Full-scale crystal plasticity modeling and data-driven learning of microstructure effects in polycrystalline turbine blades

• Full-scale crystal plasticity simulations of a turbine blade with up to 40,000 grains. • Interface-aware meshing strategy for embedding realistic microstructures into complex geometries. • Dislocation-density based crystal plasticity model including GNDs under finite thermomechanical loads. • Grain-wise parallel domain decomposition solved with Adaptive Multipreconditioned FETI (AMPFETI). • Graph neural networks trained to predict grain-averaged fields from morphological descriptors.