基于卷积神经网络的石英纤维复合材料损伤缺陷太赫兹智能识别

doi:10.3969/j.issn.1000-1565.2022.06.015

河北大学学报(自然科学版) ›› 2022, Vol. 42 ›› Issue (6): 665-672.DOI: 10.3969/j.issn.1000-1565.2022.06.015

• • 上一篇

基于卷积神经网络的石英纤维复合材料损伤缺陷太赫兹智能识别

李涛¹,薛刚²,霍自祥²,王保民¹,李晓岭²,杨召南²

收稿日期:2022-06-15 发布日期:2023-02-22
通讯作者: 霍自祥(1966—)
作者简介:李涛(1982—),男,河北满城人,邯郸学院讲师,博士,主要从事太赫兹时域光谱方向研究.
E-mail:litao9106@126.com
基金资助:
邯郸市科学技术与发展计划项目(21422031026)

Intelligent identification of damage defects in quartz fiber composites using terahertz technique based on convolutional neural network

LI Tao¹, XUE Gang², HUO Zixiang², WANG Baomin¹, LI Xiaoling², YANG Zhaonan²

1. Department of Software, Handan University, Handan 056005, China; 2. Department of Computer, Handan Vocational College of Science and Technology, Handan 056046, China

Received:2022-06-15 Published:2023-02-22

摘要/Abstract

摘要： 利用太赫兹时域光谱对石英纤维复合材料(quartz fiber reinforced polymer, QFRP)内部分层缺陷进行检测,通过搭建一维卷积神经网络模型,实现不同位置和不同深度损伤缺陷的准确识别,验证结果准确率在90%以上.根据识别结果构建复合材料的缺陷检测图像与实际太赫兹成像图结果一致,且具有高清晰度和对比度.太赫兹技术结合卷积神经网络能够实现非极性材料的智能识别.

关键词: 太赫兹时域光谱, 石英纤维复合材料, 分层缺陷, 智能识别

Abstract: In this paper, terahertz time-domain spectroscopy was used to detect the delamination defects in quartz fiber reinforced polymer(QFRP), and a one-dimensional convolutional neural network was built to realize the accurate identification of damage defects at different positions and depths, and the accuracy of the verification results was more than 90%. The defect detection image of the composite constructed according to the recognition results has high definition and contrast, which was consistent with the actual terahertz image. Terahertz technology combined with convolutional neural network can realize the intelligent recognition of non-polar materials.

Key words: terahertz time domain spectroscopy, quartz fiber composite, lamination defect, intelligent identification

中图分类号:

TP391.4
V19

李涛,薛刚,霍自祥,王保民,李晓岭,杨召南. 基于卷积神经网络的石英纤维复合材料损伤缺陷太赫兹智能识别[J]. 河北大学学报(自然科学版), 2022, 42(6): 665-672.

LI Tao, XUE Gang, HUO Zixiang, WANG Baomin, LI Xiaoling, YANG Zhaonan. Intelligent identification of damage defects in quartz fiber composites using terahertz technique based on convolutional neural network[J]. Journal of Hebei University(Natural Science Edition), 2022, 42(6): 665-672.

参考文献

[1] 谢菲,夏洪山.石英纤维复合材料作为高频透波雷达天线罩的研究[J].四川兵工学报, 2015, 36(3): 147-150. DOI: 10.11809/scbgxb2015.03.038.
[2] ZHANG J, LI W, CUI H L, et al. Nondestructive evaluation of carbon fiber reinforced polymer composites using reflective terahertz imaging[J]. Sensors, 2016, 16(6): 875. DOI: 10.3390/s16060875.
[3] 许景周,张希成.太赫兹科学技术和应用[M].北京:北京大学出版社, 2007.
[4] 梁达川,关松.太赫兹波无损检测技术及其应用[J].光电技术应用, 2018, 33(6): 1-8. DOI: 10.3969/j.issn.1673-1255.2018.06.001.
[5] ZANDONELLA C. Terahertz imaging: T-ray specs[J]. Nature, 2003, 424(6950): 721-722. DOI: 10.1038/424721a.
[6] STOIK C, BOHN M, BLACKSHIRE J. Nondestructive evaluation of aircraft composites using reflective terahertz time domain spectroscopy[J]. NDT&E INT, 2010, 43(2): 106-115. DOI: 10.1016/j.ndteint.2009.09.005.
[7] OSPALD F, ZOUAGHI W, BEIGANG R, et al. Aeronautics composite material inspection with a terahertz time-domain spectroscopy system[J]. Optical Engineering, 2013, 53(3): 031208. DOI: 10.1117/1.OE.53.3.031208.
[8] 廖晓玲.太赫兹光谱技术在复合材料无损检测中的应用研究[D].杭州:中国计量学院, 2015.
[9] 郭小弟,王强,谷小红,等.基于太赫兹时域光谱的玻璃纤维缺陷深度检测及数据分析[J].红外技术, 2016, 38(7): 602-606.
[10] 张瑾.纤维增强复合材料的太赫兹无损检测研究[D].长春:吉林大学, 2016.
[11] 邢砾云.航空泡沫芯材及夹层结构的太赫兹无损检测研究[D].长春:吉林大学, 2016.
[12] ZHANG D D, REN J J, LI L J, et al. Terahertz non-destructive testing technology for glass fiber honeycomb composites[J]. ACTA PHOTONICA SINIC, 2019, 48(2): 212002. DOI: 10.3788/gzxb20194802.0212002.
[13] IBRAHIM M E, HEADLAND D, WITHAYACHUMNANKUL W, et al. Nondestructive testing of defects in polymer-matrix composite materials for marine applications using terahertz waves[J]. J Nondestruct Eval, 2021, 40(2): 1-11. DOI: 10.1007/s10921-021-00767-9.
[14] LI T, HE J N, ZHANG L. et al. Fast quantitative analysis of hidden dangerous substances in mail based on specific interval PLS[J]. J Infrared Millim Terahertz Waves, 2021,42: 572-587. DOI: 10.1007/s10762-021-00790-x.
[15] WANG Q, LIU Q H, XIA R C, et al. Automatic defect prediction in glass fiber reinforced polymer based on THz-TDS signal analysis with neural networks[J]. Infrared Phys Techn, 2021, 115: 103673. DOI: 10.1016/j.infrared.2021.103673.
[16] 郑莉,刘闯,任姣姣,等.基于LDA-CPSO-SVM优化的多层胶接结构脱粘缺陷识别方法[J].光子学报, 2021, 50(12): 114-121. DOI: 10.3788/gzxb20215012.1212004.
[17] LI T, XU Y Y, LUO J L, et al. A method of amino acid terahertz spectrum recognition based on the convolutional neural network and bidirectional gated recurrent network model[J]. Sci Program, 2021, 2021(10): 1-7. DOI: 10.1155/2021/2097257.
[18] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6), 84-90. DOI: 10.1145/3065386.
[19] MIKOLOV T, DEORAS A, STEFAN K, et al. Empirical evaluation and combination of advanced language modeling techniques[C] //Interspeech, Conference of the International Speech Communication Association, 2012: 605-608. DOI: 10.1.1.226.3005.
[20] GUO X M, WU F, TANG X Y. Fingerprint pattern identification and classification[C] //14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery(ICNC-FSKD)2018, Huangshan, China, IEEE, 2018: 1045-1050. DOI: 10.1109/FSKD.2018.8687199.
[21] WANG Q, LIU Q H, XIA R C, et al. Defect depth determination in laser infrared thermography based on LSTM-RNN[J]. IEEE Access, 2020, 8: 153385-153393. DOI: 10.1109/ACCESS.2020.3018116.
[22] MUNIR N, KIM H J, PARK J, et al. Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions[J]. Ultrasonics,2019, 94: 74-81. DOI: 10.1016/j.ultras.2018.12.001.
[23] DEMACHI K, HORI T, PERRIN S. Crack depth estimation of non-magnetic material by convolutional neural network analysis of eddy current testing signal[J]. J Nucl Sci Technol, 2020, 57(4): 401-407. DOI: 10.1080/00223131.2019.1691076.
[24] TOWNSEND J T. Theoretical analysis of an alphabet confusion matrix[J]. Atten Percept Psychophys, 1971, 9(1): 40-50. DOI: 10.3758/BF03213026. (

基于卷积神经网络的石英纤维复合材料损伤缺陷太赫兹智能识别

Intelligent identification of damage defects in quartz fiber composites using terahertz technique based on convolutional neural network

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价