A Multi-Scale Computational Scheme for Prediction of High-Cycle Fatigue Damage in Metal Alloy Components 2024-26-0430

Aerospace structural components grapple with the pressing issue of high-cycle fatigue-induced micro-crack initiation, especially in high-performance alloys like Titanium and super alloys. These materials find critical use in aero-engine components, facing a challenging combination of thermo-mechanical loads and vibrations that lead to gradual dislocations and plastic strain accumulation around stress-concentrated areas. The consequential vibration or overload instances can trigger minor cracks from these plastic zones, often expanding unpredictably before detection during subsequent inspections, posing substantial risks. Effectively addressing this challenge demands the capability to anticipate the consequences of operational life and aging on these components. It necessitates assessing the likelihood of crack initiation due to observed in-flight vibration or overload events. Such predictive abilities have far-reaching implications for refining material processing, design strategies, and inspection protocols. However, the intricate nature of fatigue damage prediction, compounded by manufacturing-induced microstructures, complex stress scenarios, and associated uncertainties, adds to the complexity. While current assumptions treat material elastic moduli as isotropic on a macroscopic level, the microstructural perspective reveals anisotropic and inhomogeneous plastic shear strain components, originating from microstructure variations and grain dislocations. Statistical modeling of these microstructures is pivotal, requiring synthetic counterparts mirroring experimental ones. To address this, a multi-scale computational methodology emerges, accurately estimating fatigue-induced plastic strain by linking microstructure-derived damage evolution to macroscopic damage. This approach incorporates a polygonal finite element scheme, enabling the modeling of intricate grain geometries. Experimental fatigue tests validate the accuracy of fatigue damage predictions. Ultimately, this multi-scale computational framework bridges the macroscopic material behavior with microstructural influences, diminishing the reliance on extensive empirical data from numerous tests. Moving forward, a comprehensive strategy involving systematic data generation and validation tests is outlined, laying the foundation for robust predictive capabilities. By assimilating microstructural insights into macroscopic predictions, this approach promises to revolutionize aerospace component design and longevity assessment.