Automatic analysis and anomaly detection system of transverse electron beam profile based on advanced and interpretable deep learning architectures

• Autorzy: Piekarski Michał , Jaworek-Korjakowska Joanna , Wawrzyniak Adriana

Identyfikatory

DOI

10.2478/jaiscr-2024-0008

• Języki publikacji: EN

Abstrakty

EN

The National Synchrotron Radiation Center SOLARIS, ranked among the top infrastructures of that type worldwide, is the only one located in Central-Eastern Europe, in Poland. The SOLARIS Center, with six fully operational beamlines, serves as a hub for research across a diverse range of disciplines. This cutting-edge facility fosters innovation in fields like biology, chemistry, and physics as well as material engineering, nanotechnology, medicine, and pharmacology. With its advanced infrastructure and multidisciplinary approach, the SOLARIS Center enables discoveries and pushes the boundaries of knowledge. The most important aspect of such enormous research as well as industry infrastructure is to provide stable working conditions for the users and the conducted projects. Due to its unique properties, problem complexities, and challenges that require advanced approaches, the problem of anomaly detection and automatic analysis of signals for the beam stability assessment is still a huge challenge that has not been fully developed. To increase the effectiveness of centers with advanced research infrastructure we focus on the automatic analysis of transverse beam profiles generated by the Pinhole diagnostic beamline. Pinhole beamlines are typically installed in the middle and high-energy synchrotrons to thoroughly analyze emitted X-rays and therefore assess electron beam quality. To address the problem we take advantage of AI solutions including up-to-date pre-trained convolutional neural network (CNN) models among others EfficientNetB0-B4-B6, InceptionV3 and DenseNet121. In this research, we propose the benchmark for Pinhole image classification including data preprocessing, model implementation, training process, hyperparameter selection as well as testing phase. Created from scratch database contains over one million transverse beam profile images. The proposed solution, based on the InceptionV3 architecture, classifies pinhole beamline images with 94.1% accuracy and 96.6% precision which is a state-of-the-art result in this research area. Finally, we employed interpretability algorithms to perform an analysis of the models and achieved results.

Słowa kluczowe

EN

signal analysis; anomaly detection; transfer learning; convolutional neural networks; Synchrotron Radiation System

• Wydawca: University of Social Sciences
• Czasopismo: Journal of Artificial Intelligence and Soft Computing Research
• Rocznik: 2024
• Tom: Vol. 14, No. 2
• Strony: 139--156
• Opis fizyczny: Bibliogr. 32 poz., rys.

Twórcy

autor

Piekarski Michał

• AGH University in Krakow, Department of Automatic Control and Robotics, Al. Adama Mickiewicza 30, 30-059, Krakow, Poland
• National Synchrotron Radiation Center SOLARIS, Jagiellonian University Czerwone Maki 98, 30-392, Krakow, Poland

• https://orcid.org/0000-0001-9391-4263

autor

Jaworek-Korjakowska Joanna

• AGH University in Krakow, Department of Automatic Control and Robotics, Center of Excellence in Artificial Intelligence Al. Adama Mickiewicza 30, 30-059, Krakow, Poland

• https://orcid.org/0000-0003-0146-8652

autor

Wawrzyniak Adriana

• National Synchrotron Radiation Center SOLARIS, Jagiellonian University Czerwone Maki 98, 30-392, Krakow, Poland

• https://orcid.org/0000-0003-1807-3817

Bibliografia

• [1] M. Piekarski, J. Jaworek-Korjakowska, A. I. Wawrzyniak, and M. Gorgon Convolutional neural network architecture for beam instabilities identification in synchrotron radiation systems as an anomaly detection problem, Measurement, vol. 165, p. 108116, 2020. [Online]. Available: https://www.sciencedirect.com/science/rticle/pii/S0263224120306540
• [2] M. Braei and S. Wagner, Anomaly detection in univariate time-series: A survey on the state-of-the-art, 2020. [Online]. Available: https://arxiv.org/abs/2004.00433
• [3] A. Wawrzyniak, A. Marendziak, A. Kisiel, P. Borowiec, R. Nietubyc, J. Wiechecki, K. Karaś, K. Szamota-Leandersson, M. Zajac, C. Bocchetta, and M. Stankiewicz, Solaris a new class of low energy and high brightness light source, Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, 01 2017.
• [4] S. Hulbert and G. Williams, 1 - synchrotron radiation sources, in Vacuum Ultraviolet Spectroscopy, J. Samson and D. Ederer, Eds. Burlington: Academic Press, 2000, pp. 1–25. [Online]. Available: https://www.sciencedirect.com/science/article/pii/B9780126175608500025
• [5] R. Panaś, A. Curcio, and A. Wawrzyniak, Lumos: A visible diagnostic beamlinefor the solaris storage ring, 05 2021, pp. 2901–3903.
• [6] G. Pang, C. Shen, L. Cao, and A. V. D. Hengel, Deep learning for anomaly detection, ACM Computing Surveys, vol. 54, no. 2, pp. 1–38, mar 2021. [Online]. Available: https://doi.org/10.11452F3439950
• [7] M. Segovia and J. Garcia-Alfaro, Design, modeling and implementation of digital twins, Sensors, vol. 22, p. 5396, 07 2022.
• [8] D. Kafkes and M. Schram, Developing robust digital twins and reinforcement learning for accelerator control systems at the fermilab booster, 2021. [Online]. Available: https://arxiv.org/abs/2105.12847
• [9] S. Pioli, B. Buonomo, F. Cardelli, P. Ciuffetti, L. Foggetta, and C. Di Giulio, Machine learning based middle-layer for autonomous accelerator operation and control, 2021.
• [10] L. Vera Ramirez, T. Birke, G. Hartmann, R. Müller, ¨M. Ries, A. Schälicke, and P. Schnizer, Machine Learning Tools Improve BESSY II Operation, JACoW, vol. ICALEPCS2021, p. THAL01, 2022.
• [11] R. Pirayesh, S. Biedron, J. Diaz Cruz, M. MartinezRamon, and S. Sosa Guitron, Achieving Optimal Control of LLRF Control System with Artificial Intelligence, in 17th International Conference on Accelerator and Large Experimental Physics Control Systems, 2020, p. MOPHA114.
• [12] T. Dewitte, W. Meert, E. Van Wolputte, and P. Van Trappen, Anomaly detection for cern beam transfer installations using machine learning, in Proceedings of the 17th International Conference on Accelerator and Large Experimental Control Systems (ICALEPCS 2019). JACoW, 2019, pp. 1066–1070.
• [13] J. Edelen, K. Brown, K. Bruhwiler, E. Carlin, C. Hall, and V. Schoefer, Machine Learning Based Tuning and Diagnostics for the ATR Line at BNL, JACoW, vol. ICALEPCS2021, p. THAL04, 2022.
• [14] A. Pollard, D. Dunning, and M. Maheshwari, Learning to lase: Machine learning prediction of fel beam properties, 2021.
• [15] K. Agari, H. Akiyama, Y. Morino, Y. Sato, and A. Toyoda, An Application of Machine Learning for the Analysis of Temperature Rise on the Production Target in Hadron Experimental Facility at J-PARC, in 17th International Conference on Accelerator and Large Experimental Physics Control Systems, 2020, p. WEMPL001.
• [16] S. Takano, Beam diagnostics with synchrotron radiation in light sources, 2010.
• [17] Y.-B. Leng, G.-Q. Huang, M.-Z. Zhang, Z.-C. Chen, J. Chen, and K.-R. Ye, The beam-based calibration of an x-ray pinhole camera at SSRF, Chinese Physics C, vol. 36, no. 1, pp. 80–83, jan 2012. [Online]. Available: https://doi.org/10.1088%2F1674-1137%2F36%2F1%2F014
• [18] A. Kisiel, A. Marendziak, M. Ptaszkiewicz, and A. I. Wawrzyniak, X-ray pinhole camera for emittance measurements in solaris storage ring, 2019.
• [19] K. Weiss, T. Khoshgoftaar, and D. Wang, A survey of transfer learning, Journal of Big Data, vol. 3, 05 2016.
• [20] R. Mormont, P. Geurts, and R. Maree, Comparison of deep transfer learning strategies for digital pathology, in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2018.
• [21] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, Rethinking the inception architecture for computer vision, 2015. [Online]. Available: https://arxiv.org/abs/1512.00567[22] G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, Densely connected convolutional networks, 2016. [Online]. Available: https://arxiv.org/abs/1608.06993
• [23] M. Tan and Q. V. Le, Efficientnet: Rethinking model scaling for convolutional neural networks, 2019. [Online]. Available: https://arxiv.org/abs/1905.11946
• [24] F. Chollet, Deep learning with python, 2017.
• [25] R. Y. Rubinstein and D. P. Kroese, The crossentropy method, in Information Science and Statistics, 2004.
• [26] C. M. Bishop and N. M. Nasrabadi, Pattern recognition and machine learning, J. Electronic Imaging, vol. 16, p. 049901, 2007.
• [27] D. P. Kingma and J. Ba, Adam: A method forstochastic optimization, 2014. [Online]. Available: https://arxiv.org/abs/1412.6980
• [28] S. Chakraborty, R. Tomsett, R. Raghavendra,D. Harborne, M. Alzantot, F. Cerutti, M. Srivastava, A. Preece, S. Julier, R. M. Rao, T. D. Kelley, D. Braines, M. Sensoy, C. J. Willis, and P. Gurram,Interpretability of deep learning models: A surveyof results, in 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internetof People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 2017, pp. 1–6.
• [29] R. R. Selvaraju and et al., Grad-cam: Why did you say that? visual explanations from deep networks via gradient-based localization, 2016.
• [30] M. T. Ribeiro and et al., Why should i trust you?: Explaining the predictions of any classifier, ACM KDD: Knowledge Discovery and Data Mining,2016.
• [31] L. van der Maaten and G. Hinton, Viualizing data using t-sne, Journal of Machine Learning Research, vol. 9, pp. 2579–2605, 11 2008.
• [32] L. McInnes, J. Healy, and J. Melville, Umap: Uniform manifold approximation and projection fordimension reduction, 2018. [Online]. Available:https://arxiv.org/abs/1802.03426

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).