|
|
Material Sciences 2021
面向CFRP的多关联层特征提取方法
|
Abstract:
碳纤维增强复合材料(CFRP)广泛应用于各大工业领域，其生产和制造技术已经成为我国战略新兴产业的重要支柱之一。然而，关于复合材料疲劳损伤检测的研究至今仍缺乏成熟的理论体系。因此，对于CFRP检测方法的研究受到了广泛的关注。本文针对该材料，将深度学习引入检测系统，利用自编码器提取CFRP缺陷特征，实现缺陷识别并按照位置分类的目的。实验结果表明，自编码器通过提取CFRP缺陷的数据特征不仅对缺陷有很好的辨识能力，而且可以应用于缺陷分类，效果良好。
CFRP is widely used in various industrial fields. Its production and manufacturing technology has become one of the important pillars of strategic emerging industries in China. However, the re-search on fatigue damage detection of composite materials is still lack of mature theoretical system. Therefore, the research on the CFRP detection method has been widely concerned. In this paper, deep learning is introduced into the detection system, the CFRP defect characteristics are extracted by Auto-Encoder, and the purpose of defect identification and location classification is realized. The experimental results show that the Auto-Encoder can not only identify the defects but also can be used in defect classification by extracting the data characteristics of CFRP defects.
| [1] | 杜善义, 王彪, 编著. 复合材料细观力学[M]. 北京: 科学出版社, 1998. |
| [2] | Mazumdar, S.K. (2002) Composites Manufacturing: Materials, Product, and Process Engineering. CRC Press Ltd., Boca Raton. https://doi.org/10.1201/9781420041989 |
| [3] | Lee, S. (1984) A Review of Newly Developed Composite Materials. NRCC, NAE ST 358. |
| [4] | 尹武良. 低频电磁传感检测技术: 设计、分析、计算与应用[M]. 北京: 科学出版社, 2010. |
| [5] | Sun, Y., Wang, X. and Tang, X., (2014) Deep Learning Face Representation from Predicting 10,000 Classes. 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, 23-28 June 2014, 1891-1898.
https://doi.org/10.1109/CVPR.2014.244 |
| [6] | Schmidhuber, J. (2015) Deep Learning in Neural Networks: An Overview. Neural Network, 61, 85-117.
https://doi.org/10.1016/j.neunet.2014.09.003 |
| [7] | Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S. and Darrell, T. (2014) Ca？e: Convolutional Architecture for Fast Feature Embedding. Proceedings of the 22nd ACM International Conference on Multimedia. 20 June 2014, 675-678. https://doi.org/10.1145/2647868.2654889 |
| [8] | Zhu, P., Cheng, Y., Banerjee, P., Tamburrino, A. and Deng, Y. (2019) A Novel Machine Learning Model for Eddy Current Testing with Uncertainty. NDT and E International, 101, 104-112.
https://doi.org/10.1016/j.ndteint.2018.09.010 |
| [9] | Yin, L., Ye, B., Zhang, Z., Tao, Y., Xu, H., Salas Avila, J. and Yin, W. (2019) A Novel Feature Extraction Method of Eddy Current Testing for Defect Detection Based on Machine Learning. NDT and E International, 107, Article ID: 102108. https://doi.org/10.1016/j.ndteint.2019.04.005 |
| [10] | Pasadas, D., Ramos, H., Feng, B., Baskaran, P. and Ribeiro, A. (2020). Defect Classification with SVM and Wideband Excitation in Multilayer Aluminum Plates. IEEE Transactions on Instrumentation and Measurement, 69, 241-248.
https://doi.org/10.1109/TIM.2019.2893009 |
