Acoustic Emissions Detection and Ranging of Cracks in Metal Tanks Using Deep Learning

Cardarilli, G.C.; Nunzio, Luca Di; Fazzolari, Rocco; Giardino, Daniele; Matta, Marco; Re, M.; Spanò, Sergio

doi:10.1007/978-3-030-37277-4_37

Cited by 2 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, AE technology and deep learning are linked and have been adopted by many researchers [7,[35][36][37][38][39][40][41]. The conversion of time series data into two-dimensional image data using short fast Fourier transform, wavelet transform, and the classification of acoustic emission data [42][43][44][45][46] using two-dimensional convolutional neural networks (CNNs) is a common method that has been used by many researchers.…”

Section: Introductionmentioning

confidence: 99%

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

Fu-ping

Jiang

et al. 2022

Materials

View full text Add to dashboard Cite

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.

show abstract

Section: Introductionmentioning

confidence: 99%

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

Fu-ping

Jiang

et al. 2022

Materials

View full text Add to dashboard Cite

show abstract

“…Detection of cracks in shafts represents one of the most reliable fault diagnosis methods, as it helps in preventing catastrophic failures from taking place [1,2]. Thus, many techniques have been developed to predict the occurrence of cracks in shafts; for instance, Shen and Taylor [3] introduced a method for simple crack identification in a beam with one pair of symmetric cracks.…”

Section: Introductionmentioning

confidence: 99%

Estimation the Natural Frequencies of a Cracked Shaft Based on Finite Element Modeling and Artificial Neural Network

Flaieh

Hamdoon

Jaber

2020

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

The early detection of faults in rotating systems considers an integral approach that has received considerable attention from the industrial sector, as it contributes to preventing catastrophic failures in machines. In this research, the natural frequencies of a shaft, when it is healthy and when cracks with different depths are introduced, have been calculated. The deviation of the computed natural frequencies from the healthy ones is counted as a sign of the presence of an abnormality in the system. For this intention, the finite element analysis (FEA) method based on ANSYS software has been utilized to obtain the first five natural frequencies of the shaft when there is a crack of different severity at different positions. The results of the FEA are used for designing an artificial neural network (ANN) model that can be easily used to predict the first five natural frequencies of the shaft based on just the crack's position and depth. Finally, the predicted natural frequencies by the deigned ANN have been compared to their peers that were computed using the FEA method. The absolute error percentage has then been calculated and used to get an indication of how close the result of both techniques is. The recorded highest error percentage was 0.67%, which is quite small and referring to that the designed ANN can accurately predict the natural frequencies of rotating systems.

show abstract

Acoustic Emissions Detection and Ranging of Cracks in Metal Tanks Using Deep Learning

Cited by 2 publications

References 15 publications

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

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

Estimation the Natural Frequencies of a Cracked Shaft Based on Finite Element Modeling and Artificial Neural Network

Contact Info

Product

Resources

About