FATIGUE LIFE PREDICTION IN 7075 T651 ALUMINUM ALLOY: EFFECTS OF APPLIED STRESS
Abstract
This study examines the relationship between applied stress (σa [MPa]) and fatigue life (Nf) in 7075-T651 aluminum alloy using both physics-based and machine learning approaches. Fatigue tests were conducted under bending and torsional loading at multiple time stages, and the fatigue life was modeled on a log10(Nf) scale. A Basquin-type regression (log10(Nf) ~ stress amplitude) captured the expected monotonic stress–life trend, explaining approximately one-third of the variance (R² ≈ 0.33) with an RMSE of 0.71 log cycles (≈3.2 × 10⁶ cycles) and an MAE of 0.58 log cycles (≈1.1 × 10⁶ cycles). An optimized XGBoost model trained with applied stress and time stage reduced the absolute error (MAE ≈ 7.6 × 10⁵ cycles) but accounted for less variance overall (R² ≈ 0.17; RMSE ≈ 0.84 log cycles, ≈2.4 × 10⁶ cycles). Feature importance analysis confirmed that applied stress was the dominant predictor, with time stage contributing only marginally. These findings show that while XGBoost achieved modest improvements in raw error, the Basquin regression remained more interpretable and physically consistent, underscoring the importance of benchmarking ML models against established fatigue mechanics.
Related
- Pore suppression in adjustable ring-mode laser lap welding of dissimilar-thickness 1060 aluminum alloy
- Influence of temperature and strain rate on the competition between dynamic precipitation and dynamic dissolution in an aged 7xxx Aluminum alloy
- Ultrarobust, Rapidly Recoverable Superhydrophobic Coatings with Self-Similar Micro/Nanostructures and Reinforced Interface for Aluminum Alloy Protection
- Automated Burr Quality Classification in Drilling of AA6013 Aluminum Alloy Using YOLOv10n
- Nanoscale perspective on the stress-corrosion cracking behavior of a peak-aged 7XXX-Al alloy
BibTeX
@article{iwuji2026fatigue,
title = {{FATIGUE LIFE PREDICTION IN 7075 T651 ALUMINUM ALLOY: EFFECTS OF APPLIED STRESS}},
author = {Kingsley Obinna Iwuji},
journal = {GLOBAL JOURNAL OF ENGINEERING AND TECHNOLOGY ADVANCES},
year = {2026},
month = sep,
doi = {10.30574/gjeta.2026.28.3.0243},
url = {https://hal.science/hal-05847467v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - FATIGUE LIFE PREDICTION IN 7075 T651 ALUMINUM ALLOY: EFFECTS OF APPLIED STRESS
AU - Kingsley Obinna Iwuji
PY - 2026
DA - 2026/09/30
JO - GLOBAL JOURNAL OF ENGINEERING AND TECHNOLOGY ADVANCES
DO - 10.30574/gjeta.2026.28.3.0243
UR - https://hal.science/hal-05847467v1
AB - This study examines the relationship between applied stress (σa [MPa]) and fatigue life (Nf) in 7075-T651 aluminum alloy using both physics-based and machine learning approaches. Fatigue tests were conducted under bending and torsional loading at multiple time stages, and the fatigue life was modeled on a log10(Nf) scale. A Basquin-type regression (log10(Nf) ~ stress amplitude) captured the expected monotonic stress–life trend, explaining approximately one-third of the variance (R² ≈ 0.33) with an RMSE of 0.71 log cycles (≈3.2 × 10⁶ cycles) and an MAE of 0.58 log cycles (≈1.1 × 10⁶ cycles). An optimized XGBoost model trained with applied stress and time stage reduced the absolute error (MAE ≈ 7.6 × 10⁵ cycles) but accounted for less variance overall (R² ≈ 0.17; RMSE ≈ 0.84 log cycles, ≈2.4 × 10⁶ cycles). Feature importance analysis confirmed that applied stress was the dominant predictor, with time stage contributing only marginally. These findings show that while XGBoost achieved modest improvements in raw error, the Basquin regression remained more interpretable and physically consistent, underscoring the importance of benchmarking ML models against established fatigue mechanics.
ER -