DOI | Trouver le DOI : https://doi.org/10.1109/TIM.2023.3300463 |
---|
Auteur | Rechercher : Yun, HongguangIdentifiant ORCID : https://orcid.org/0000-0003-4552-2913; Rechercher : Feng, KeIdentifiant ORCID : https://orcid.org/0000-0003-2338-5161; Rechercher : Rayhana, RakibaIdentifiant ORCID : https://orcid.org/0000-0002-1512-4335; Rechercher : Pant, Shashank1Identifiant ORCID : https://orcid.org/0000-0003-3271-5011; Rechercher : Genest, Marc1Identifiant ORCID : https://orcid.org/0000-0001-6301-6235; Rechercher : Liu, ZhengIdentifiant ORCID : https://orcid.org/0000-0002-7241-3483 |
---|
Affiliation | - Conseil national de recherches du Canada. Aérospatiale
|
---|
Bailleur de fonds | Rechercher : National Research Council of Canada |
---|
Format | Texte, Article |
---|
Sujet | composite laminate; damage localization; data fusion; guided wave; neural networks; ultrasonics; location awareness; transducers; feature extraction; data models; artificial neural networks; signal processing; acoustics |
---|
Résumé | Ultrasonic guided wave (UGW)-based damage localization on plate-like composite structures plays a vital role in the structural health monitoring (SHM) of aircraft structures. Precisely locating the damage requires full utilization of high-dimensional UGW signals as well as low-dimensional transducer coordinates. However, current deep learning (DL)-based methods cannot incorporate transducer coordinates in the neural networks. To address this issue, this article proposes a novel multidimensional data fusion neural network framework for damage localization on plate-like composite structures using UGW. The proposed framework includes an encoder and a Fourier feature projection head to integrate high-dimensional wave signals and low-dimensional coordinates. A multilayer perceptron (MLP) is adopted as a decoder to learn features from the encoder and the projection head. Comprehensive experiments demonstrate that the proposed method achieves the state-of-the-art results with less than 2 mm absolute distance error. Moreover, a discussion regarding data availability in the training process is performed. The proposed method demonstrates superior robustness over the state-of-the-art methods with limited training data. |
---|
Date de publication | 2023-08-01 |
---|
Maison d’édition | IEEE |
---|
Dans | |
---|
Langue | anglais |
---|
Publications évaluées par des pairs | Oui |
---|
Exporter la notice | Exporter en format RIS |
---|
Signaler une correction | Signaler une correction (s'ouvre dans un nouvel onglet) |
---|
Identificateur de l’enregistrement | 803fe418-6e53-4ab4-b360-918db732418a |
---|
Enregistrement créé | 2024-06-27 |
---|
Enregistrement modifié | 2024-06-27 |
---|