Convolutional neural network-based pattern recognition in natural circulation instability images
Reconhecimento de padrões baseado em rede neural convolucional em imagens de instabilidade de circulação natural
Palavras-chave:
Convolutional Neural Network, Two-phase flow, Pattern Recognition, Natural CirculationResumo
Heat removal systems employing natural circulation are key in new nuclear power plant designs for mitigating accidents. This study applies Convolutional Neural Networks (CNNs) to classify 'chugging' instability phases, analyzing 1152 two-phase flow images from a Natural Circulation Circuit. Three CNN models, including one incorporating transfer learning from the ImageNet database, were trained via five-fold cross-validation to fine-tune hyperparameters. This involved comparing models with and without transfer learning against a baseline linear model. A model using a pre-trained Resnet50 with transfer learning accurately classified all 230 samples, outperforming the baseline linear model with an F1-Score of 0.859. The results endorse the use of CNNs with transfer learning for thermohydraulic image analysis in identifying natural circulation instability stages.
Downloads
Referências
Boure, J. A., Bergles, A. E., and Tong, L. S.: Review of Two-phase flow instability, Nuclear Engineering and Design © North-Holland Publishing Company, 25, 165–192, https://doi.org/10.1016/0029-5493(73)90043-5,1973.
Chollet, F.: Deep Learning with Python, Second Edition, 2021.
Du, M., Yin, H., Chen, X., and Wang, X.: Oil-in-water two-phase flow pattern identification from experimental snapshots using convolutional neural network, IEEE Access, 7, https://doi.org/10.1109/ACCESS.2018.2888733, 2018.
Ghanbarzadeh, S., Hanafizadeh, P., and Hassan Saidi, M.: Intelligent image-based gas-liquid two-phase flow regime recognition, Journal of Fluids Engineering, Transactions of the ASME, 134, https://doi.org/10.1115/1.4006613, 2012.
Glorot, X. and Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks, in: Proceedings of Machine Learning Research, 9: 249-256, https://proceedings.mlr.press/v9/glorot10a.html, 2010.
Gonzalez, R. C. and Woods, R. E.: Digital Image Processing (3rd Edition), 2007.
Goodfellow, I., Bengio, Y., and Courville, A.: Deep learning An MIT Press Book, 2016.
Han, B., Ge, B., Wang, F., Gao, Q., Li, Z., and Fang, L.: Void fraction detection technology of gas-liquid two-phase bubbly flow based on convolutional neural network, Exp Therm Fluid Sci, 142, https://doi.org/10.1016/j.expthermflusci.2022.110804, 2023.
He, K., Zhang, X., Ren, S., and Sun, J.: ResNet-dev, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9908 LNCS, 2016.
Hobold, G. M. and da Silva, A. K.: Automatic detection of the onset of film boiling using convolutional neural networks and Bayesian statistics, Int J Heat Mass Transf, 134, https://doi.org/10.1016/j.ijheatmasstransfer.2018.12.070, 2019.
Huang, G., Ji, H., Huang, Z., Wang, B., and Li, H.: Flow regime identification of mini-pipe gas-liquid two-phase flow based on textural feature series, in: Conference Record - IEEE Instrumentation and Measurement Technology Conference, https://doi.org/10.1109/IMTC.2011.5944218, 2011.
International Atomic Energy Agency: Natural circulation data and methods for advanced water cooled nuclear power plant designs, VIENNA, https://www-pub.iaea.org/MTCD/Publications/PDF/te_1281_prn.pdf, 2002.
International Atomic Energy Agency: Natural circulation in water cooled nuclear power plants: phenomena models and methodology for system reliability assessments., International Atomic Energy Agency, 635 pp., https://www-pub.iaea.org/MTCD/Publications/PDF/TE_1474_web.pdf, 2005.
Jia Deng, Wei Dong, Socher, R., Li-Jia Li, Kai Li, and Li Fei-Fei: ImageNet: A large-scale hierarchical image database, https://doi.org/10.1109/cvprw.2009.5206848, 2009.
Kadish, S., Schmid, D., Son, J., and Boje, E.: Computer Vision-Based Classification of Flow Regime and Vapor Quality in Vertical Two-Phase Flow, Sensors, 22, https://doi.org/10.3390/s22030996, 2022.
Kingma, D. P. and Ba, J. L.: Adam: A method for stochastic optimization, in: 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, https://doi.org/10.48550/arXiv.1412.6980, 2015.
Krizhevsky, A., Sutskever, I., and Hinton, G. E.: ImageNet classification with deep convolutional neural networks, in: Advances in Neural Information Processing Systems, https://doi.org/10.1145/3065386, 2012.
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P.: Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86, https://doi.org/10.1109/5.726791, 1998.
LeCun, Y., Bengio, Y., and Hinton, G.: Deep learning, Nature, 436-444, https://doi.org/10.1038/nature14539, 2015.
Lin, M., Chen, Q., and Yan, S.: Network In Network, http://arxiv.org/abs/1312.4400, 2013.
Mesquita, R. N., Libardi, R. M. P., Masotti, P. H. F., Sabundjian, G., Andrade, D. A., Umbehaun, P. E., Torres, W. M., Conti, T. N., Macedo, L. A., and Engineering Conferences International.: Two-phase Flow Patterns Recognition and Parameters Estimation Though Natural Circulation Test Loop Image Analysis, in: 7th ECI International Conference on Boiling Heat Transfer 2009, 541–547, https://inis.iaea.org/collection/NCLCollectionStore/_Public/40/079/40079244.pdf, 2009.
de Mesquita, R. N., Masotti, P. H. F., Penha, R. M. L., Andrade, D. A., Sabundjian, G., Torres, W. M., and Macedo, L. A.: Classification of natural circulation two-phase flow patterns using fuzzy inference on image analysis, Nuclear Engineering and Design, 250, https://doi.org/10.1016/j.nucengdes.2012.06.014, 2012.
de Mesquita, R. N., Castro, L. F., Torres, W. M., Rocha, M. D. S., Umbehaun, P. E., Andrade, D. A., Sabundjian, G., and Masotti, P. H. F.: Classification of natural circulation two-phase flow image patterns based on self-organizing maps of full frame DCT coefficients, Nuclear Engineering and Design, 335, https://doi.org/10.1016/j.nucengdes.2018.05.019, 2018.
Murphy, K. P.: Machine Learning: A Probabilistic Perspective. Cambridge, MA: MIT Press, https://www.cs.ubc.ca/~murphyk/MLbook/pml-toc-1may12.pdf, 2012.
Nayak, A. K. and Vijayan, P. K.: Flow instabilities in boiling two-phase natural circulation systems: A review, https://doi.org/10.1155/2008/573192, 2008.
Nie, F., Wang, H., Song, Q., Zhao, Y., Shen, J., and Gong, M.: Image identification for two-phase flow patterns based on CNN algorithms, International Journal of Multiphase Flow, 152, https://doi.org/10.1016/j.ijmultiphaseflow.2022.104067, 2022.
Pang, B., Nijkamp, E., and Wu, Y. N.: Deep Learning With TensorFlow: A Review, Journal of Educational and Behavioral Statistics, 45, 227–248, https://doi.org/10.3102/1076998619872761, 2020.
Qiumei, Z., Yukun, H., Fenghua, W., Zhang, P., and Chao, L.: Liqnet: A real-time monitoring network for two-phase flow patterns, Flow Measurement and Instrumentation, 90, https://doi.org/10.1016/j.flowmeasinst.2023.102313, 2023.
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A. C., and Fei-Fei, L.: ImageNet Large Scale Visual Recognition Challenge, Int J Comput Vis, 115, https://doi.org/10.1007/s11263-015-0816-y, 2015.
Schott, S. M. C.: Classificação de Padrões de Escoamento Bifásico Por Meio de Redes Neurais Convolucionais, Master thesis, USP - University of São Paulo, São Paulo, https://doi.org/https://doi.org/10.11606/D.85.2021.tde-10062022-125100, 2021.
Schott, S. M. C., Da Silva, M. C. B., and De Mesquita, R. N.: Classification of two-phase flow instability phases using convolutional neural networks, in: 10o edition of the international Nuclear Atlantic Conference - INAC 2021, https://inac2021.aben.com.br/resumos/R0290-1.pdf, 2021.
Seal, M. K., Noori Rahim Abadi, S. M. A., Mehrabi, M., and Meyer, J. P.: Machine learning classification of in-tube condensation flow patterns using visualization, International Journal of Multiphase Flow, 143, https://doi.org/10.1016/j.ijmultiphaseflow.2021.103755, 2021.
Shanthi, C. and Pappa, N.: An artificial intelligence based improved classification of two-phase flow patterns with feature extracted from acquired images, ISA Trans, 68, https://doi.org/10.1016/j.isatra.2016.10.021, 2017.
Shi, L.: Fuzzy recognition for gas-liquid two-phase flow pattern based on image processing, in: 2007 IEEE International Conference on Control and Automation, ICCA, https://doi.org/10.1109/ICCA.2007.4376595, 2007.
Silva Junior, L. H., Barbosa, J. R., and da Silva, A. K.: Multi-parameter classification and quantification of R-134a condensation using machine learning, Appl Therm Eng, 231, 120880, https://doi.org/10.1016/j.applthermaleng.2023.120880, 2023.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R.: Dropout: A simple way to prevent neural networks from overfitting, Journal of Machine Learning Research, 15, https://jmlr.org/papers/v15/srivastava14a.html, 2014.
Theodoridis, S.: Neural Networks and Deep Learning, in: Machine Learning, 875–936, https://doi.org/10.1016/b978-0-12-801522-3.00018-5, 2015.
Weiss, K., Khoshgoftaar, T. M., and Wang, D. D.: A survey of transfer learning, J Big Data, 3, https://doi.org/10.1186/s40537-016-0043-6, 2016.
Xiao, J., Luo, X., Feng, Z., and Zhang, J.: Using artificial intelligence to improve identification of nanofluid gas-liquid two-phase flow pattern in mini-channel, AIP Adv, 8, https://doi.org/10.1063/1.5008907, 2018.
Xu, H., Tang, T., Zhang, B., and Liu, Y.: Identification of two-phase flow regime in the energy industry based on modified convolutional neural network, Progress in Nuclear Energy, 147, https://doi.org/10.1016/j.pnucene.2022.104191, 2022.
Zhang, J., Wei, X., and Wang, Z.: The Recognition Algorithm of Two-Phase Flow Patterns Based on GoogLeNet+5 Coord Attention, Micromachines (Basel), 14, https://doi.org/10.3390/mi14020462, 2023.
