International Journal of Material Forming最新文献

{"title":"Tool concepts and part characteristics of simultaneously formed sheet metal and long fiber-reinforced thermoplastic components","authors":"Amir Hajdarevic,&nbsp;Xiangfan Fang","doi":"10.1007/s12289-026-02082-y","DOIUrl":"10.1007/s12289-026-02082-y","url":null,"abstract":"<div><p>This study investigates a hybrid forming process for the simultaneous manufacturing of complex parts consisting of sheet metal and long fiber-reinforced thermoplastics (LFT), specifically glass fiber-reinforced polyamide 6 (PA6 GF40). This approach integrates sheet metal forming and LFT compression molding within a single tool and process step. Adhesive bonding between the metal and LFT was achieved during the process using a bonding agent previously applied to the sheet metal. Various tool concepts were developed and experimentally evaluated to form different open-ended, complex-shaped hybrid parts in a single-step process. Special attention was given to sealing concepts to prevent the press-out of LFT during forming. The manufactured hybrid parts were analyzed with respect to sheet metal forming behavior, geometrical accuracy, and LFT filling using optical measurement and cross-sectional analysis. Finally, a longitudinal control arm for the multi-link rear axle of the current VW Golf was manufactured and evaluated. The results demonstrate that the new integrated tool design enables reliable forming of hybrid parts with complex geometries. Excellent forming results were obtained for DC04 and DP600 steels, while DP800 steel showed localized thinning and partial underforming. The developed tool concepts ensured effective sealing and prevented LFT leakage, enabling full cavity filling when appropriate process parameters were selected, thereby proving the concept for real industrial parts. Thus, this process offers significant potential for lightweight multi-material structures in automotive applications by reducing process steps.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12289-026-02082-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Underwater explosive forming of thin metallic plates using repeated blast loading: improving die filling while controlling thinning","authors":"Mohammad Kouzehgaran,&nbsp;Hossein Khodarahmi,&nbsp;Milad Sadegh-Yazdi,&nbsp;Mojtaba Ziya-Shamami,&nbsp;Tohid Mirzababaie Mostofi","doi":"10.1007/s12289-026-02080-0","DOIUrl":"10.1007/s12289-026-02080-0","url":null,"abstract":"<div><p>Explosive forming offers a flexible route for shaping metallic sheets into dies, but a single explosion at room temperature can limit formability and die filling. This study experimentally and numerically investigates underwater forming of a 3 mm thick plate into a female die using repeated blast loading. Metallic specimens were subjected to single, double, and triple explosions, and the outcomes were assessed in terms of sheet thinning and die-filling percentage. In single-shot forming, increasing the charge mass beyond 5 g caused reverse deflection in the central region, led to a maximum thinning of 36.7%, and still failed to achieve full die filling (81.4%). In contrast, the use of second and third underwater blasts enabled complete die filling with reduced thinning and a lower total explosive charge. A coupled Eulerian–Lagrangian (CEL) model in ABAQUS/Explicit was employed to examine deformation velocity, Von Mises stress (SMises), equivalent plastic strain (PEEQ), and the Johnson–Cook damage initiation parameter (JCCRT). Double-explosion tests showed higher SMises, PEEQ, and JCCRT, while no red (damaged) zones appeared on either face of specimens formed with triple explosions. Using a rupture limit of 29% based on the material’s elongation, the region with thinning beyond this limit was shorter for triple than for double explosions. These findings highlight the benefits of repeated, especially triple, underwater blasts for improving die filling while controlling thinning in explosive forming.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physics-informed hybrid learning for bending analysis of multidirectional porous FG shells","authors":"Soufiane Benounas,&nbsp;Mohamed-Ouejdi Belarbi,&nbsp;Smain Bezzina,&nbsp;Ahmed-Amine Daikh,&nbsp;Li Li,&nbsp;Eugenio Ruocco,&nbsp;Mohammed Sid Ahmed Houari,&nbsp;Aman Garg","doi":"10.1007/s12289-026-02072-0","DOIUrl":"10.1007/s12289-026-02072-0","url":null,"abstract":"<div><p>The bending response of metal foam shells is dictated by a complex interplay of geometric curvature and graded porosity, which complicates accurate modeling. To address this issue, we create a thorough Physics-Informed (PI) Machine Learning framework that integrates an improved higher-order shear deformation theory with a finite element formulation to forecast the bending behavior of multidirectional functionally graded (MFG) porous metal foam doubly curved shells. A four-node quadrilateral element is developed to model both hard-core and soft-core architectures with five different porosity distribution patterns. The model's predictions match up very well with the benchmark results, which shows that it is numerically strong across a wide range of geometries and boundary conditions. Five hybrid machine learning surrogate models, PI-ANN + GBM, PI-ANN + RF, PI-ANN + DNN, PI-ANN + DT, and PI-ANN + XGBoost are trained on simulation data to predict displacements and stresses. Among them, PI-ANN + XGBoost and PI-ANN + DNN achieve the highest predictive accuracy, with R² values exceeding 0.98 and the lowest RMSE values. These models effectively capture the nonlinear stress–strain relationships while maintaining physical consistency through embedded physics constraints. Overall, the proposed framework modeling exhibits advanced hybrid learning to make better predictions and make predictions that are physically accurate.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Formability improvement mechanism of local laser hardened aluminium alloy sheet with gradient material model","authors":"Xiao-song Zhang,&nbsp;Shuai-feng Chen,&nbsp;Hong-wu Song,&nbsp;Xue-jian Yang,&nbsp;Wen-long Xie,&nbsp;Bao-cheng Yang,&nbsp;Shi-hong Zhang,&nbsp;Zuo-shan Wei,&nbsp;Dorel Banabic","doi":"10.1007/s12289-026-02084-w","DOIUrl":"10.1007/s12289-026-02084-w","url":null,"abstract":"<div><p>To solve the difficulty of forming complex parts of aluminium alloy, a method was proposed using local hardening by laser heat treatment (LHT) to improve the formability of the Al alloy sheet. In order to explore the influence of laser local hardening on deep drawing forming performance, the temperature field of laser treatment was simulated based on the measured data, and a sheet model coupled with mechanical property gradient was established. In addition, the mechanism of the improvement of material formability by the change of constitutive parameters was revealed by the combination of simulation and experiment. The results show that laser scanning can generate a stable gradient temperature field with little influence on the surrounding area. Furthermore, the multi-pass scanning allows for the design of property gradient differential sheets with varying ranges. It is proved that the LHT can effectively improve the plastic strength and the local hardening ability of Al alloy. The results of deep drawing simulations and tests show that the LHT mainly affects the forming ability of differential aluminium alloy sheets by regulating the gradient distribution of strain hardening exponent <i>n</i> and strain hardening coefficient <i>k</i>. The gradient decreasing distribution of <i>n</i> value from the heat treatment center to the surrounding area can not only improve the forming ability of the core area, but also alleviate the strain concentration in the periphery, so as to achieve a more uniform strain distribution in the large deformation zone. The gradient distribution of <i>k</i> can change the position and order of the yield behavior of the plate, and then change the material flow sequence, thus slowing down the material flow at the bottom fillet with higher <i>k</i> value and reducing the tendency of thinning and cracking. In particular, it is found that the gradient distribution of <i>k</i> value is the dominant factor to improve the forming ability of poor aluminium alloy sheet. When the gradient of <i>k</i> value distribution between the laser strengthening area and the surrounding material reaches the critical value, the fracture position of the deep drawing part will shift, which greatly delays the occurrence of necking and cracking processes and improves the overall forming ability of the sheet.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermal uniformity of plastic sheets in thermoforming: A combined simulation and experimental study","authors":"Peiyong Ni,&nbsp;Zhuonan Chen,&nbsp;Xiangli Wang,&nbsp;Xuewen Zhang,&nbsp;Xiang Li","doi":"10.1007/s12289-026-02086-8","DOIUrl":"10.1007/s12289-026-02086-8","url":null,"abstract":"<div><p>Plastic thermoforming relies heavily on uniform heating to ensure quality and efficiency, while uneven temperature distribution caused by improper thermal control remains a key engineering bottleneck. This paper presents a combined numerical and experimental investigation on the thermal heating characteristics of four typical thermoplastic sheets, namely polyethylene terephthalate (PET), polyvinyl chloride (PVC), polypropylene (PP) and polystyrene (PS), during infrared radiation heating. A transient thermal finite element model was established to simulate the temperature field evolution, and the Taguchi method was adopted to quantitatively evaluate the effects of heating temperature, heating distance, heating time and material type on temperature uniformity. The results show that heating distance exerts the most significant effect on sheet temperature uniformity, followed by heating temperature and material type, whereas heating time shows the least influence. Specifically, increasing the heating distance from 100 mm to 200 mm reduces temperature non-uniformity by approximately 24% relative to its initial value, which corresponds to a reduction of about 13% per 50 mm increment. A low temperature range between 250 °C and 300 °C contributes to better uniform heating. Among the four materials, PS exhibits the best comprehensive thermal response, while PP requires targeted optimization of heating parameters due to its combination of relatively low thermal diffusivity and high specific heat capacity. Furthermore, the dimensional change rate is positively correlated with temperature uniformity, suggesting that improving heating uniformity is essential to suppress thermally induced deformation. This provides theoretical guidance for the improved design of heating parameters and precise thermal control in plastic thermoforming.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel composite punch design for current-assisted bending forming","authors":"Guoxin Pang,&nbsp;Guoqing Chen,&nbsp;Minghao Li,&nbsp;Xuewen Chen,&nbsp;Junqing Guo,&nbsp;Nan Xiang,&nbsp;Yanchao Jiang,&nbsp;Hairui Zhang","doi":"10.1007/s12289-026-02079-7","DOIUrl":"10.1007/s12289-026-02079-7","url":null,"abstract":"<div><p>To address the challenges of poor formability, large springback, and low precision in the room-temperature V-bending of high-strength 2A97 Al-Li alloy, this study innovatively proposes a novel “metal matrix + rounded graphite” composite punch design for pulse current-assisted V-bending experiments. The effects of different mold structures and current paths on the temperature distribution, bending limit, forming precision, and microstructural evolution of 2A97 Al-Li alloy during forming were investigated. The results indicated that compared with room-temperature bending, the current-assisted bending process significantly reduces the bending load of 2A97 Al-Li alloy, increases the bending limit to 31° (whereas room-temperature bending results in fracture at only 115°), and improves forming precision. The “metal matrix + rounded graphite” composite punch enables precise local high-temperature control in the bending deformation zone, which enhances the plastic deformation capacity of the deformation zone and reduces the risk of cracking. Furthermore, this structure notably decreases the current density required for self-resistance heating of the alloy—only 11 A/mm² is needed to reach the target forming temperature of 400 °C, far lower than the 50 A/mm² required by traditional metal punches—and reduces the bending springback angle to 1°, corresponding to a 66%–80% improvement in bending precision compared with the 3°–5° springback angle of conventional metal punches.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization of the microstructure and mechanical properties of SAPH440 dual-phase steel for automotive parts","authors":"B. Zibaei,&nbsp;A. Kamyabi-Gol,&nbsp;M. Mazinani,&nbsp;B. Beidokhti","doi":"10.1007/s12289-026-02078-8","DOIUrl":"10.1007/s12289-026-02078-8","url":null,"abstract":"<div><p>The effect of various variables, including intercritical annealing temperature, holding time, initial condition, and quenching media, on the mechanical properties of an industrial-grade SAPH440 low-carbon steel sheet was investigated. The main objective was to improve the overall mechanical properties of the steel by simultaneously increasing the ultimate tensile-to-yield strength ratio (T/Y), ductility (in terms of reduction in area), and work hardening rate. The results showed that the initial condition of the steel was the most effective parameter (~ 55%), and holding time at an intercritical temperature was the least effective parameter (~ 0%). The effect of intercritical annealing time on the microstructure and mechanical properties was evaluated during uniaxial tensile loading. After annealing, five steel specimens were heated to an intercritical temperature of 730 °C for 5, 10, 15, 20, and 30 min., and subsequently water-quenched. It was found that the yield point phenomenon in specimens with a holding time of 5–15 min. was similar to that of the fully annealed specimen due to insufficient volume fraction of martensite (&lt; 5%). Cementite spheroidization in these three specimens was found to be responsible for the observed increase in the post-neck region with holding time.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Influence of the edge magnetic shield on transverse flux induction heating of endless-rolling strip","authors":"FangZhou Shi,&nbsp;MengYe Wang,&nbsp;Xinmiao Shao,&nbsp;WenShuai Wang,&nbsp;Meng Yan,&nbsp;Ce Ji,&nbsp;Huagui Huang","doi":"10.1007/s12289-026-02074-y","DOIUrl":"10.1007/s12289-026-02074-y","url":null,"abstract":"<div><p>Endless rolling represents a key development direction in short-process steel manufacturing, driven by its significant advantages: high dimensional accuracy, superior uniformity of microstructure and properties, high yield, and low process energy consumption. To satisfy the stringent temperature control requirements across the entire strip cross-section, these production lines are equipped with induction heating modules. However, the intrinsic magnetic field distribution of Transverse Flux Induction Heating (TFIH), combined with the skin effect, leads to severe overheating at the strip edges, which compromises temperature control accuracy across the strip’s width. This study, therefore, investigates the magnetic shield as a core control element to elucidate how its geometric parameters and spatial positioning regulate transverse temperature uniformity. For this purpose, an experimental platform for moving TFIH was constructed. The results indicate that an increase in the horizontal extension of the shield beyond the strip edge (<i>L</i>₁) elevated the temperature rise in the edge region from 766.0 °C to 794.0 °C. Conversely, increasing the coverage width over the strip (<i>L</i>₂) shifted the thermal peak further from the strip edge, causing the standard deviation of the transverse temperature to first decrease and then increase. Furthermore, increasing the vertical shield-to-strip distance (<i>H</i>₁) from 7.5 mm to 20 mm significantly reduced the transverse temperature standard deviation from 4.2 to 2.4, whereas increasing the shield thickness (H) across the same range yielded only a modest reduction to 3.65. The accuracy of the finite element analysis was validated through a comparison of simulation and experimental results. The findings of this study establish a robust theoretical foundation for the design and optimization of high-performance TFIH systems.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interpretable failure-load prediction of externally bonded FRP-strengthened reinforced concrete beams using mechanics-based machine learning and consensus sensitivity analysis","authors":"Amit Thoriya,&nbsp;Husain Rangwala,&nbsp;Tarak Vora,&nbsp;Bharat Shah","doi":"10.1007/s12289-026-02075-x","DOIUrl":"10.1007/s12289-026-02075-x","url":null,"abstract":"<div><p>Accurate prediction of the failure load of reinforced concrete (RC) beams strengthened with externally bonded Fiber-Reinforced Polymer (FRP) composites remains challenging because four competing failure modes—concrete crushing, FRP rupture, intermediate crack debonding, and plate-end debonding—are governed by different but interacting flexural, bond, and anchorage mechanisms. Existing machine learning (ML) studies on this application rely exclusively on single-method interpretability and underutilise mechanics-based feature engineering. This study develops an interpretable ML framework using 624 externally bonded FRP-strengthened RC beam specimens. Sixteen mechanics-based engineered features encoding force equilibrium, stiffness compatibility, and geometric ratios are engineered, forming a 39-feature input space. Four deterministic ensemble models (Random Forest, XGBoost, LightGBM, CatBoost) and the probabilistic NGBoost are trained with Optuna hyperparameter optimization. XGBoost achieves the highest R² of 0.9743 (RMSE = 15.73 kN), while CatBoost yields the lowest MAE of 9.21 kN and the most uniform failure-mode-wise accuracy. A four-method consensus sensitivity analysis combining SHAP, permutation importance, Morris screening, and Sobol indices consistently ranks the total tensile force capacity T_total = A_sf_y + A_ff_fu as the most influential mechanics-guided predictive descriptor within the assembled database. Five of the top ten consensus-ranked features are mechanics-based engineered quantities. LIME analysis of three representative specimens provides local examples that are broadly consistent with the global sensitivity trends. Separately, NGBoost with cross-conformal calibration provides 95% prediction intervals with 97.6% empirical coverage and a mean width of 70.9 kN, the first reported coverage-guaranteed uncertainty quantification for this domain. All models substantially outperform ACI 440.2R-23, demonstrating 44–49% improvement in R² and 45–64% reduction in MAE.</p></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Composition-holdout validation of gaussian-process surrogates for reliability-aware multi-property prediction in AA6061 hybrid composites","authors":"Obinna Onyebuchi Barah,&nbsp;Abdulrazak Jinadu Otaru,&nbsp;Ige Bori,&nbsp;Sami F. Khalil","doi":"10.1007/s12289-026-02081-z","DOIUrl":"10.1007/s12289-026-02081-z","url":null,"abstract":"<div><p>Machine-learning studies of composite materials can report excellent accuracy even when repeated specimens from the same formulation are distributed across training and test sets, because the design variable itself is not truly withheld. This study quantifies that risk using an experimentally generated AA6061 dataset containing 12 independently prepared specimens at 0, 5, 10, and 15 wt% total hybrid reinforcements. The reinforcement comprised plantain stem fibre ash, eucalyptus wood ash, and periwinkle shell particulate at a fixed 1:1:1 mass ratio. Ultimate tensile strength, Brinell hardness, and Charpy impact energy were modelled using linear regression, quadratic ridge regression, and Gaussian-process regression (GPR) with Matérn-5/2 and radial-basis kernels. Two validation tasks were deliberately separated. Replicate-holdout validation tested new specimens at known compositions, whereas leave-one-composition-out validation tested an unseen reinforcement loading. Matérn GPR achieved R² values of 0.992, 0.994, and 0.994 for tensile strength, hardness, and impact energy under replicate holdout, but the corresponding composition-holdout R² values fell to -0.070, -0.699, and − 0.874. Nominal 95% GPR interval coverage also decreased to 50%, 75%, and 75%, respectively. After fitting all available data, the conservative multi-property screening score was highest near the experimentally observed H10 region. Because the model did not recover the H10 peak when that composition was withheld, this result is interpreted only as a model-smoothed description of the measured response and not as evidence of an optimized formulation. Within this sparse experimental dataset, the findings demonstrate the practical importance of aligning the validation unit with the intended composite-design task and identify 2.5, 7.5, and 12.5 wt% as informative next compositions for expanding the experimentally supported design space.</p><h3>Graphical Abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":591,"journal":{"name":"International Journal of Material Forming","volume":"19 3","pages":""},"PeriodicalIF":2.6,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}