Abstract
Acoustic emission (AE) technology is of great importance for damage detection and classification in carbon fiber reinforced composite materials. In this study, state-of-the-art deep learning (DL) models for time series were employed to classify three types of damage data, namely fiber breakage, matrix cracking, and delamination, obtained from tensile damage tests. The raw AE time series data were used as inputs to compare the classification performance of eight different deep learning models including FCN, ResNet, XResNet, LSTM_FCN, InceptionTime, XceptionTime, mWDN, and LSTM. The evaluation and analysis showed that XceptionTime, InceptionTime, and ResNet models achieved better training, validation, and testing accuracy, enabling accurate classification of composite material damage. Notably, these three models also demonstrated ideal classification performance for imbalanced data in the composite material AE dataset, providing reference methods for imbalanced data classification problems with small sample sizes in practical engineering applications.
REFERENCES
Chawla, K.K., Composite Materials, Cham: Springer, 2019, pp. 297–311.
Gao, Y., Xiao, D., He, T., Lin, Y., Li, N., Ye, Q., and Wang, Y., Identification of damage mechanisms of carbon fiber reinforced silicon carbide composites under static loading using acoustic emission monitoring, Ceram. Int., 2019, vol. 45, pp. 13847–13858. https://doi.org/10.1016/j.ceramint.2019.04.082
Pinho, S.T., Robinson, P., and Iannucci, L., Fracture toughness of the tensile and compressive fibre failure modes in laminated composites, Compos. Sci. Technol., 2006, vol. 66, pp. 2069–2079. https://doi.org/10.1016/j.compscitech.2005.12.023
Padmaraj, N.H., Pai, D.K., Shreepannaga, S., and Kini, M.V., Fatigue behaviour and damage characterization of quasi-isotropic carbon/epoxy laminates, Cogent Eng., 2022, vol. 9, no. 1, p. 2077680. https://doi.org/:10.1080/23311916.2022.2077680
Turon, A., Camanho, P.P., Costa, J., and Davila, C.G., A damage model for the simulation of delamination in advanced composites under variable-mode loading, Mech. Mater., 2006, vol. 38, pp. 1072–1089. https://doi.org/10.1016/j.mechmat.2005.10.003
Chelliah, S.K., Kannivel, S.K., and Vellayaraj, A., Characterization of failure mechanism in glass, carbon and their hybrid composite laminates in epoxy resin by acoustic emission monitoring, Nondestr. Test. Eval., 2019, vol. 34, pp. 254–266. https://doi.org/10.1080/10589759.2019.1590829
Gul, S., Tabrizi, I.E., Okan, B.S., Kefal, A., and Yildiz, M., An experimental investigation on damage mechanisms of thick hybrid composite structures under flexural loading using multi-instrument measurements, Aerospace Sci. Technol., 2021, vol. 117, p. 106921. https://doi.org/10.1016/j.ast.2021.106921
Jinachandran, S. and Rajan, G., Fibre Bragg grating based acoustic emission measurement system for structural health monitoring applications, Materials, 2021, vol. 14. https://doi.org/10.3390/ma14040897
Behnia, A., Ranjbar, N., Chai, H.K., and Masaeli, M., Failure prediction and reliability analysis of ferrocement composite structures by incorporating machine learning into acoustic emission monitoring technique, Constr. Build. Mater., 2016, vol. 122, pp. 823–832. https://doi.org/10.1016/j.conbuildmat.2016.06.130
Lissek, F., Haeger, A., Knoblauch, V., Hloch, S., Pude, F., and Kaufeld, M., Acoustic emission for interlaminar toughness testing of CFRP: Evaluation of the crack growth due to burst analysis, Compos. B Eng., 2018, vol. 136, pp. 55–62. https://doi.org/10.1016/j.compositesb.2017.10.012
Liu, H. and Zhang, Y., Image-driven structural steel damage condition assessment method using deep learning algorithm, Measurement, 2019, vol. 133, pp. 168–181. https://doi.org/10.1016/j.measurement.2018.09.081
Krizhevskii, A., Sutskever, I., and Geoffrey E. Hinton, Imagenet classification with deep convolutional neural networks, Commun. ACM, 2012, vol. 60, no. 6, pp. 84–90. https://doi.org/10.1145/3065386
Ebrahimkhanlou, A., Schneider, M.B., Dubuc, B., and Salamone, S., A deep learning framework for acoustic emission sources localization and characterization in complex aerospace, Mater. Eval., 2021, vol. 79, pp. 391–400. https://doi.org/10.32548/2021.me-04179
Sathiyamurthy, R., Duraiselvam, M., and Sevvel, P., Acoustic emission based deep learning technique to predict adhesive bond strength of laser processed CFRP composites, FME Trans., 2020, vol. 48, pp. 611–619. https://doi.org/10.5937/fme2003611S
Haile, M.A., Zhu, E., Hsu, C., and Bradley, N., Deep machine learning for detection of acoustic wave reflections, Struct. Health Monit., 2020, vol. 19, pp. 1340–1350. https://doi.org/10.1177/1475921719881642
Sikdar, S., Liu, D., and Kundu, A., Acoustic emission data based deep learning approach for classification and detection of damage-sources in a composite panel, Compos. B Eng., 2022, vol. 228. https://doi.org/10.1016/j.compositesb.2021.109450
Daugela, A., Chang, C.H., and Peterson, D.W., Deep learning based characterization of nanoindentation induced acoustic events, Mater. Sci. Eng. A, 2021, vol. 800, p. 140273. https://doi.org/10.1016/j.msea.2020.140273
Ebrahim, S.A., Poshtan, J., Jamali, S.M., and Ebrahim, N.A., Quantitative and qualitative analysis of time-series classification using deep learning, IEEE Access, 2020, vol. 8, pp. 90202–90215. https://doi.org/10.1109/ACCESS.2020.2993538
Fawaz, H.I., Forestier, G., Weber, J., Idoumghar, L., and Muller, P., Deep learning for time series classification: a review, Data Min. Knowl. Discovery, 2019, vol. 33, pp. 917–963. https://doi.org/10.1007/s10618-019-00619-1
Guo, F., Li, W., Jiang, P., Chen, F., and Liu, Y., Deep learning approach for damage classification based on acoustic emission data in composite materials, Materials, 2022, vol. 15. https://doi.org/10.3390/ma15124270
Tu, N.D.K., Noh, M., Ko, Y., Kim, J., Kang, C.Y., and Kim, H., Enhanced electromechanical performance of P(VDF-TrFE-CTFE) thin films hybridized with highly dispersed carbon blacks, Compos. B Eng., 2018, vol. 152, pp. 133–138. https://doi.org/10.1016/j.compositesb.2018.06.036
Zeng, J., Gao, W., and Liu, F., Interfacial behavior and debonding failures of full-scale CFRP-strengthened H-section steel beams, Compos. Struct., 2018, vol. 201, pp. 540–552. https://doi.org/10.1016/j.compstruct.2018.06.045
Pennecchi, F.R., Kuselman, I., Di Rocco, A., Hibbert, D.B., Sobina, A., and Sobina, E., Specific risks of false decisions in conformity assessment of a substance or material with a mass balance constraint—A case study of potassium iodate, Measurement, 2021, vol. 173. https://doi.org/10.1016/j.measurement.2020.108662
Liu, W. and Chen, P., Theoretical analysis and experimental investigation of the occurrence of fiber bridging in unidirectional laminates under Mode I loading, Compos. Struct., 2021, vol. 257. https://doi.org/10.1016/j.compstruct.2020.113383
Kundu, A., Sikdar, S., Eaton, M., and Navaratne, R., A generic framework for application of machine learning in acoustic emission-based damage identification, 13th Int. Conf. Damage Assess. Struct. (Porto, 2020), pp. 244–262.
Li, H., Zhang, K., Cheng, H., Suo, H., Cheng, Y., and Hu, J., Multi-stage mechanical behavior and failure mechanism analysis of CFRP/Al single-lap bolted joints with different seawater ageing conditions, Compos. Struct., 2019, vol. 208, pp. 634–645. https://doi.org/10.1016/j.compstruct.2018.10.044
Aljazaeri, Z.R., Janke, M.A., and Myers, J.J., A novel and effective anchorage system for enhancing the flexural capacity of RC beams strengthened with FRCM composites, Compos. Struct., 2019, vol. 210, pp. 20–28. https://doi.org/10.1016/j.compstruct.2018.10.110
Jierula, A., Wang, S., Oh, T., Lee, J., and Lee, J.H., Detection of source locations in RC columns using machine learning with acoustic emission data, Eng. Struct., 2021, vol. 246, p. 112992. https://doi.org/10.1016/j.engstruct.2021.112992
Muir, C., Swaminathan, B., Almansour, A.S., Sevener, K., Smith, C., Presby, M., Kiser, J.D., Pollock, T.M., and Daly, S., Damage mechanism identification in composites via machine learning and acoustic emission, NPJ Comput. Mater., 2021, vol. 7. https://doi.org/10.1038/s41524-021-00565-x
Nelon, C., Myers, O., and Hall, A., The intersection of damage evaluation of fiber-reinforced composite materials with machine learning: A review, J. Compos. Mater., 2022, vol. 56, pp. 1417–1452. https://doi.org/10.1177/00219983211037048
Wang, X.C.M.C., Impact damage detection of CFRP laminates using a convolutional neural network based on wavelet packet decomposition, Compos. Struct., 2021, p. 114665. https://doi.org/10.1016/j.compstruct.2021.114665
Lissner, M., Erice, B., Alabort, E., Thomson, D., Cui, H., Kaboglu, C., Blackman, B.R.K., Gude, M., and Petrinic, N., Multi-material adhesively bonded structures: Characterisation and modelling of their rate-dependent performance, Compos. B Eng., 2020, vol. 195. https://doi.org/10.1016/j.compositesb.2020.108077
Fang, C.H.Y.S., Structural health monitoring of composite laminates using an acoustic emission-based recurrent neural network, J. Intel. Mat. Syst. Struct., 2021, pp. 3–16. https://doi.org/10.1177/1045389X20922419
Han, J.Z.B.L., A convolutional neural network for classification of acoustic emission signals in composite delamination detection, Compos. Sci. Technol., 2021, p. 108795. https://doi.org/10.1016/j.compscitech.2020.108795
Kumar, V., Yokozeki, T., Karch, C., Hassen, A.A., Hershey, C.J., Kim, S., Lindahl, J.M., Barnes, A., Bandari, Y.K., and Kunc, V., Factors affecting direct lightning strike damage to fiber reinforced composites: A review, Compos. B Eng., 2020, vol. 183. https://doi.org/10.1016/j.compositesb.2019.107688
Xu, J.H.Q.L., A novel framework for identifying damage mechanisms in CFRP composites using deep learning techniques, Compos. B Eng., 2020, p. 107710. https://doi.org/10.1016/j.compositesb.2019.107710
Arribasplata-Seguin, A., Quispe-Dominguez, R., Tupia-Anticona, W., and Acosta-Sullcahuaman, J., Rotational molding parameters of wood-plastic composite materials made of recycled high density polyethylene and wood particles, Compos. B Eng., 2021, vol. 217. https://doi.org/10.1016/j.compositesb.2021.108876
Tinkloh, S., Wu, T., Troester, T., and Niendorf, T., A micromechanical-based finite element simulation of process-induced residual stresses in metal-CFRP-hybrid structures, Compos. Struct., 2020, vol. 238. https://doi.org/10.1016/j.compstruct.2020.111926
Wang, Y., Chi, Z., and Liu, J., On buckling behaviors of a typical bending-dominated periodic lattice, Compos. Struct., 2021, vol. 258. https://doi.org/10.1016/j.compstruct.2020.113204
Giordano, A., Mao, L., and Chiang, F., Full-field experimental analysis of a sandwich beam under bending and comparison with theories, Compos. Struct., 2021, vol. 255. https://doi.org/10.1016/j.compstruct.2020.112965
Hamel, C.M., Kuang, X., and Qi, H.J., Modeling the dissolution of thermosetting polymers and composites via solvent assisted exchange reactions, Compos. B Eng., 2020, vol. 200. https://doi.org/10.1016/j.compositesb.2020.108363
Huang, H.Z.X.S., A hybrid method of 1D-CNN and BiLSTM for acoustic emission signal classification and damage diagnosis of CFRP composites, Compos. Struct., 2021, p. 112836. https://doi.org/10.1016/j.compstruct.2020.112836
Machine Learning, in: Encyclopedia of Animal Cognition and Behavior, Vonk, J., Shackelford, T.K., Eds., Cham: Springer, 2022, p. 4038.
Chollet, F., Deep Learning with Python, Shelter Island: Manning Publ. Co., 2018.
Krizhevsky, A., Sutskever, I., and Hinton, G.E., ImageNet classification with deep convolutional neural networks, Commun. ACM, 2017, vol. 60, pp. 84–90. https://doi.org/10.1145/3065386
Tsai, L., Fast AI Deep Learning from the Foundations. https://www.fast.ai/. Accessed March 29, 2023.
Grandini, M., Bagli, E., and Visani, G., Metrics for multi-class classification: An overview, arXiv, 2020, vol. 17.
Funding
This work was supported by ongoing institutional funding. No additional grants to carry out or direct this particular research were obtained.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors declare that they have no conflicts of interest.
Rights and permissions
About this article
Cite this article
Guo, F., Li, W., Jiang, P. et al. Deep Learning for Time Series-Based Acoustic Emission Damage Classification in Composite Materials. Russ J Nondestruct Test 59, 665–676 (2023). https://doi.org/10.1134/S1061830923600314
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1061830923600314