Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network
Multi-layer lightweight composite structures are widely used in the field of aviation and aerospace during the processes of manufacturing and use, and, as such, they inevitably produce defects, damage, and other quality problems, creating the need for timely non-destructive testing procedures and th...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Materials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1944/17/4/839 |
| _version_ | 1797297646409351168 |
|---|---|
| author | Dandan Zhang Lulu Li Jiyang Zhang Jiaojiao Ren Jian Gu Lijuan Li Baihong Jiang Shida Zhang |
| author_facet | Dandan Zhang Lulu Li Jiyang Zhang Jiaojiao Ren Jian Gu Lijuan Li Baihong Jiang Shida Zhang |
| author_sort | Dandan Zhang |
| collection | DOAJ |
| description | Multi-layer lightweight composite structures are widely used in the field of aviation and aerospace during the processes of manufacturing and use, and, as such, they inevitably produce defects, damage, and other quality problems, creating the need for timely non-destructive testing procedures and the convenient repair or replacement of quality problems related to the material. When using terahertz non-destructive testing technology to detect defects in multi-layer lightweight composite materials, due to the complexity of their structure and defect types, there are many signal characteristics of terahertz waves propagating in the structures, and there is no obvious rule behind them, resulting in a large gap between the recognition results and the actual ones. In this study, we introduced a U-Net-BiLSTM network that combines the strengths of the U-Net and BiLSTM networks. The U-Net network extracts the spatial features of THz signals, while the BiLSTM network captures their temporal features. By optimizing the network structure and various parameters, we obtained a model tailored to THz spectroscopy data. This model was subsequently employed for the identification and quantitative analysis of defects in multi-layer lightweight composite structures using THz non-destructive testing. The proposed U-Net-BiLSTM network achieved an accuracy of 99.45% in typical defect identification, with a comprehensive F1 score of 99.43%, outperforming the CNN, ResNet, U-Net, and BiLSTM networks. By leveraging defect classification and thickness recognition, this study successfully reconstructed three-dimensional THz defect images, thereby realizing quantitative defect detection. |
| first_indexed | 2024-03-07T22:24:19Z |
| format | Article |
| id | doaj.art-fccecf8d5ab94933838778bbc3f111d8 |
| institution | Directory Open Access Journal |
| issn | 1996-1944 |
| language | English |
| last_indexed | 2024-03-07T22:24:19Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Materials |
| spelling | doaj.art-fccecf8d5ab94933838778bbc3f111d82024-02-23T15:25:32ZengMDPI AGMaterials1996-19442024-02-0117483910.3390/ma17040839Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM NetworkDandan Zhang0Lulu Li1Jiyang Zhang2Jiaojiao Ren3Jian Gu4Lijuan Li5Baihong Jiang6Shida Zhang7Key Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaInstitute of Aerospace Special Materials and Technology, Beijing 100074, ChinaKey Laboratory of Optoelectronic Measurement and Control and Optical Information Transmission Technology, Ministry of Education, Changchun University of Science and Technology, Changchun 130022, ChinaMulti-layer lightweight composite structures are widely used in the field of aviation and aerospace during the processes of manufacturing and use, and, as such, they inevitably produce defects, damage, and other quality problems, creating the need for timely non-destructive testing procedures and the convenient repair or replacement of quality problems related to the material. When using terahertz non-destructive testing technology to detect defects in multi-layer lightweight composite materials, due to the complexity of their structure and defect types, there are many signal characteristics of terahertz waves propagating in the structures, and there is no obvious rule behind them, resulting in a large gap between the recognition results and the actual ones. In this study, we introduced a U-Net-BiLSTM network that combines the strengths of the U-Net and BiLSTM networks. The U-Net network extracts the spatial features of THz signals, while the BiLSTM network captures their temporal features. By optimizing the network structure and various parameters, we obtained a model tailored to THz spectroscopy data. This model was subsequently employed for the identification and quantitative analysis of defects in multi-layer lightweight composite structures using THz non-destructive testing. The proposed U-Net-BiLSTM network achieved an accuracy of 99.45% in typical defect identification, with a comprehensive F1 score of 99.43%, outperforming the CNN, ResNet, U-Net, and BiLSTM networks. By leveraging defect classification and thickness recognition, this study successfully reconstructed three-dimensional THz defect images, thereby realizing quantitative defect detection.https://www.mdpi.com/1996-1944/17/4/839composite structuresdefect detectionterahertz non-destructive testingU-NetBiLSTM |
| spellingShingle | Dandan Zhang Lulu Li Jiyang Zhang Jiaojiao Ren Jian Gu Lijuan Li Baihong Jiang Shida Zhang Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network Materials composite structures defect detection terahertz non-destructive testing U-Net BiLSTM |
| title | Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network |
| title_full | Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network |
| title_fullStr | Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network |
| title_full_unstemmed | Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network |
| title_short | Quantitative Detection of Defects in Multi-Layer Lightweight Composite Structures Using THz-TDS Based on a U-Net-BiLSTM Network |
| title_sort | quantitative detection of defects in multi layer lightweight composite structures using thz tds based on a u net bilstm network |
| topic | composite structures defect detection terahertz non-destructive testing U-Net BiLSTM |
| url | https://www.mdpi.com/1996-1944/17/4/839 |
| work_keys_str_mv | AT dandanzhang quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT lululi quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT jiyangzhang quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT jiaojiaoren quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT jiangu quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT lijuanli quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT baihongjiang quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork AT shidazhang quantitativedetectionofdefectsinmultilayerlightweightcompositestructuresusingthztdsbasedonaunetbilstmnetwork |