Substantial advances in fatigue estimation of defective materials can be attained through the employment of a Physics-Informed Neural Network (PINN). The fundamental strength of such a framework is the ability to account for several defect descriptors while maintaining predictions physically sound. The first objective of the present work is the assessment of the PINN estimated fatigue life variability due to uncertainties carried by the inputs. Additionally, a set of sensitivity indices are employed to explore the influence of defect descriptors in fatigue life. The work suggested that some traditionally neglected defect descriptors may play a relevant role under specific circumstances.
Quantification of uncertainty in a defect-based Physics-Informed Neural Network for fatigue evaluation and insights on influencing factors
Tognan A.;Salvati E.
2023-01-01
Abstract
Substantial advances in fatigue estimation of defective materials can be attained through the employment of a Physics-Informed Neural Network (PINN). The fundamental strength of such a framework is the ability to account for several defect descriptors while maintaining predictions physically sound. The first objective of the present work is the assessment of the PINN estimated fatigue life variability due to uncertainties carried by the inputs. Additionally, a set of sensitivity indices are employed to explore the influence of defect descriptors in fatigue life. The work suggested that some traditionally neglected defect descriptors may play a relevant role under specific circumstances.File | Dimensione | Formato | |
---|---|---|---|
1-s2.0-S0013794423005532-main.pdf
accesso aperto
Tipologia:
Versione Editoriale (PDF)
Licenza:
Creative commons
Dimensione
763.98 kB
Formato
Adobe PDF
|
763.98 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.