Ultrasonic signal classification and porosity testing for CFRP materials via artificial neural network

Chen, Dongkangkang; Zhou, Yufeng; Wang, Wei; Zhang, Yumin; Deng, Yansha

doi:10.1016/j.mtcomm.2021.103021

Cited by 9 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 12 illustrates the accuracy curves for both the training and validation sets. The accuracy reached its peak at around epoch 25, reaching approximately 99.8% for the training set and around 91.4% for the validation set; compared to the ANN regression model [16] and similar CNN models [23] in the same field, the accuracy of our model is competitive. This confirms that, at epoch 25, the model, indeed, achieved global optimization.…”

Section: Depth Measurement Resultsmentioning

confidence: 85%

“…In this study, the neural network preparation was conducted using TensorFlow, training was conducted with various hyperparameters, and the determination of whether the model achieved the global optimum was based on the loss function values of the training set, as illustrated in Figure 10. From the graph, it can be observed that the three batch sizes selected in this study (16,32, and 64) had minimal impact on the loss function. Even when the initial learning rates were set to 0.0001, 0.0002, and 0.0004, the loss function did not reach its minimum value even after 100 epochs, indicating that the model did not converge to the global optimum.…”

Section: Neural Network Optimizationmentioning

confidence: 91%

“…In recent years, the integration of artificial neural networks with ultrasonic detection has undergone significant advancements, which can accurately identify potential features of signals from different sources and have been widely used in the fields of defective signal identification and SNR enhancement, especially in high-attenuation, strong-interference detection scenarios, such as those involving composite materials [15][16][17]. Artificial neural networks, represented by BPNN, require the manual feature extraction of ultrasound signals, and their accuracy is affected by the type of extracted features [18][19][20].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Depth Evaluation of Tiny Defects on or near Surface Based on Convolutional Neural Network

Fei,

Cao,

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

This paper proposes a method for the detection and depth assessment of tiny defects in or near surfaces by combining laser ultrasonics with convolutional neural networks (CNNs). The innovation in this study lies in several key aspects. Firstly, a comprehensive analysis of changes in ultrasonic signal characteristics caused by variations in defect depth is conducted in both the time and frequency domains, based on discrete frequency spectra and original A—scan signals. Continuous wavelet transform (CWT) is employed to obtain wavelet time–frequency maps, demonstrating the consistent characteristics of this image with crack depth variations. A crucial innovation in this research involves the targeted design and optimization of the model based on the characteristics of ultrasonic signals and dataset size. This includes aspects such as data preparation, CNN architecture construction, and hyperparameter selection. The model is tested using a random validation set, which effectively demonstrates the CNN model’s validity and high precision. The proposed method enables the recognition and depth assessment of tiny defects on or near surfaces.

show abstract

Section: Depth Measurement Resultsmentioning

confidence: 85%

Section: Neural Network Optimizationmentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Depth Evaluation of Tiny Defects on or near Surface Based on Convolutional Neural Network

Fei,

Cao,

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Several Non-Destructive Techniques (NDT) are developed for SHM such as Acoustic Emission (AE) technique, 10 electromagnetic testing 11 and ultrasonic methods. 12 The AE technique, which is generally external, presents one of the most widely used methods that permits to control the damage evolution in materials through the detection of transient ultrasonic waves generated by damage development under load. 13,14 Many approaches are proposed in the literature.…”

Section: Introductionmentioning

confidence: 99%

In-situ health monitoring of thermoplastic bio-composites using acoustic emission

Allagui

Mahi

Rebière

et al. 2023

Journal of Thermoplastic Composite Materials

View full text Add to dashboard Cite

This article aims to accomplish research’s that has been achieved recently in the scientific community on the manufacturing monitoring and structural health monitoring (SHM) of composite materials using in-situ piezoelectric sensors. A series of specimens of flax/Elium composite laminates with and without integrated piezoelectric sensors were subject to static and fatigue tensile tests. The effects of introducing the sensor into the composite materials were studied. Moreover, the Acoustic Emission (AE) method was used during experimental tests to follow the gradual degradation and damage evolution in the bio-composites. The AE data were analyzed by the k-means unsupervised pattern recognition algorithm. Moreover, the characteristics of the acoustic signals detected from specimens with and without integrated sensors were compared. The experimental results show that the integration of piezoelectric sensors within the flax/Elium composite without degrading its mechanical properties is possible. In addition, the integrated sensor shows good results regarding class detection and separation. The in-situ health monitoring methods performed on bio-composites presents an excellent alternative to the conventional technique in which sensors are generally bonded to the material surface.

show abstract

“…In this study, ultrasonic testing (UT) testing with automated defect identification algorithm is proposed for detection of defective anchor bolts. UT is a common non-destructive (NDT) method which relies on the propagation of high frequency wave signals within the material to detect any defect or deformity [2]- [5]. This study evaluates several features extracted from ultrasonic A-scan signal obtained from defective and non-defective anchor bolts.…”

Section: Introductionmentioning

confidence: 99%

Distinguishing Defective Anchor Bolts with Features Extracted from Ultrasonic Signal

Talib¹,

Lim²,

Liew³

2022

eJNDT

View full text Add to dashboard Cite

Regular inspection on the condition of civil structure is critical to ensure structural integrity, prevent catastrophic failure and minimize downtime for emergency repair. Hitherto, the inspection on the anchor bolts relies on visual observation, which does not provide sufficient information to distinguish between normal and defective bolts. This paper aims to explore the classification of different types of defective anchor bolts based on ultrasonic signal through discrete wavelet transform (DWT). Multiple features are extracted from the ultrasonic signal to identify defective anchor bolts, thus improving the reliability of the inspection process.

show abstract

Ultrasonic signal classification and porosity testing for CFRP materials via artificial neural network

Cited by 9 publications

References 30 publications

Depth Evaluation of Tiny Defects on or near Surface Based on Convolutional Neural Network

Depth Evaluation of Tiny Defects on or near Surface Based on Convolutional Neural Network

In-situ health monitoring of thermoplastic bio-composites using acoustic emission

Distinguishing Defective Anchor Bolts with Features Extracted from Ultrasonic Signal

Contact Info

Product

Resources

About