Deep Learning Approach for Damage Classification Based on Acoustic Emission Data in Composite Materials

Abstract

Damage detection and the classification of carbon fiber-reinforced composites using non-destructive testing (NDT) techniques are of great importance. This paper applies an acoustic emission (AE) technique to obtain AE data from three tensile damage tests determining fiber breakage, matrix cracking, and delamination. This article proposes a deep learning approach that combines a state-of-the-art deep learning technique for time series classification: the InceptionTime model with acoustic emission data for damage classification in composite materials. Raw AE time series and frequency-domain sequence data are used as the input for the InceptionTime network, and both obtain very high classification performances, achieving high accuracy scores of about 99%. The InceptionTime network produces better training, validation, and test accuracy with the raw AE time series data than it does with the frequency-domain sequence data. Simultaneously, the InceptionTime model network shows its potential in dealing with data imbalances.

Study of the Effect of NaOH Treatment on the Properties of GF/VER Composites Using AE Technique

“…As shown in Figure 4 , the main sources of AE signals in fiber-reinforced composite materials include matrix cracking, interface debonding, fiber pullout, fiber relaxation, fiber breakage, and matrix delamination. During the entire damage process of composite materials, damage signals may be generated from a single source or multiple sources simultaneously [ 50 , 51 , 52 ].…”

Classification of early mechanical damage over time in pears based on hyperspectral imaging and transfer learning

“…Precision was used to evaluate the quality of the true positives that obtained positive predictions, and recall was used to evaluate the quality of the positive predictions. The F 1‐score is between 0 and 1, which is the summed average of precision and recall (Guo et al., 2022). Accuracy, precision, recall, F 1‐score, and specificity were calculated as follows: $$$$\begin{equation}Accuracy = \frac{{TP{\rm{ }} + {\rm{ }}TN}}{{TP + TN + FP + FN}}\end{equation}$$$$ $$$$\begin{equation}Precision = \frac{{TP}}{{TP + FP}}\end{equation}$$$$ $$$$\begin{equation}Recall = \frac{{{\rm{ }}TP}}{{TP + FN}}\end{equation}$$$$ $$$$\begin{equation}F1{\rm{ - }}score = \frac{{2 \times \left( {precision \times recall} \right){\rm{ }}}}{{precision + recall}}\end{equation}$$$$ $$$$\begin{equation}Specificity = \frac{{{\rm{ }}TN}}{{{\rm{ }}TN + FP}}\end{equation}$$$$where TP is the true positive, FP is the false positive, TN is the true negative, and FN is the false negative.…”

“…Precision was used to evaluate the quality of the true positives that obtained positive predictions, and recall was used to evaluate the quality of the positive predictions. The F1-score is between 0 and 1, which is the summed average of precision and recall (Guo et al, 2022). Accuracy, precision, recall, F1-score, and specificity were calculated as follows:…”

Acoustic-Signal-Based Damage Detection of Wind Turbine Blades—A Review

“…The deep learning model (Figure 13) has excellent automatic mining capability in image training, due to which the AE signal waveform can also be used as a training sample. Guo et al [178] proposed a deep learning method for detection of fiber breakage, matrix cracking, and delamination, namely the InceptionTime model. The model contains five inception modules to reduce overfitting of small datasets and filtering of different lengths.…”