This work examines how rolling speed, feeding rate, and pass schedule—with a constant total reduction—affect the stress–strain fields, rolling force, and texture evolution of Al–Cu–Sc alloy sheets. A coupled finite element (FEM) and viscoplastic self-consistent (VPSC) framework is employed and compared with EBSD measurements to connect macroscopic fields with microscale texture changes. Results indicate that increasing rolling speed raises the effective strain rate and deformation heating, which lowers peak rolling force and improves in-plane stress homogenization on the RD–ND plane, while enhancing surface–core incompatibility and residual-stress gradients along the ND–TD direction. A higher feeding rate mainly intensifies work hardening, slightly elevates rolling force, and promotes near-surface stress/strain localization; in contrast, multi-pass schedules redistribute deformation between passes and reduce macroscopic stress concentration. Texture analyses show a speed-induced rotation from $⟨001⟩$ toward $⟨111⟩$ orientations, strengthening shear-related components; KAM maps suggest increased local orientation gradients consistent with higher stored energy. The simulations capture the principal experimental trends across conditions, supporting the use of the combined framework for trend-level process guidance. Overall, the findings clarify parameter–microstructure relationships and provide a basis for designing rolling routes that balance force reduction, stress uniformity, and texture control in Al–Cu–Sc sheets.

This work examines how rolling speed, feeding rate, and pass schedule—with a constant total reduction—affect the stress–strain fields, rolling force, and texture evolution of Al–Cu–Sc alloy sheets. A coupled finite element (FEM) and viscoplastic self-consistent (VPSC) framework is employed and compared with EBSD measurements to connect macroscopic fields with microscale texture changes. Results indicate that increasing rolling speed raises the effective strain rate and deformation heating, which lowers peak rolling force and improves in-plane stress homogenization on the RD–ND plane, while enhancing surface–core incompatibility and residual-stress gradients along the ND–TD direction. A higher feeding rate mainly intensifies work hardening, slightly elevates rolling force, and promotes near-surface stress/strain localization; in contrast, multi-pass schedules redistribute deformation between passes and reduce macroscopic stress concentration. Texture analyses show a speed-induced rotation from 〈001〉 toward 〈111〉 orientations, strengthening shear-related components; KAM maps suggest increased local orientation gradients consistent with higher stored energy. The simulations capture the principal experimental trends across conditions, supporting the use of the combined framework for trend-level process guidance. Overall, the findings clarify parameter–microstructure relationships and provide a basis for designing rolling routes that balance force reduction, stress uniformity, and texture control in Al–Cu–Sc sheets.

Microstructure-resolved phase-field models capture crack–reinforcement interactions in hyperelastic composites, but repeated platelet-scale fracture calculations are costly for structural analysis. This work develops a microstructure-informed effective phase-field model for hexagonal boron nitride (hBN)-reinforced silicone rubber. Fractography supports a predominantly matrix-dominated, reinforcement-deflected fracture idealization and motivates a mechanical-role-separated closure. A finite-strain phase-field branch represents the fracture-active response, while a low-order auxiliary branch represents reinforcement-mediated load transfer. Both contribute to the effective stress and share one phase-field state, but only the tensile energy of the fracture-active branch drives fracture evolution. Four closure parameters are constrained using 33 proportional multiaxial loading paths and the complete pre-peak histories of all four in-plane first Piola–Kirchhoff stress components. A coefficient-separated reconstruction accelerates the offline search, while the spatial phase-field formulation is retained for structural calculations. The identified parameters are frozen and transferred to re-entrant-corner and edge-defect problems and to a separate five-repeat-per-formulation tensile dataset. Eleven of the 12 structural comparisons have pre-peak NRMSE below 3%, with a maximum of 3.732%, and retain the dominant localization patterns. For the 11 cases with complete terminal reference responses, fracture-displacement differences remain within approximately 3%. Experimental peak-normalized pre-peak RMSE ranges from 1.20% to 2.98%. Structural calculations are typically accelerated by about 4–7×, reaching 11.04× in the fastest benchmark. The framework therefore provides a compact, fully offline route to structure-level fracture prediction without online microstructural solves.

Designing hyperelastic porous microstructures under finite strain is challenging because bending, buckling, contact, and densification interact to produce nonconvex and one-to-many relationships between topology and response. We present HyperDiff, a conditional diffusion framework that reformulates inverse design as probabilistic sampling rather than deterministic regression. A compact B-spline encoding of the target force--displacement curve captures the system's energy-evolution trend and provides temporal and mechanical context that guides denoising toward physically consistent configurations with the desired multistage deformation behavior. The workflow integrates Gaussian random field (GRF)-based topology generation, constitutive calibration, large-deformation finite-element simulations, and quasi-static compression experiments. Across held-out and interpolated targets, the generated microstructures accurately reproduce sequential deformation stages and global responses, with deviations typically below 10%, while preserving manufacturability and one-to-many design diversity. The current implementation focuses on two-dimensional unit cells under quasi-static compression. However, the framework is extensible to 3D, multiresolution, and multiphysics systems. By combining physics-aware conditioning with generative sampling, HyperDiff provides a practical front end for mechanics-based design workflows, including the design of programmable soft actuators and impact-energy absorbers with tunable plateaus, as well as the rapid exploration of nonlinear architected materials for soft and deformable systems.

Designing hyperelastic porous microstructures under finite strain is challenging because bending, buckling, contact, and densification interact to produce nonconvex and one-to-many relations between topology and response. We present HyperDiff, a conditional diffusion framework that reformulates inverse design as probabilistic sampling rather than deterministic regression. A compact B-spline encoding of the target force--displacement curve captures the system’s energy-evolution trend, providing temporal and mechanical context that guides the denoising process toward physically consistent configurations with the desired multi-stage deformation behavior. The workflow integrates Gaussian random field (GRF)-based topology generation, constitutive calibration, large-deformation finite-element simulations, and quasi-static compression experiments. Across held-out and interpolated targets, the generated microstructures accurately reproduce sequential deformation stages (bending-buckling-densification) and global responses, with deviations typically below 10%, while preserving manufacturability and one-to-many design diversity. The current implementation focuses on two-dimensional unit cells under quasi-static compression, yet the framework is extensible to 3D, multi-resolution, and multi-physics systems. By combining physics-aware conditioning with generative sampling, HyperDiff establishes a practical front end for mechanics-based design workflows, applicable to programmable soft actuators, impact-energy absorbers with tunable plateaus, and rapid exploration of nonlinear architected materials for soft and deformable systems.