7050 aluminum alloy, vital in aerospace, is susceptible to high-temperature creep damage. Accurate creep life prediction is crucial, but traditional methods and machine learning face challenges: limited data and, for GANs, pattern collapse.
Methods
To address the above issues, researchers from East China University of Science and Technology proposed a creep fracture life prediction method named AP-GAN-DNN, which combines the affinity propagation (AP) clustering algorithm, GAN, and deep neural networks (DNN) to predict creep fracture life more accurately with small samples.
The AP clustering algorithm trains independent GAN models for each cluster and generates synthetic data that is highly similar to the real data. Finally, the DNN model is trained using the synthetic data and predicted using real creep fracture life data.
Highlights
The proposed AP-GAN-DNN method is able to predict creep fracture life more accurately in the case of small samples, which solves the complex nonlinear problem and the problem of test data dispersion.
Compared with traditional physical methods and machine learning models, the AP-GAN-DNN model has better generalization and robustness for creep life prediction.
AP-GAN data enhancement method can significantly reduce the "pattern collapse" effect of GAN and improve the performance of prediction algorithms.
The proposed AP-GAN-DNN method provides a new approach to address the issue of creep fracture life prediction using machine learning algorithms with small samples.
Prof. Fu-Zhen Xuan is currently the President of East China University of Science and Technology. He is a recipient of the National Science Fund for Distinguished Young Scholars and the Chang Jiang Scholars Program. He has also been awarded the first and second prizes of the National Science and Technology Progress Award. His research interests cover structural failure analysis, life prediction, and intelligent detection.
Welcome to the International Center for Creep Prediction ICCP! We share the latest research and insights on high-temperature materials, strength, Creep, and Fundamental Theory and experiments of Materials. We welcome you to share your research for free! Join us in advancing the field!
