Research on Ultrasonic Guided Wave Structure Damage Localization Method Using 2-D Image Fusion and Dual-Channel Heterogeneous Neural Networks
IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 3515811-3515811
DOI: 10.1109/tim.2026.3699627 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Yong Chang und 4 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Fund of Henan Key Laboratory of Superhard Abrasives and Grinding Equipment, Henan University of Technology.
Abstract (Originalsprache)
To achieve accurate localization of structural damage, a novel ultrasonic guided wave (UGW) detection method based on 2-D image fusion technology and a dual-channel heterogeneous neural network is introduced in this study. First, the method innovatively combines the Gramian angular field (GAF) and Markov transition field (MTF) techniques to convert 1-D guided wave time-series signals into 2-D images with complementary features. In this way, the spatiotemporal texture information and dynamic statistical features of guided wave detection signals are better preserved, providing high-information-content data samples for a deep learning (DL) model. Second, based on the different image features of GAF and MTF, a dual-channel heterogeneous neural network architecture is constructed to achieve efficient recognition of the spatiotemporal and statistical information features of the images. Finally, a convolutional block attention module is introduced, which applies adaptive attention weighting to the dual-channel features to achieve deep fusion and enhance the damage characteristics. The results of the experiments on aluminum plate damage detection demonstrate that the proposed method enables high-precision localized structural damage detection, providing an effective tool for UGW detection.
Förderung
- National Natural Science Foundation of China
- Fund of Henan Key Laboratory of Superhard Abrasives and Grinding Equipment, Henan University of Technology
Zitieren
APA
Chang, Y., Liu, H., Li, N., Du, Y., & Liu, Z. (2026). Research on Ultrasonic Guided Wave Structure Damage Localization Method Using 2-D Image Fusion and Dual-Channel Heterogeneous Neural Networks. IEEE Transactions on Instrumentation and Measurement, 75, 3515811-3515811. https://doi.org/10.1109/tim.2026.3699627
Vancouver
Chang Y, Liu H, Li N, Du Y, Liu Z. Research on Ultrasonic Guided Wave Structure Damage Localization Method Using 2-D Image Fusion and Dual-Channel Heterogeneous Neural Networks. IEEE Transactions on Instrumentation and Measurement. 2026;75:3515811-3515811. doi:10.1109/tim.2026.3699627
BibTeX
@article{Chang2026,
author = {Chang, Yong and Liu, Haoyuan and Li, Nana and Du, Yihang and Liu, Zongyao},
title = {Research on Ultrasonic Guided Wave Structure Damage Localization Method Using 2-D Image Fusion and Dual-Channel Heterogeneous Neural Networks},
journal = {IEEE Transactions on Instrumentation and Measurement},
year = {2026},
volume = {75},
pages = {3515811--3515811},
doi = {10.1109/tim.2026.3699627}
}
EUVIMED