A Transformer-based neural network for automatic delamination characterization of quartz fiber-reinforced polymer curved structure using improved THz-TDS
Qiuhan Liu , Qiang Wang , Jiansheng Guo , Wenquan Liu , Ruicong Xia , Jiayang Yu , Xinghao Wang
{"title":"A Transformer-based neural network for automatic delamination characterization of quartz fiber-reinforced polymer curved structure using improved THz-TDS","authors":"Qiuhan Liu , Qiang Wang , Jiansheng Guo , Wenquan Liu , Ruicong Xia , Jiayang Yu , Xinghao Wang","doi":"10.1016/j.compstruct.2024.118272","DOIUrl":null,"url":null,"abstract":"<div><p>Quartz fiber-reinforced polymer (QFRP) is a vital non-polar material used in aviation wave-transparent structural components. Automatic characterization of delamination defects in QFRP is critical to aviation structural component safety. Terahertz time-domain spectroscopy (THz-TDS) is one of the new non-destructive testing (NDT) methods with highly accurate characterization of internal defects in non-polar material. Hence, attempts to extract features of THz time-domain signals for automatic characterization have been made by using deep learning algorithms. In this work, a Transformer-based neural network to classify the THz time-domain signals collected from a QFRP curved structure for automatic characterization of pre-embedded delamination defects has been reported. A THz-TDS system combined with a collaborative robot for collecting the THz signals from QFRP curved structure has been built. An automatic characterization method framework is developed. Results show that the precision rates of Transformer-based neural network for <span><math><msub><mrow><mn>1</mn></mrow><mrow><mi>s</mi><mi>t</mi></mrow></msub></math></span> delamination to <span><math><msub><mrow><mn>5</mn></mrow><mrow><mi>t</mi><mi>h</mi></mrow></msub></math></span> delamination are 1.0, 1.0, 1.0, 0.985, 1.0, and <span><math><msub><mrow><mi>F</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> score of it is 0.982. During the process of testing, delamination defects inside the QFRP curved structure were visualized using pixels with different colors. Results indicate that the Transformer-based neural network can characterize all pre-embedded delamination defects while minimizing false identification of non-defective areas, performing outstanding generalization.</p></div>","PeriodicalId":281,"journal":{"name":"Composite Structures","volume":null,"pages":null},"PeriodicalIF":6.3000,"publicationDate":"2024-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Composite Structures","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0263822324004008","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATERIALS SCIENCE, COMPOSITES","Score":null,"Total":0}
引用次数: 0
Abstract
Quartz fiber-reinforced polymer (QFRP) is a vital non-polar material used in aviation wave-transparent structural components. Automatic characterization of delamination defects in QFRP is critical to aviation structural component safety. Terahertz time-domain spectroscopy (THz-TDS) is one of the new non-destructive testing (NDT) methods with highly accurate characterization of internal defects in non-polar material. Hence, attempts to extract features of THz time-domain signals for automatic characterization have been made by using deep learning algorithms. In this work, a Transformer-based neural network to classify the THz time-domain signals collected from a QFRP curved structure for automatic characterization of pre-embedded delamination defects has been reported. A THz-TDS system combined with a collaborative robot for collecting the THz signals from QFRP curved structure has been built. An automatic characterization method framework is developed. Results show that the precision rates of Transformer-based neural network for delamination to delamination are 1.0, 1.0, 1.0, 0.985, 1.0, and score of it is 0.982. During the process of testing, delamination defects inside the QFRP curved structure were visualized using pixels with different colors. Results indicate that the Transformer-based neural network can characterize all pre-embedded delamination defects while minimizing false identification of non-defective areas, performing outstanding generalization.
期刊介绍:
The past few decades have seen outstanding advances in the use of composite materials in structural applications. There can be little doubt that, within engineering circles, composites have revolutionised traditional design concepts and made possible an unparalleled range of new and exciting possibilities as viable materials for construction. Composite Structures, an International Journal, disseminates knowledge between users, manufacturers, designers and researchers involved in structures or structural components manufactured using composite materials.
The journal publishes papers which contribute to knowledge in the use of composite materials in engineering structures. Papers deal with design, research and development studies, experimental investigations, theoretical analysis and fabrication techniques relevant to the application of composites in load-bearing components for assemblies, ranging from individual components such as plates and shells to complete composite structures.