Damage Quantification Method for Aircraft Structures Based on Multitask CNN-LSTM and Transfer Learning
IEEE Sensors Journal · 15.03.2024 · 24 (6) · 9217-9228
DOI: 10.1109/jsen.2024.3360109 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 15.03.2024. Verfasst von Weihan Shao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 37 Literaturangaben. Gefördert von: Natural Science Foundation of China, Chinese Aeronautical Establishment, Natural Science Foundation of Fujian Province.
Abstract (Originalsprache)
Damage quantification based on Lamb waves is one of the research hotspots in the field of aerospace structural health monitoring (SHM). Deep learning (DL) is an efficient method to identify damage-related features from Lamb waves’ complex responses. In this article, a multitask convolutional neural networks and long-term and short-term memory networks (CNNs-LSTM) damage quantification method combining transfer learning is proposed, which directly uses the Lamb waves signal in the original discrete-time domain to predict the size and location of damage. The 1-D convolutional neural network (1D-CNN) is used to achieve damage size prediction, which can not only learn the corresponding features but also avoid wasting training resources. For damage location prediction, a multitask CNN-LSTM network architecture is established. Two parallel branches can output the coordinates of damage in ${x}$ - and ${y}$ -directions at the same time, to locate the damage at any location within the structure. To prove the reliability and generalization ability of the method, three datasets are collected through experiments. The three datasets are derived from two aluminum plates and one composite laminate. The model trained on the first aluminum plate is defined as the pre-training model, its structure and weight are extracted, and then the transfer learning method is used to realize the structural damage location identification of aluminum plate-aluminum plate and aluminum plate-composite laminate, which is of certain value for the research and application of transfer learning theory in damage quantification.
Förderung
- Natural Science Foundation of China
- Chinese Aeronautical Establishment
- Natural Science Foundation of Fujian Province
- Innovation Foundation for Young Scholar of Xiamen
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Shao, W., Sun, H., Wang, Y., & Qing, X. (2024). Damage Quantification Method for Aircraft Structures Based on Multitask CNN-LSTM and Transfer Learning. IEEE Sensors Journal, 24(6), 9217-9228. https://doi.org/10.1109/jsen.2024.3360109
Vancouver
Shao W, Sun H, Wang Y, Qing X. Damage Quantification Method for Aircraft Structures Based on Multitask CNN-LSTM and Transfer Learning. IEEE Sensors Journal. 2024;24(6):9217-9228. doi:10.1109/jsen.2024.3360109
BibTeX
@article{Shao2024,
author = {Shao, Weihan and Sun, Hu and Wang, Yishou and Qing, Xinlin},
title = {Damage Quantification Method for Aircraft Structures Based on Multitask CNN-LSTM and Transfer Learning},
journal = {IEEE Sensors Journal},
year = {2024},
volume = {24},
number = {6},
pages = {9217--9228},
doi = {10.1109/jsen.2024.3360109}
}
EUVIMED