A tensile properties-related fatigue strength predicted machine learning framework for alloys used in aerospace

• Published in: Engineering fracture mechanics 2024-05, Vol.301, p.110057, Article 110057
• Main Authors: Fan, Jiangbo, Wang, Zhangwei, Liu, Changqi, Shi, Duoqi, Yang, Xiaoguang
• Format: Article
• Language: English
• Subjects: Aerospace alloys; Fatigue strength; Machine learning framework; Tensile properties
• ISSN: 0013-7944; 1873-7315
• DOI: 10.1016/j.engfracmech.2024.110057

•Fatigue strength prediction machine learning framework related to tensile properties used in aerospace was established.•Gaussian noise was introduced in the input features to test the robustness of proposed machine learning framework.•ResNet model has better predictive accuracy compared to other machine learning models. A tensile properties-related fatigue strength prediction framework based on machine learning (ML) methods was proposed. Firstly, 200 data containing six materials used in aerospace were collected. Secondly, tensile properties-related ML framework, including residual neural network (ResNet) model, gradient boosting decision tree (GBDT) model and LightGBM (LGB) model, was established to predict fatigue strength under symmetric cycles (R = -1) after features correlation analysis. Then, Gaussian noise was introduced into the features to prove the robustness and feature sensitivity of proposed ML framework. Finally, material characteristics of fatigue resistance in the next generation were discussed after the analysis of the correlation between the tensile properties and fatigue strength. Compared with ± 50 % error band of classical prediction models, the prediction results of proposed ML framework located in the ± 10 % error band, especially for the special crystal orientations and stress ratios. In practical applications, the tensile properties-related ML framework could be applied efficiently to materials fatigue strength prediction used in aerospace due to its robust prediction accuracy and generalization ability without additional parameters fitting.