To enhance the accuracy of Finite Element Analysis (FEA) used for validating product performance under high-speed impact conditions (such as collision, explosion, or drop) where materials exhibit different stress-strain behavior due to high strain rates compared to static conditions, this study conducted high-speed tensile testing to analyze the strain rate dependency of metallic sheets, where, unlike conventional methods that simply calculate strain rate based on test speed and initial gauge length, Digital Image Correlation (DIC) was utilized to precisely measure the actual full-field strain distribution and accurately derive the strain rate for each condition, allowing for a comparative analysis of the differences in the resulting stress-strain curves, while further validating the appropriate gauge length by comparing theoretical and DIC-based measured values and assessing the strain rate differential between the uniform and localized elongation regions (at maximum tensile strength and immediately before fracture), ultimately establishing a methodology for measuring the accurate strain rate through optimal gauge length determination in high-speed tensile tests.

Aluminum nitride (AlN) ceramics exhibit high thermal conductivity and insulation but require additives for densification. In this study, AYS4 and AYS5 specimens, prepared by adding Y₂O₃ and Sm₂O₃ and heat-treated above 1900 ℃, were evaluated along with related compositions. AYS4 and AYS5 achieved >99% of theoretical density, with thermal conductivity of 116~117 W/m·K, thermal expansion of 4.3-4.48 ppm/℃, and flexural strength of 291~295 MPa. With 2 wt% SiO₂ added, AYS6 showed improved strength (335 MPa) but reduced conductivity (89 W/m·K) due to glassy phases, highlighting a trade-off between properties. For ultra-high temperature AlN shaft bonding, Y₂O₃-CaO-Al₂O₃ with 10 wt% SiO₂ was used. Refining particle size from D50 ≈ 2.4 ㎛ to 1.4 ㎛ enabled defect-free joints at 1600℃. These results confirm that additive selection and particle refinement are critical for optimizing AlN ceramics for semiconductor and high-power device applications.

In this study, combined drawing-bending sheet metal forming process is suggested to prevent wrinkling in the heat shield forming for aircraft wheel component. When two-step bending process-that are composed of wheel arc bending followed by outer edge rim bending process-is applied, wrinkling occurs in the edge rim that are priorly bended during the arc bending due to the compressive deformation mode through the circumferential direction. Combined sheet metal forming process composed of simultaneous arc bending and edge drawing was investigated via finite element simulations. In the simulations, it is clearly shown that this process prevents wrinkling in the rim edge owing to the deformation mode applied in the edge rim drawing process. Then, an actual sheet metal forming tool was made, and a trial component was shaped via a newly developed process. The result shows that the heat shield component was fabricated without any wrinkles, like the finite element analysis.

A new low-alloy ER8 steel, referred to as ER8M (Modified ER8), was developed by tailoring its chemical composition to enhance mechanical strength and hot workability for railway wheel applications. Based on isothermal hot compression tests, a deformation processing map was established to identify optimal hot forging parameters, clearly distinguishing stable and unstable deformation regimes. Finite element-based forging simulations using the Forge software were conducted to validate the process window and to develop robust multi-stage forming strategies. Additionally, a parametric study was performed to investigate the effects of varying friction coefficients on folding behavior and load response. The results suggest that ER8M possesses high potential for industrial implementation in railway wheel forging processes, offering enhanced mechanical performance and greater process reliability.

Laser surface carburizing with graphene (GN) has been investigated to incorporate graphene's outstanding features into the substrate. The poor wettability of graphene with metals presents challenges for researchers regarding carbon diffusion into the Fe-substrate. Therefore, in this study, the two commonly used elements that act as the glue between graphene and metals in metal composite processing, i.e. Cu and Ni nanoparticles, were added to the graphene in the laser surface carburizing of AISI 430 for the first time. The study focused on adding Cu and Ni to graphene to increase the hardness of the alloy at greater depths. When the graphene layer thickness was optimized, the hardness at the interface and depth was higher in the Cu-GN case due to increased dissolution of Cu into Fe and a higher amount of oxygen released and dissolved during the diffusion of Cu-GN in the substrate. The hardness was greater at depth in Ni-GN when using a thinner layer of material, although the highest hardness was still less than that of Cu-GN with a larger layer thickness.

In this study, crush performance of two aluminum extrusions under four different extrusion conditions and three different heat-treatment conditions are investigated. Tensile tests were performed to obtain yield strength (YS), and V-bending tests were performed to obtain bending angle (BA). Crush tests were then performed to obtain different crush performance indicators. First indicator is crush grade (CG), which indicates apparent degree of deformation. Second indicator is mean crush force (MCF) between two primary peaks which gives load-bearing capacity under early stage of tests. Furthermore, relationships between YS, BA, and crush performance indicators are quantitatively investigated using Pearson's correlation coefficient. Results show that YS-MCF and BA-CG relationships are strongly correlated, implying that crush performance could be predicted cost-effectively from tensile and bending tests.

This study investigates the effect of normalizing processing parameters on ferrite/pearlite band structures in S355NL structural steel and establishes an optimized condition on the basis of fundamental mechanism. The ferrite/pearlite banding, induced by microsegregation during solidification and subsequent deformation, induced mechanical anisotropy and degrades impact toughness, particularly in the transverse direction. This study revealed that the continuity of band structure decreased with increasing normalizing temperature, with complete elimination achieved at 910℃. In contrast, normalizing duration exerted negligible influence on the band structure. In a large-scale forged ring made from S355NL steel, accelerated cooling significantly enhanced band suppression. This was distinctly observed after water spray cooling. Mechanical-property analysis indicated that the elimination of ferrite/pearlite bands at 910℃ significantly improved transverse-direction toughness and mitigated anisotropy, while tensile properties barely changed with the presence of these bands. Consequently, high-temperature normalizing at 910℃ combined with rapid cooling is identified as an effective processing strategy to ensure the toughness and isotropy required for offshore wind structural applications.

The hatch distance is a key process parameter in the directed energy deposition (DED) process that directly affects surface roughness and dimensional accuracy. Conventional approaches determine the optimal hatch distance through trial and error, which is time- and cost-intensive. In this study, a three-dimensional thermal analysis model was developed to predict the hatch distance in the DED process. The model was constructed using SYSWELD V16.0 and calibrated based on single-bead experiments. Inconel 718 was used as the material, and temperature-dependent thermal properties were obtained from JmatPro software. A Gaussian volumetric moving heat source was applied to simulate the actual heat input. The finite element analysis qualitatively predicted the occurrence of valley regions, verifying the validity of the model. The proposed method enables estimation of suitable hatch distance ranges prior to multi-track deposition, thereby reducing experimental efforts and improving process design efficiency in DED applications.

This study investigates the phase-specific mechanical behavior of hot-rolled DP980 steel through microstructural characterization, nanoindentation testing, and a reverse algorithm-based property extraction method. FE-SEM analysis revealed a pronounced banded microstructure consisting of ferrite and martensite aligned along the thickness direction. High-magnification images showed the coexistence of fresh and tempered martensite within martensitic regions, indicating heterogeneous transformation and tempering histories. WDS elemental mapping further demonstrated localized segregation of carbon and manganese in martensite bands, confirming that solute partitioning during hot rolling contributed to the formation of the banded structure. Phase-dependent mechanical properties-including elastic modulus, hardness, yield strength, and strain-hardening exponent-were quantitatively derived from nanoindentation data using the reverse algorithm. The resulting property distributions allowed the microstructure to be classified into four groups: ferrite, ferrite-dominant, martensite-dominant, and martensite. These clusters reflect variations in local phase fraction and tempering degree, highlighting the microstructural heterogeneity inherent to DP980 steel. Using the phase-specific stress-strain curves as input, the mixture rule accurately predicted the macroscopic tensile behavior of DP980 steel, showing close agreement with experimental measurements. These results demonstrate that the proposed nanoindentation-reverse algorithm framework provides a reliable and quantitative approach for linking microstructural features to bulk mechanical performance in dual-phase steels.

In this study, uniaxial tension and plane strain specimens in the dome-punch tests are newly proposed using finite element analysis to characterize the ductile fracture criteria of metal sheets. Conventional fracture tests require finite element analysis to determine the strain and stress at the fracture point, which is located at the center of the specimen in the thickness direction. In contrast, in the case of the dome-punch test, the strain at the fracture location can be measured directly using the DIC (digital image correlation), since the maximum strain is applied at the surface of the test specimen. To optimize the specimen geometries, three different shapes of uniaxial tension and plane strain dome-punch test specimens are proposed and compared through the finite element simulations. Mechanical properties of the STS430 stainless steel sheets, including the Hosford-Coulomb fracture criterion, are identified and used in the simulations. Subsequently, the newly designed specimens were verified through the actual dome-punch tests. The DIC-measured major/minor strain paths at the fracture initiation location show that the intended deformation states were successfully achieved in the dome-punch test.

Steel scrap represents a critical recyclable feedstock in the steel industry, serving as a key raw material in electric arc furnace steelmaking processes while significantly reducing CO₂ emissions and waste generation. Reliable classification and grading of steel scrap are essential for quality assurance in the steel recycling chain. However, conventional manual inspection depends heavily on operator expertise and becomes inefficient and inconsistent when large volumes must be processed. To overcome these challenges, this study, for the first time, proposes a panoptic segmentation-based framework for steel scrap classification, which simultaneously identifies both steel scrap types (semantic information) and individual instances (instance information) for all scrap objects in an image. The proposed model adopts the MaskDINO architecture with a Transformer encoder-decoder and a ResNet-50 backbone, and is trained on high-resolution 4K steel scrap images of truck cargos collected under diverse stockyard conditions, including daytime, nighttime, and rainy weather. Experimental results demonstrate that our model achieves 80.2% accuracy in counting steel scrap objects and 86.9% accuracy in area ratio estimation across diverse steel scrap classes. Notably, the model exhibits robust performance even in challenging scenarios involving overlapping or irregularly shaped steel scrap objects. These results indicate that our approach can serve as a practical foundation for large-scale, consistent, and real-time steel scrap grading in industrial environments.

This study addresses the design of the first drawing die in split drawing operations for automotive crossmembers manufactured from high-strength steel sheets. To mitigate forming defects, the die addendum shape was systematically optimized. The design parameters - punch opening width, upper bar radius, wall angle, and lower die radius - were defined to characterize the addendum geometry. The maximum values of thinning, wrinkling, and springback observed after the drawing process were individually considered as objective functions and minimized through the optimization procedure. The optimization was conducted using the Sigma module in AutoForm software. In addition, an impact analysis was performed to evaluate the influence of design parameters on forming defects. The results indicate that the optimized addendum design reduces forming defects compared to the initial design without an addendum.