Abstract. The finite element method (FEM) combined with the crystal plasticity (CP) model was used to predict crystallographic texture evolution during multi-pass asymmetric rolling of the polycrystalline aluminum with tilted entry into the rolling gap. The FEM macro-scale model of the rolling process with various asymmetry was performed using Abaqus commercial software. The asymmetry was introduced by difference in the diameters of cooperating rolls and various tilt angles of the rolled bar entry into the rolling gap. Predicted stress distribution over sample volume was used as the boundary conditions in the crystal plasticity meso-scale model of the plastic deformation. The CP meso-scale model developed by Leffers and Wierzbanowski [1], [2] was used as a post-processing procedure. The predicted evolution of the rolling texture distribution on the rolled bar thickness turned out to be in good agreement with that measured for specimens rolled with the same parameters as the ones used for modeling. Impact of the rolling parameters on the texture evolution was found.

Abstract. Synergies between neural networks and traditional surrogate modelling techniques have emerged as the forefront of data-driven engineering. Neural network-based surrogate models, trained on carefully selected experimental data or high-fidelity simulations, can predict behaviours of complex systems with remarkable speed and accuracy. This review examines the current state and recent developments in neural surrogate technologies, highlighting their expanding roles in engineering design optimisation and generation. It also covers various feature engineering methods for representing 3D geometries, the principles of neural surrogate modelling, and the potential of emerging AI-driven design tools. While feature engineering remains a challenge, especially in parameterising complex designs for machine learning, recent advancements in code/language-based representations offer promising solutions for digitalising various design scenarios. Moreover, the emergence of AI-driven design tools, including text-to-CAD models powered by large language models, enables engineers to rapidly generate and evaluate innovative design concepts. Neural surrogate modelling has the potential to transform engineering workflows. Continued research into geometric feature engineering, along with the integration of AI-driven design tools, will speed up the use of neural surrogate models in engineering designs.

Abstract. Twin roll casting (TRC) has a big potential to produce thin strip material from magnesium alloys and enables the production in an economic manner. However, the final properties of TRC strips, such as microstructure and texture, are influenced by the twin roll casting conditions. In this work the development of a Mg-6.8Y-2.5Zn-0.4Zr alloy (WZ73) during different twin roll casting conditions, varying the twin roll casting speed, were studied. As a reference the strain and strain rate were determined. After twin roll casting the microstructure is inhomogeneous over the strip thickness and consists of a network-like structure of the LPSO phases and the α-Mg matrix. The α-Mg matrix is made up of dobulites (flake-like structures), which is already known for this alloy. [1,2] Typical defects of the twin roll cast strips were observed as well. It was also revealed that the twin roll casting conditions have a big influence on the precipitation, morphology, and phase fraction of the LPSO phases. For example, the phase fraction increases with the strain decreasing whilst the thickness of the precipitated phases increases with an increased strain. In all samples kink bands and yttrium enriched precipitations within the network like structures were detected. No dynamic recrystallization nor grain boundaries were detected. The resulting textures revealed the activation of basal slip and non-basal slip, but the intensities are small, regardless of the twin roll casting conditions.

Abstract. The manufacturing processing and the mechanical properties of particle reinforced metal matrix composites are strongly dependent on their microstructural characteristics. In this research, 3D models in microscale of tungsten alloys reinforced by Zr(Y)O2 particles (W-Zr(Y)O2) were established to investigate their uniaxial compression deformation behaviours. The effects of particles contents, size and their distribution on the compressive properties of W-Zr(Y)O2 alloy were discussed. The mechanical behaviours of the reinforced W alloys were improved by increasing the content of Zr(Y)O2 particles. With the same particles content, the strength of the reinforced alloys increased with smaller size particles. With the same particle size and content, the stress concentration was reduced with more homogeneous distribution of the reinforced particles. The predicted strengths with 3D models are compared with the experiment data, which exhibits high prediction accuracy.