Crack Growth Rate Model Derived from Domain Knowledge-Guided Symbolic Regression

Abstract Machine learning (ML) has powerful nonlinear processing and multivariate capabilities, so it been widely utilised in the fatigue field. However, most ML methods are inexplicable black-box models that difficult to apply engineering practice. Symbolic regression (SR) is an interpretable machine method for determining optimal fitting equation datasets. In this study, domain knowledge-guided SR was used determine a new crack growth (FCG) rate model. Three terms of variable subtree Δ K , R -ratio, th were obtained by analysing eight traditional semi-empirical FCG models. Based on test data from other literature, model constructed using Al-7055-T7511. It subsequently extended alloys (Ti-10V-2Fe-3Al, Ti-6Al-4V, Cr-Mo-V, LC9cs, Al-6013-T651, Al-2324-T3) multiple linear regression. Compared with three models, yielded higher prediction accuracy. This result demonstrates potential building