TY - JOUR
T1 - A generalized machine learning framework to estimate fatigue life across materials with minimal data
AU - Srinivasan, Dharun Vadugappatty
AU - Moradi, Morteza
AU - Komninos, Panagiotis
AU - Zarouchas, Dimitrios
AU - Vassilopoulos, Anastasios P.
PY - 2024
Y1 - 2024
N2 - In this research, a generalized machine learning (ML) framework is proposed to estimate the fatigue life of epoxy polymers and additively manufactured AlSi10Mg alloy materials, leveraging their failure surface void characteristics. An extreme gradient boosting algorithm-based ML framework encompassing Synthetic Minority Over-sampling TEchnique (SMOTE), categorical data encoding, and external loop cross-validation is developed to evaluate the fatigue life across materials. The influence of different training strategies based on materials, input features, encoding method, and data standardization on the model performance is explored. Additionally, the importance of anti-data-leakage and anti-overfitting measures over the ML model performance is addressed. The result shows that the data-leakage-free, external loop cross-validated model can estimate the fatigue life of selective epoxy polymers and metal alloys with an average R2 of 0.71 ± 0.06 using a mere 12 to 27 experimental data points per material category. Whereas the model trained with data-leakage and overfitting results in high R2 of 0.9.
AB - In this research, a generalized machine learning (ML) framework is proposed to estimate the fatigue life of epoxy polymers and additively manufactured AlSi10Mg alloy materials, leveraging their failure surface void characteristics. An extreme gradient boosting algorithm-based ML framework encompassing Synthetic Minority Over-sampling TEchnique (SMOTE), categorical data encoding, and external loop cross-validation is developed to evaluate the fatigue life across materials. The influence of different training strategies based on materials, input features, encoding method, and data standardization on the model performance is explored. Additionally, the importance of anti-data-leakage and anti-overfitting measures over the ML model performance is addressed. The result shows that the data-leakage-free, external loop cross-validated model can estimate the fatigue life of selective epoxy polymers and metal alloys with an average R2 of 0.71 ± 0.06 using a mere 12 to 27 experimental data points per material category. Whereas the model trained with data-leakage and overfitting results in high R2 of 0.9.
KW - Composites
KW - Fatigue
KW - Machine learning
KW - Metal alloys
KW - Minimal data
KW - Void
UR - http://www.scopus.com/inward/record.url?scp=85205685671&partnerID=8YFLogxK
U2 - 10.1016/j.matdes.2024.113355
DO - 10.1016/j.matdes.2024.113355
M3 - Article
AN - SCOPUS:85205685671
SN - 0264-1275
VL - 246
JO - Materials and Design
JF - Materials and Design
M1 - 113355
ER -