We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Analysis of High-Cycle Fatigue Life Prediction of 304 Stainless Steel Based on Deep Learning.
- Authors
Duan, Hongyan; He, Hong; Yue, Shunqiang; Cao, Mengjie; Zhao, Yingjian; Zhang, Zengwang; Liu, Yang
- Abstract
In this century, deep learning has been widely used due to the rapid popularization of the Internet and the improvement of computer performance. This dataset has been established by collecting the high-cycle fatigue test data of 304 stainless steel. The datasets were preprocessed, then inputted into the back propagation neural network model, a fuzzy neural network model, and a long short-term memory neural network (LSTM) model for training and testing. The reliability and generalization of the three models have been verified by high-cycle fatigue experiments, and the prediction effects of the three models compared. The results show that the LSTM model in the deep learning model has better prediction accuracy for high-cycle fatigue life, is superior to the other two machine learning models in terms of generalization and accuracy, and the correlation coefficient (R2) of the final prediction result was 0.9786.
- Subjects
DEEP learning; FATIGUE life; MACHINE learning; STAINLESS steel; FUZZY neural networks; BACK propagation
- Publication
JOM: The Journal of The Minerals, Metals & Materials Society (TMS), 2023, Vol 75, Issue 11, p4586
- ISSN
1047-4838
- Publication type
Article
- DOI
10.1007/s11837-023-06042-8