Effect of columnar crystal ratio on microstructure, composition segregation, precipitate distribution and microtexture of Hi-B steel hot rolled plate

This study presents a systematic investigation into the evolution of non-metallic inclusions during the industrial continuous casting production of high-carbon chromium bearing steel. Conventional 2D characterization techniques and X-ray-based Micro-CT were employed to compare the number, type, size, and spatial distribution of inclusions throughout the LF-RH double-refining process. The results show that Al2O3 and MgO·Al2O3 are the dominant inclusion types formed during melting and refining, with MgO·Al2O3 exhibiting a strong thermodynamic tendency for formation. Due to differences in the resolution of the two characterization methods and variations in inclusion characteristics among the samples, inclusions with an equivalent diameter larger than 5 μm show better consistency between the results obtained from the two techniques. This finding revises the assumption that X-ray-based CT techniques inherently yield larger equivalent inclusion diameters. The spatial distribution and morphological characteristics of inclusions are identified as key influencing factors. In ultra-clean bearing steel characterized by non-uniform inclusion distribution and predominantly spherical oxide inclusions, when only inclusions larger than the critical equivalent diameter of 5 μm are considered, the equivalent diameters obtained from 3D characterization are significantly smaller than those derived from 2D methods. This study provides vital industrial data for the future development of precise multidimensional characterization of inclusions in ultra-clean special steels. It also supports the establishment of a reliable 2D-3D conversion model that incorporates characteristic parameters.

Mold flux crystallization behavior dictates its continuous casting metallurgical performance, yet how static magnetic fields affect its crystallization kinetics remains unclear. This study explored 0 and 40 mT magnetic field effects on mold flux isothermal crystallization at 1150–1300 °C via FactSage simulations, isothermal experiments and microstructural characterizations. Magnetic fields altered crystallization kinetics and structural evolution, advancing nucleation, prolonging total crystallization time at all temperatures but 1300 °C and changing pathways, lowering activation energy and pre-exponential factor to facilitate nucleation but reduce growth rate. They suppressed phase separation and coarsening for uniform, dense microstructures, with no obvious effects on precipitate phases or elemental segregation. This work supports optimized magnetic field application in continuous casting mold flux design theoretically and experimentally.

In industrial environments, the internal quality of continuously cast steel shapes is typically evaluated afterwards by means of macroetch testing. Nevertheless, this approach becomes increasingly difficult as the product dimensions grow larger. When process parameters change — for example, with an increase in productivity — it is advisable to have a virtual macroetch available to perform a preliminary quality assessment.

Purpose The purpose of this study is to address the instability of meniscus behavior and the difficulty of matching optimal electromagnetic braking (EMBr) parameters under high-speed continuous casting conditions. A novel decision-based EMBr control method is proposed to achieve stable meniscus behavior and improved initial shell quality in slab continuous casting. Design/methodology/approach A transient three-dimensional multiphase and multiphysics coupling model of the mold (MPF-Mold) is developed to analyze the effects of casting speed and double-ruler EMBr (FC-Mold) parameters on molten steel flow and meniscus behavior. Eight evaluation indicators are extracted, and a combined principal component analysis-technique for order preference by similarity to ideal solution (PCA-TOPSIS) approach is used for weight assignment and multi-objective optimization. Findings The results indicate that the uniformity index of initial shell thickness U(Thick shell) has the highest weight (0.451), followed by maximum meniscus velocity and level fluctuation amplitude. The optimal EMBr condition (lower field 0.33T, upper field two-thirds of the lower) reduces meniscus velocity from 0.52 m/s to 0.34 m/s, limits level fluctuation to ± 2 mm, increases initial shell thickness by 0.5 mm and improves U(Thick shell) from 0.842 to 0.895. Originality/value This study integrates numerical simulation with decision analysis for the first time to establish a rapid matching method of EMBr parameters for meniscus control. The developed PCA-TOPSIS-based optimization framework provides a practical and efficient strategy for stabilizing the meniscus and enhancing slab surface quality in high-speed continuous casting.

Abstract TThe aging process has significant effects on the microstructures and properties of copper-based alloys. Furthermore, the effects of aging are increased by conventional solid-solution processing to obtain these materials as supersaturated matrices. In this study, it was observed that the electrical conductivity (85 %IACS) and tensile strength (830 MPa) of a peak-aged, cold-drawn Cu-4.5wt.%Ag alloy prepared without solid-solution treatment were higher than those of the corresponding alloy prepared with solid-solution treatment. The eutectic-, grain-, precipitation-, and dislocation-strengthening effects were investigated via SEM, EBSD, TEM, and XRD analyses. There exist supersaturated Ag atoms in the as-cast Cu-Ag alloy prepared by controlled cooling continuous casting. During the peak-aging process, solute Ag atoms in the as-cast Cu-Ag alloy precipitated from this supersaturated matrix.This precipitation behavior ultimately achieved a strengthening effect equivalent to that of the alloy prepared without solid-solution treatment. The difference in the tensile strengths of the Cu-4.5wt.%Ag alloys fabricated via three different routes was attributed to precipitation and dislocation phenomena. Furthermore, residual solute Ag atoms were found to dramatically decrease electrical conductivity, which explains why the alloy without solid-solution treatment exhibited high conductivity.

GRPO-TST: Group relative policy optimized time series tokenization for long-horizon equipment-state forecasting in continuous casting

Abstract To address the limitations of traditional continuous casting billet surface defect detection methods in practical applications, such as insufficient real-time performance and low detection accuracy, this study proposes an improved YOLOv8-based billet surface defect detection model, termed LRCR-YOLOv8. First, a Lightweight Spatial Cross-Down sampling (LSCD) structure
 is introduced into the network neck, which effectively reduces model parameters and computational complexity while maintaining detection accuracy. Second, an RFCAConv module is embedded into the backbone network to enhance the capability of capturing fine grained surface features of cast billets. Finally, based on the C2f module, a C2f_RFCBAMConv structure is constructed. which consists of one 3 × 3 convolutional layer and two 1 × 1 convolutional layers, thereby further optimizing feature extraction and improving the overall detection performance. Experimental results on the publicly available NEU-DET dataset demonstrate that the proposed model achieves superior detection performance across six typical defect detection categories. Specifically, the precision, recall, and mAP reach 77.8%, 73%, and 77.2%, respectively, representing improvements of 1.0 percentage point, 0.5 percentage point, and 2.7 percentage points over the original model. These results effectively validate the practical applicability and technical advancement of the proposed method in billet surface defect detection, providing technical, support for early warning and accurate identification of surface defects in the continuous casting process.

Explainable deep learning-aided design of alloyed carbon steel with improved hot ductility for manufacturing scalability

A continual causal learning architecture with multiscale graph attention for robust mold level fluctuation prediction in smart continuous casting systems

To address the existing challenges in mold slag thickness measurement-such as the susceptibility of contact sensors to high-temperature degradation and the limitation of non-contact methods to detecting only the upper slag surface-this study proposes an integrated approach that fuses millimeter-wave radar and eddy current sensors for measuring mold slag thickness in a continuous casting mold. The method innovatively combines two sensing principles: the millimeter-wave radar employs an improved FFT-CZT2 high-precision ranging algorithm to perform high-resolution scanning of the solid slag upper surface, reconstructing its topography (error: ±1 mm), while Mel-frequency cepstral coefficients (MFCC) are applied to extract features from the radar intermediate-frequency signals, combined with an enhanced PSO-BP neural network algorithm to predict the thickness of the solid slag layer (error: ±5 mm). Concurrently, an eddy current sensor monitors the liquid slag-molten steel interface position (error: ±1 mm). Through dual-sensor data fusion, the upper surface topography data and solid slag thickness obtained from the radar are spatially registered in three dimensions with the molten steel level information derived from the eddy current sensor. This integration ultimately enables the non-contact synchronous measurement of three key parameters within the mold: solid slag layer thickness, liquid slag layer thickness inversion, and molten steel level. Furthermore, by reconstructing the upper slag surface morphology, the method successfully resolves practical issues such as uneven material distribution, local material deficiency, or excessive feeding. Preliminary experimental verification confirms that the proposed method maintains stable performance even under high-temperature and complex environmental conditions. It thus provides a real-time, accurate, and full-cross-section monitoring solution for mold slag in continuous casting, offering significant practical value for the development of smart steel plants.

Abstract Unsteady bulging in continuous casting of steel slabs refers to transient bending of the strand shell between the rolls of the secondary cooling zone. It can cause defects in the final product, impacting surface quality and structural integrity, and is thus unwanted. Because roll pitches significantly influence unsteady bulging, optimizing these pitches might avoid or at least reduce this phenomenon already in the design process. In this respect, the literature in this field lacks a systematic method for selecting the roll pitches. The current work proposes a two-step multi-fidelity method for selecting individual roll pitches for each roll pair of a continuous slab caster, based on an optimization problem and an unsteady bulging model . Simulation results using a model validated at a real plant show that the optimized roll pitches can avoid unsteady bulging for a wide range of system parameters and are, in this respect, superior to uniform roll pitches. The proposed method can be readily transferred to other continuous slab casters, particularly supporting the design of new installations. Dedicated experimental validation of the optimized roll pitch layouts in existing plants would require major plant modifications and is therefore outside the scope of this work.

For the submerged entry nozzle (SEN) samples obtained during continuous casting of Ti-IF steels with different Ti contents, the characteristics of the SEN deposits were systematically studied using X-ray fluorescence, X-ray diffraction, and scanning electron microscope with energy-dispersive spectroscopy (SEM-EDS). The cleanliness of various Ti-IF steel slabs was analyzed using automatic scanning microscope, SEM-EDS, and oxygen/nitrogen analyzer. The experimental results show that the nozzle deposits for the Ti-IF steel with high Ti content can be divided into three layers. However, for the Ti-IF steel with low Ti content, it consists only of two layers. As the Ti content in the Ti-IF steel increases, the Ti element in the nozzle deposits aggregating significantly, and the cleanliness of Ti-IF steel slabs worsens gradually, and the possibility and severity of the nozzle clogging also increase accordingly. Therefore, two different clogging mechanisms for Ti-IF steels with various Ti contents were obtained in this paper.

Macrosegregation in continuous casting slabs remains a critical defect that adversely affects the homogeneity and mechanical properties of the final rolled products. Industrial experiments were conducted on E355 steel continuous casting slabs to investigate the effects of electromagnetic stirring (EMS) and soft reduction (SR) on the evolution of slab macrosegregation. Furthermore, the inheritance of segregation from the slab to the rolled plate was analyzed. The results indicate that the equiaxed crystal ratio increases and the centerline segregation decreases with increasing stirring intensity. The application of both secondary EMS and SR minimized the centerline segregation in the slab. When the current intensity was increased from 0 A to 320 A in continuous stirring mode, the equiaxed crystal fraction increased from 22.52% to 32.52%, and the centerline segregation index decreased from 1.23 to 1.17. Compared with the continuous stirring mode, the alternating stirring mode promoted a more pronounced increase in the equiaxed crystal ratio and a further reduction in the centerline segregation. The centerline segregation in the slab correlates with the banded structure observed in the rolled plate. A higher degree of slab centerline segregation corresponds to a more severe banded structure and greater fluctuations in the mechanical properties of the plate. Through parameter optimization, the recommended settings are an alternating stirring mode with a current of 320 A at 5 Hz and an SR amount of 3 mm. Under these optimized conditions, the equiaxed crystal ratio of the slab increased to 35.22%, the centerline segregation index dropped to 1.15, and the banded structure in the rolled plate was reduced to grade 2.0. Consequently, the standard deviations of the tensile strength and elongation were 8.03 MPa and 1.1%, respectively.

Abstract Oxide scale formation during thin-slab continuous casting has a complex structure, which is influenced by mold flux contamination, that modifies interfacial reactions during solidification, subsequent reheating, and descaling. While individual aspects of the oxidation behavior of carbon steel have been previously examined, the synergetic effects of mold flux contamination during continuous casting and subsequent reheating on scale modification and the efficiency of hydraulic descaling remain inadequately studied. This study quantitatively examines the effect of flux composition on oxide scale evolution, adhesion, and hydraulic removal in low-carbon steel under simulated industrial conditions. Slab samples with as-cast, cleaned, and flux-coated surfaces were reheated to 1065 °C in a controlled oxidizing atmosphere and immediately descaled using a computer numerical control (CNC)-controlled high-pressure water-jet system. The developed procedures closely simulate industrial conditions. The resulting scale morphologies and residuals after descaling were analyzed using cross-sectional scanning electron microscopy (SEM), Raman spectroscope and quantified through image J analysis. The results demonstrate that the mold flux composition significantly affects the structural evolution and adhesion characteristics of the oxide scale, thereby influencing its hydraulic removal performance.

Heterogeneous Graph Modeling for Predictive Safety Control in the Continuous Casting Process

Multi-scale modeling and machine-learning-assisted optimization of convex-roll end-of-solidification reduction in ultra-thick continuous-casting steel slabs

Rapid velocity magnitude flow field prediction in electromagnetically stirred continuous casting via a proper orthogonal decomposition-based reduced-order model

Quantitative correlation between solidification front steel speed and solidified shell surface temperature in a slab continuous casting mold

The Al-Cu-Mg-Ag alloys have excellent specific strength, good heat resistance and have a wide range of applications in the aerospace and automotive industries. However, industrial production of such alloys is a great challenge owing to the addition of Ag, which limits their widespread application. In this work, the industrial continuous cast and continuous extrusion (Conform) processes were employed to prepare Al-5Cu-0.4Mg-0.1Zr (-0.4Ag) alloys, and the effects of Ag addition on the microstructural characteristics and mechanical properties during processing and heat treatment were investigated. The results indicated that Ag addition significantly refined grain size, increased high-angle grain boundary fraction and grain orientation difference in as-cast Al-5Cu-0.4Mg-0.1Zr (-0.4Ag) alloys, and suppressed excessive grain coarsening during homogenizing annealing. During Conform, Ag further refined grain size, increased subgrain number and enhanced grain orientation difference in extruded alloys. For the aging heat treatment, the T6 process demonstrated superior strengthening effects compared to the T5 process. Following T6 treatment, Ag promoted efficient and uniform precipitation of the Ω (Al2CuMgAg) phase and then significantly enhanced peak hardness (160 HV) and tensile strength (511.46 ± 2.06 MPa). Additionally, Ag accelerated second-phase dissolution throughout the entire process and produced finer, denser ductile dimples on tensile fracture surfaces to gain good strength-ductility balance.