Novel Probabilistic Multi-Scale Modeling and Sensing Framework for Fatigue Life Prediction of Aerospace Structures and Materials: DCT Project.

摘要

This project entails three complementary tasks with the objective of integrating them into a unified platform for advancing a novel probabilistic multi-scale modeling and sensing framework for fatigue life prediction. In the first task (Ghosh, JHU), and integrated system of novel spatial-temporal multi- scale computational models for image based modeling of deformation and fatigue crack nucleation in polycrystalline titanium alloys is developed. A novel wavelet transformation based multi-time scaling algorithm and homogenized continuum anisotropic plasticity and probabilistic fatigue crack nucleation models are developed. In the second task (Rokhlin, OSU), a comprehensive experimental methodology and inverse models are developed for structural health monitoring and noninvasive assessment of fatigue sensitive microstructures in Ti alloys. The aim is to provide realistic data input for computational predictive models under development in this program. In the third task (Millwater, UTSA), the distribution of weakest-link type material properties for larger material volumes is predicted from evaluation of small microstructural models with Extreme Value Theory. A property distribution associated with a grain boundary is used to extrapolate the distribution for a grain which is then used to predict the property distribution for a volume of the polycrystalline material.