examine the wettability and spreading behavior of copper alloy droplets during solidification on metal substrates, confirming the robustness of the computational model. The results indicate that the copper substrate, due to its high thermal conductivity, exhibits faster heating and cooling rates and maintains lower temperatures compared to a steel substrate. Increasing the coating thickness reduces heat transfer efficiency, however, the combined effects of interfacial thermal resistance and coating thermal resistance diminish the marginal gains in thermal insulation. Furthermore, increasing substrate surface roughness impedes liquid alloy flow due to microscale asperities, promoting the formation of localized eddy currents and stagnation zones that enhance heat accumulation and localized heat transfer to the substrate. Simulations further reveal that the number of secondary droplets formed during splashing is lower on steel substrates than on copper substrates, and the droplet spreading distance increases with impact velocity within the range of 0 similar to 0.3 m/s. These findings provide valuable insights into the heat transfer and solidification mechanisms of copper alloys during twin-roll strip casting.

Overcoming the strength–ductility–toughness trade-off in ultra-high-strength steels remains a major challenge for aerospace applications, especially for aircraft landing gear steels. In this work, a V–Nb–Mo microalloyed medium-Mn steel was designed, and a lamellar martensitic microstructure with an excellent strength–toughness synergy was obtained through low-temperature forging followed by quenching, deep cryogenic treatment, and tempering. Lowering the forging temperature promoted prior austenite grain (PAG) refinement, which favored a close-packed (CP) group-dominated martensite variant selection mode and significantly increased the density of high-angle grain boundaries (HAGBs). In addition, owing to phase transformation inheritance, martensite laths were arranged preferentially along specific crystallographic orientations, resulting in a highly oriented lamellar structure. This unique microstructure enhanced strength and crack resistance through grain refinement and interfacial strengthening. More importantly, it promoted coordinated plastic deformation by facilitating long-range dislocation glide and a more homogeneous dislocation distribution, thereby alleviating the ductility loss induced by excessive dislocation accumulation. Consequently, the low-temperature forged specimen (L) achieved a yield strength of 1934 MPa, an ultimate tensile strength of 2292 MPa, a total elongation of 17%, and a room-temperature impact toughness of 46.46 J/cm2, outperforming the widely used 300M steel. These findings clarify the strengthening and toughening mechanisms of the lamellar martensitic structure and provide guidance for the design of next-generation high-performance landing gear steels.

To address the lack of a systematic and energy-efficient design methodology for casting roll cooling channels in twin-roll strip casting, this study proposes an integrated framework combining numerical simulation, data-driven modeling, and multi-objective optimization. The axial and circumferential cooling are decoupled and systematically analyzed to clarify their distinct roles in cooling uniformity, intensity, and energy consumption. Axial cooling is investigated through numerical simulations of different inlet structures, and an optimized guide-plane inlet is proposed to ensure stable flow and reduce pumping power. For circumferential cooling, an artificial neural network model is developed using channel diameter, spacing, distance from the roll surface, and cooling water flow rate as design parameters. Based on 6720 parameter combinations, a practical design scheme is developed to simplify the cooling channel configuration while maintaining cooling performance. A process window for roll sleeve thinning and water flow rate is defined to ensure long-term thermal stability. Under the constraint that the circumferential temperature difference remains below 1 K, multi-objective optimization achieves a 57.83 K reduction in maximum roll surface temperature and a 74.14% decrease in pumping power. This work provides a systematic, energy-efficient approach for casting roll cooling, offering practical guidance for industrial implementation.

To address the challenges of high residual stress and significant quenching distortion in thin-gauge martensitic steel plates, this study systematically investigated a novel temperature-controlled quenching and deformation (TCQD) synergistic control process. Four distinct processes with varying deformation conditions were designed and compared with the conventional quenching-tempering (Q-T) method. The results show that the TCQD process can produce quenched martensitic steel with an optimized microstructure and mechanical properties without the need for tempering, with only minimal reductions in tensile strength (4.7%) and hardness (1.4%). Notably, it also reduces the residual stress by 7.9% and the dislocation density by 24.2%, while it improves the ductility by 8.6% and the toughness by 3.6%. The slow cooling within the martensitic transformation range in the TCQD process effectively suppresses the transformation, which is the fundamental reason for the relatively low dislocation density and residual stress in the new process. Moreover, variations in the transformation driving force and cementite precipitation behavior under different processing conditions are the main factors responsible for the differences in dislocation density and residual stress among the samples produced by the new process. The overall improvement in mechanical properties stems from the optimized regulation of nano-carbide precipitation behavior and size (reduced by 24.4%), a significantly decreased dislocation density, and an increased high-angle grain boundary (HAGB) density due to the absence of tempering (increased by 37.1%). This study suggests that this new process has the potential to replace the conventional Q-T process, substantially increase production efficiency, and provide a promising solution for quenching thin-gauge martensitic steel plates with high flatness and low residual stress.