期刊
COMPOSITES PART B-ENGINEERING
卷 257, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.compositesb.2023.110694
关键词
Terahertz wave; Non-destructive evaluation; Convolutional neural network; Micro-defect; Probability map
This study presents a non-destructive evaluation method using a CNN-based THz signal processing to distinguish overlapping signals from micro-defects in GFRP composites. Theoretical models were devised to describe the propagation of the THz signal and its interaction with the micro-defects. THz signals were transformed into 2D spectrograms and used to train a CNN deep learning model for classifying the size and thickness of the defects. The CNN model achieved a 95% accuracy in differentiating micro-defects thinner than 20 μm.
This study presents a non-destructive evaluation (NDE) method that utilizes a convolutional neural network (CNN) based terahertz (THz) signal processing to quantitively distinguish overlapping signals from micro-defects in glass fiber reinforced polymer (GFRP) composites. Theoretical electro-magnetic wave propagation models for describing the propagation of the THz signal into the GFRP composite were devised considering the interaction of THz with the micro-defects in the GFRP composite. THz signals were simulated with respect to the thickness of the defects, and transformed into two-dimensional (2D) spectrograms using a short-term Fourier transform (STFT). The converted 2D-STFT images were used to train the CNN deep learning model to classify the size and thickness of the defects. A probability map of the classification of the defect thickness was derived from the trained CNN deep learning model. Using the probability map based on the CNN model, the depth and size of the micro-defects thinner than 20 & mu;m inside the GFRP composites could be successfully differentiated with 95% accuracy. This research proposes a promising approach for analyzing micro-defects in composite materials by quantitatively evaluating the extent of signal overlap, going beyond the traditional approach of training the model based solely on the signal.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据