Revisiting Cylindrical Buckling under Axial Compression using Explainable Machine Learning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Revisiting Cylindrical Buckling under Axial Compression using Explainable Machine Learning

Résumé

For more than a century, the deterministic solution for the buckling of thin cylindrical shells has yielded divergent results when compared to experiments. The disparity between classical buckling theory and experimental findings has prompted an investigation into the underlying reasons for this discrepancy. Further, the classical buckling theory might not fully capture the true complexity of the relationship between material properties and geometrical parameters with the critical buckling load. To that end, this paper combines finite element simulation and explainable machine learning to uncover such a relationship. This paper employs Gaussian Process Regression (GPR) with six input variables, namely, Young's modulus, Poisson ratio, shell length, shell thickness, shell radius, and FTQC (Fabrication Tolerance Quality Class). Shapley additive explanation (SHAP) was introduced to explain the GPR model by letting the user examine every variable of the model and their interaction with another variable. The thickness of the cylinder was found to have a significant impact on the model, followed by Young's modulus, shell radius, FTQC, shell length, and Poisson ratio. Our findings reveal notable distinctions from classical theory, especially in the manner in which both the shell radius and length contribute to the critical buckling load. Nevertheless, the explanations offered by the GPR model are consistent with the classical theory concerning the trend of the influence of Young's modulus, Poisson's ratio, and thickness. The significance of FTQC also underscores the importance of factoring in manufacturing uncertainty when aiming for a specific critical buckling load in the design process.


Fichier sous embargo
Fichier sous embargo
0 2 7
Année Mois Jours
Avant la publication
lundi 8 juillet 2024
Fichier sous embargo
lundi 8 juillet 2024
Connectez-vous pour demander l'accès au fichier

Dates et versions

hal-04512431 , version 1 (12-04-2024)

Identifiants

Citer

Muhammad Wafi, Pramudita Palar, Mohammad Robani, Annisa Jusuf, Zuhal Lavi, et al.. Revisiting Cylindrical Buckling under Axial Compression using Explainable Machine Learning. Scitech, Jan 2024, Orlando, United States. ⟨10.2514/6.2024-0265⟩. ⟨hal-04512431⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More