Explainable Machine Learning for Bubble Leakage Detection at Tube Array Surfaces in Pool
Source abstract
Abstract
Early detection of bubble generation from tube arrays in systems such as fast reactor steam generators, Pressurized Water Reactor (PWR) cores, and Liquefied Natural Gas (LNG) regasification units is critical for safety. While various methods have been proposed, they face challenges such as high spatial resolution requirements, rapid response times, and varying strengths and weaknesses, suggesting the need for a combined approach. This study integrates ultrasonic testing (UT) with Machine Learning (ML) to identify the presence, location, and direction of bubbles within a complex tube array that cause signal attenuation. A Convolutional Neural Network (CNN) successfully achieved 100% identification accuracy. Furthermore, a method was developed that uses an autoencoder as a feature extractor, combined with a One-Class Support Vector Machine (SVM) and k-means. This approach achieved high accuracy and a correct decision basis. It also demonstrated strong generalization, successfully detecting anomalies without requiring labels for anomalous data, enabling robust bubble identification.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Fluid Dynamics and Mixing · Ultrasound and Cavitation Phenomena · Water Systems and Optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Cooperative Major in Nuclear Energy, Graduate School of Advanced Science and Engineering, Waseda University, Tokyo 169-8555, Japan
References (53)
Show source-provided references
1. State Sensing of Bubble Jet Flow Based on Acoustic Recognition and Deep Learning10.1016/j.ijmultiphaseflow.2022.104340 ↗
2. When Bubbles Are Not Spherical: Artificial Intelligence Analysis of Ultrasonic Cavitation Bubbles in Solutions of Varying Concentrations10.1021/acs.jpcb.2c00948 ↗
3. Sun, S., Xu, F., Cai, L., Salvato, D., Dilemma, F., Capriotti, L., Xian, M., and Yao, T. (2023). An Efficient Instance Segmentation Approach for Studying Fission Gas Bubbles in Irradiated Metallic Nuclear Fuel. Sci. Rep., 13.10.1038/s41598-023-47914-y ↗
4. Liu, Z., Niu, X., Guo, X., Mu, G., and Pang, Y. (2006, January 13–16). Method for Acoustic Leak Detection of Fast Reactor Steam Generator Based on Wavelet Noise Elimination. Proceedings of the 2006 International Conference on Machine Learning and Cybernetics, Dalian, China.10.1109/ICMLC.2006.258523 ↗
5. Water Leak Detection in Steam Generator of Super Phenix10.1016/0149-1970(88)90073-X ↗
6. Demonstration of an Improved Passive Acoustic Fault Detection Method on Recordings from the Phénix Steam Generator Operating at Full Power10.1016/j.anucene.2016.10.003 ↗
7. Acoustic Leak Detection System for Sodium-Cooled Reactor Steam Generators Using Delay-and-Sum Beamformer10.1080/18811248.2010.9711932 ↗
8. Fundamental Evaluation of Hydrogen Behavior in Sodium for Sodium-Water Reaction Detection of Sodium-Cooled Fast Reactor10.1016/j.net.2023.10.043 ↗
9. Liquid Sodium—The Heat Transport Medium in Fast Breeder Reactors10.1007/BF02757649 ↗
10. Mingyu, H., Tao, W., Jinbing, C., and Xiaoran, W. (2020, January 22–24). Study on Ultrasonic Location Based on Sound Pressure and TDOA Switching. Proceedings of the 2020 Chinese Control and Decision Conference (CCDC), Hefei, China.10.1109/CCDC49329.2020.9164437 ↗
11. Griffin, J.W., Peters, T.J., Posakony, G.J., Chien, H.-T., Bond, L.J., Denslow, K.M., Sheen, S.-H., and Raptis, P. (2009). Under-Sodium Viewing: A Review of Ultrasonic Imaging Technology for Liquid Metal Fast Reactors.10.2172/1010482 ↗
12. Modelling of Ultrasonic Propagation in Turbulent Liquid Sodium with Temperature Gradient10.1063/1.4875876 ↗
13. Review on Research Progress in Boiling Acoustics10.1016/j.icheatmasstransfer.2022.106465 ↗
14. Investigation of Spatial Effects Accompanying Local Boiling of Coolant in VVER Core on the Basis of Neutron Flux Noise Analysis
15. A Review of Research on Acoustic Detection of Heat Exchanger Tube
16. Three-Dimensional Transcranial Microbubble Cavitation Localization by Four Sensors10.1109/TUFFC.2021.3091950 ↗
17. Leak Detection of Steam or Water into Sodium in Steam Generators of Liquid-Metal Fast Breeder Reactors
18. Desai, P.D., Ng, W.C., Hines, M.J., Riaz, Y., Tesar, V., and Zimmerman, W.B. (2019). Comparison of Bubble Size Distributions Inferred from Acoustic, Optical Visualisation, and Laser Diffraction. Colloids Interfaces, 3.10.3390/colloids3040065 ↗
19. High Frequency Ultrasonic Imaging10.1016/S0929-6441(09)60012-6 ↗
20. Computer Methods for Non-Invasive Measurement and Control of Two-Phase Flows: A Review Study10.5755/j01.itc.48.3.22189 ↗
21. Evaluating the Attenuation in Ultrasonic Testing of Castings
22. Classification of Ultrasonic Signals10.1504/IJMPT.2006.011267 ↗
23. Guichou, R., Tordjeman, P., Bergez, W., and Zamansky, R. (2017, January 21–22). Experimental Study of Bubble Detection in Liquid Metal. Proceedings of the VIII International Scientific Colloquium “Modelling for Materials Processing”, Riga, Latvia.10.22364/mmp2017.30 ↗
24. Sun, H., Ramuhalli, P., and Meyer, R. (2023). An Assessment of Machine Learning Applied to Ultrasonic Nondestructive Evaluation.10.2172/2320378 ↗
25. Zhang, X., Yu, Y., Yu, Z., Qiao, F., Du, J., and Yao, H. (2025). A Scoping Review: Applications of Deep Learning in Non-Destructive Building Tests. Electronics, 14.10.3390/electronics14061124 ↗
26. Matsuura, M., and Ikeda, M. (2005, January 7–12). Modification of Steam Generator System to Prevent Overheating Tube Rapture Accidents at MONJU. Proceedings of the 18th International Conference on Structural Mechanics in Reactor Technology, Beijing, China.
27. Chetal, S.C. (1997). Evolution of Design of Steam Generator for Sodium Cooled Reactors, International Atomic Energy Agency.
28. Haynes, W.M. (2016). CRC Handbook of Chemistry and Physics, CRC Press.10.1201/9781315380476 ↗
29. Klasifikasi Cacat Biji Kopi Menggunakan Metode Transfer Learning dengan Hyperparameter Tuning Gridsearch
30. Tan, M., and Le, Q. (2021, January 18–24). EfficientNetV2: Smaller Models and Faster Training. Proceedings of the 38th International Conference on Machine Learning Research, Online.
31. Chang, D.T. (2019). Bayesian Hyperparameter Optimization with BoTorch, GPyTorch, and Ax. arXiv.
32. Nonlinear Mode Decomposition and Reduced-Order Modeling for Three-Dimensional Cylinder Flow by Distributed Learning on Fugaku
33. Supervised Learning Mixing Characteristics of Film Cooling in a Rocket Combustor Using Convolutional Neural Networks10.1016/j.actaastro.2020.05.021 ↗
34. Hyperparameter Optimization in Generative Adversarial Networks (GANs) Using Gaussian AHP
35. Combining Autoencoder with Clustering Analysis for Anomaly Detection in Radiotherapy Plans10.21037/qims-22-825 ↗
36. Aktar, S., and Nur, A.Y. (2024, January 11). Advancing Network Anomaly Detection: An Ensemble Approach Combining Optimized Contractive Autoencoders and k-Means Clustering. Proceedings of the 2024 IEEE 3rd International Conference on Computing and Machine Intelligence (ICMI), Mt Pleasant, MI, USA.10.1109/ICMI60790.2024.10585975 ↗
37. Dumont, V., Ju, X., and Mueller, J. (2022). Hyperparameter Optimization of Generative Adversarial Network Models for High-Energy Physics Simulations. arXiv.10.21203/rs.3.rs-2181360/v1 ↗
38. Best Selection of Generative Adversarial Networks Hyper-Parameters Using Genetic Algorithm10.1007/s42979-021-00689-3 ↗
39. Ribeiro, M.T., Singh, S., and Guestrin, C. (2016, January 13–17). “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.10.1145/2939672.2939778 ↗
40. Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22–29). Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.10.1109/ICCV.2017.74 ↗
41. Visualizing Data Using T-SNE
42. van der Accelerating T-SNE Using Tree-Based Algorithms
43. Laguna, S., Heidenreich, J.N., Sun, J., Cetin, N., Al-Hazwani, I., Schlegel, U., Cheng, F., and El-Assady, M. (2023, January 22). ExpLIMEable: A Visual Analytics Approach for Exploring LIME. Proceedings of the 2023 Workshop on Visual Analytics in Healthcare (VAHC), Melbourne, Australia.10.1109/VAHC60858.2023.00011 ↗
44. Anwar, S., Griffiths, N., Bhalerao, A., and Popham, T.J. (2024, January 26). MASALA: Model-Agnostic Surrogate Explanations by Locality Adaptation. Proceedings of the 1st KDD Workshop on Human-Interpretable AI, Barcelona, Spain.
45. Application of Autoencoders Neural Network and K-Means Clustering for the Definition of Geostatistical Estimation Domains|IIETA
46. Nguyen, V., Viet Hung, N., Le-Khac, N.-A., and Cao, V.L. (2020). Clustering-Based Deep Autoencoders for Network Anomaly Detection, Springer International Publishing.10.1007/978-3-030-63924-2_17 ↗
47. Akbarian, H., Mahgoub, I., and Williams, A. (2024, January 28–30). Autoencoder-K-Means Algorithm for Efficient Anomaly Detection to Improve Space Operations. Proceedings of the 2024 International Conference on Smart Applications, Communications and Networking (SmartNets), Harrisonburg, VA, USA.10.1109/SmartNets61466.2024.10577704 ↗
48. Chang, L.-K., Wang, S.-H., and Tsai, M.-C. (2020). Demagnetization Fault Diagnosis of a PMSM Using Auto-Encoder and K-Means Clustering. Energies, 13.10.3390/en13174467 ↗
49. Unsupervised Deep Embedding for Clustering Analysis
50. Aytekin, C., Ni, X., Cricri, F., and Aksu, E. (2018, January 8–13). Clustering and Unsupervised Anomaly Detection with L2 Normalized Deep Auto-Encoder Representations. Proceedings of the 2018 International Joint Conference on Neural Networks (IJCNN), Rio de Janeiro, Brazil.10.1109/IJCNN.2018.8489068 ↗
51. Islam, M.M., Faruque, M.O., Butterfield, J., Singh, G., and Cooke, T.A. (2023, January 16–20). Unsupervised Clustering of Disturbances in Power Systems via Deep Convolutional Autoencoders. Proceedings of the 2023 IEEE Power & Energy Society General Meeting (PESGM), Orlando, FL, USA.10.1109/PESGM52003.2023.10253229 ↗
52. Estimating the Support of a High-Dimensional Distribution10.1162/089976601750264965 ↗
53. Support vector data description10.1023/B:MACH.0000008084.60811.49 ↗