Characterization, Modeling, and Prediction of Fatigue Damage Under Variable Amplitude Loading
Fatigue is an omnipresent problem in structural systems subjected to vibrations, and can lead to catastrophic failures, injuries, and loss of life. Quantitative prognostication of fatigue damage is still a largely unresolved and critical problem of engineering practice. This award supports fundamental research establishing the connection between material fatigue and structural dynamics. Such a link enables fatigue damage modeling and prediction and is an indispensable component of effective structural health monitoring and condition-based maintenance technologies. Successful achievement of the project objectives will enhance safety and performance of critical infrastructure and machinery, and will provide wide-ranging benefits to the U.S. economy and society. The project has a multidisciplinary nature incorporating material science, dynamical systems modeling, fatigue testing, and nonlinear time series analysis. The research approach and the project findings will also enrich engineering education and broaden participation of underrepresented groups in cross-disciplinary research.
Current understanding of fatigue mechanisms at the micro-scale is not sufficient to provide information applicable to macro-scale fatigue life estimation. Furthermore, existing models and analyses of the dynamic behavior of damaged structures do not directly yield insight into fatigue mechanisms or fatigue prediction. This award will support the study of fatigue evolution across its different stages, its interaction with structural dynamics and loads, and the effect of loads on fatigue dynamics. A combination of analytical, numerical, and experimental studies will be conducted to investigate the interaction between fatigue accumulation and structural oscillations. The findings will be employed for practical fatigue life prediction under variable amplitude loading conditions. The project will formulate a modeling approach that permits the exploitation of nonlinear load characteristics for fatigue life prediction. The resulting damage model and method for fatigue life prediction – using new, physically meaningful, and easy to estimate load metrics – will be instrumental in designing against fatigue, and enabling effective fatigue prognostication.