@Article{cmc.2026.087556,
AUTHOR = {Gang Xu, Yeda Lian, Leike Yang, Hao Li, Yonggang Yang, Lanjie Niu},
TITLE = {A Multi-Level Equivalent Driving Force Framework for Fatigue Life Prediction of Nickel-Based Single-Crystal Superalloys under Stress Ratio and Notch Effects},
JOURNAL = {Computers, Materials \& Continua},
VOLUME = {89},
YEAR = {2026},
NUMBER = {2},
PAGES = {--},
URL = {http://www.techscience.com/cmc/v89n2/68839},
ISSN = {1546-2226},
ABSTRACT = {Hot-section nickel-based single-crystal superalloy components under isothermal cyclic loading often exhibit systematic life shifts when datasets span different stress ratios and notch severities, making it difficult to maintain a globally consistent parameter set using conventional models. Because the effects of temperature, stress ratio, and stress concentration on cyclic response and damage evolution are typically nonlinear and coupled, this study proposes a multi-level equivalent driving force framework for fatigue life prediction, in which condition-induced life differences are represented as comparable shifts on a unified engineering driving-force scale. The proposed framework links the nominal cyclic response, the notch-root equivalent response, and crystallographic orientation weighting. Nominal responses from stress-controlled and strain-controlled tests are first unified using the cyclic Ramberg–Osgood relation, and the notch-root elastoplastic response is estimated using Neuber localization. The obtained notch-root equivalent stress response is then combined with the crystallographic Schmid factor to define an equivalent shear-type driving parameter for life correlation. Stress-ratio and notch corrections are introduced sequentially on a baseline master-curve framework to construct an equivalent driving force and identify one parameter set for each alloy–temperature combination. The validation results demonstrate that, at a fixed temperature, the proposed approach provides a unified description across multiple stress ratios and notch severities and maintains stable predictive consistency for combined conditions not included in calibration, with controlled prediction scatter.},
DOI = {10.32604/cmc.2026.087556}
}