Research Projects

Fatigue is one of the most common failure modes in mechanical systems. Therefore, fatigue reliability assessment is crucial, especially for safety-critical engineering applications like the aerospace industry. Irreversible deformation due to the materials’ evolving microstructure is the origin of fatigue damage. In the traditional fatigue reliability assessment, such microstructural aspects are not explicitly considered; empirical tools, such as the SN diagram and Paris’ law, are used. Consequently, fatigue reliability assessment is expensive and time-consuming in the aerospace industry due to its overreliance on material testing at different levels – coupons, elements, sub-components, components, and full-scale articles. However, coupling physics-based models (bottom-up approach) with fewer experiments can open the possibility of expedited fatigue reliability evaluation at a relatively lower cost. Different physics-based modeling strategies exist for fatigue damage estimation. Crystal plasticity is the most prominent as it models the material deformation at the mesoscopic length scale and captures anisotropy associated with elastic and plastic deformation. Crystal plasticity has been widely used in academic research; however, it has yet to realize its full potential for practical applications in industry. A lack of uncertainty quantification, challenges associated with initializing the initial state of the material (e.g., residual stress), and high computation cost have restricted its integration in regular engineering workflows. The proposed research aims to address some of these challenges by proposing (a) a rigorous uncertainty quantification framework for a crystal plasticity model of an aerospace alloy, (b) a reduced-order probabilistic crystal plasticity simulation model for component-level fatigue assessment, and (c) a methodology for incorporating statistical variability associated with residual stress within a microstructure-sensitive fatigue reliability assessment framework. With this, the proposed work aims to deliver residual stress initialized, uncertainty-quantified, physics-based, probabilistic, and computationally efficient fatigue framework for the reliability assessment of aerospace components. The novelty of this proposal lies in incorporating real-time material states, such as residual stress and microstructural variability, within a microstructure-sensitive reliability framework that enables multiscale fatigue assessment, connecting mesoscale deformation mechanics to macroscale component reliability and addressing both design and in-service reliability. The probabilistic crystal plasticity framework developed here will offer a path toward industry adoption by integrating uncertainty quantification as a trust-building measure and leveraging reduced-order modeling to make crystal plasticity accessible and practical for fatigue assessments in aerospace applications.