Twinning-induced plasticity (TWIP) steel is an ideal material for impact-resistant structures and energy absorption because of its high product of strength and elongation. However, compared with other advanced high-strength steels, the relatively low yield strength of TWIP steel is one of the important shortfalls that significantly limits its engineering applications. To enhance the comprehensive properties of TWIP steel, a machine learning design strategy that integrated comparative modelling, SHAP analysis, and multi-objective optimization were adopted in this study. Initially, various machine learning algorithms were compared for their predictive accuracy based on normalized data (273 entries) regarding the microstructure and properties of TWIP steel. Then, performance prediction models for yield strength, tensile strength, and elongation were established. SHAP analysis was subsequently employed to assess the significance and explicit laws of composition and microstructures in these three target properties, identifying key elements that enhance the overall performance. Furthermore, two new TWIP steels with high yield strengths and high products of strength and elongation were developed via multi-objective optimization. Under conventional hot forging + hot rolling + cold rolling + annealing processes, the two designed TWIP steels had yield strengths of 585 MPa and 560 MPa, tensile strengths of 1055 MPa and 1101 MPa, elongations of 55% and 58.5%, and products of strength and elongation of 58.0 GPa% and 66.4 GPa%, respectively. The yield strengths of the designed TWIP steels significantly improved while maintaining a reasonable product of strength and elongation. This work provides important references for the rational development of new TWIP steels.

In this paper, Fe-3%Si ultra-thin ribbons prepared by the planar flow casting (PFC) technique were subjected to temper rolling and annealing treatments. The microstructure and texture evolution during this process were examined through experimental measurements coupled with crystal plasticity finite element (CPFE) simulation to assess the feasibility of preparing ultra-thin non-oriented silicon steel using PFC ribbons. The results indicate that the PFC ribbons exhibit a significant columnar crystal structure, and {001}-oriented grains comprise over 30%. After being annealed, the grains with different orientations grew uniformly, the texture components were basically unchanged, and the {001} texture was well preserved. When annealing was carried out after temper rolling with a reduction rate of 7%, uneven grain growth was observed, and the growth tendency of the {001} grains, especially, surpassed that of the {111} grains, with an elevated temperature which peaked at 950 °C, where the proportion of {001} grains was maximal. When being annealed after temper rolling to 15%, grains of other orientations showed significant growth at each temperature, while the {001} grains did not show an obvious growth advantage. Utilizing the CPFE, the deformation-stored energy distribution of each characteristic-oriented grain was simulated, and it was shown that compared to the 15% rolling reduction rate, the deformation-stored energy accumulation of {001}-oriented grains after being rolled to 7% reduction was significantly lower than that of {111}-oriented grains. It suggests that the larger stored energy difference makes {001} grains show a stronger growth advantage based on the SIBM mechanism during annealing, after being rolled with a reduction rate of 7%. Overall, for the synergistic optimization of microstructure and texture, rolling with a 7% reduction rate followed by annealing at 950 °C in a hydrogen atmosphere is most advantageous.