(49-6) 06 * << * >> * Russian * English * Content * All Issues

Development of neural networks for modeling diffraction of electromagnetic radiation on a single cylinder and a group of cylindrical objects
E.E. Chitorkin 1, D.L. Golovashkin 1,2

1 Samara National Research University,
Moskovskoye Shosse 34, Samara, 443086, Russia;
2 Image Processing Systems Institute, NRC “Kurchatov Institute”,
Molodogvardeyskaya Str. 151, Samara, 443001, Russia

 PDF, 1550 kB

DOI: 10.18287/2412-6179-CO-1640

Pages: 909-915.

Full text of article: Russian language.

Abstract:
We show the effectiveness of using neural networks to model the diffraction of electromagnetic radiation on cylindrical objects, and compare various architectures of neural networks. The error of the neural-network-aided diffraction modeling is evaluated for different problem statements and varying cylinder parameters. The application of the proposed models for the case of several cylinders is also analyzed.

Keywords:
neural networks, convolutional neural networks, Maxwell’s equations.

Citation:
Chitorkin EE, Golovashkin DL. Development of neural networks for modeling diffraction of electromagnetic radiation on a single cylinder and a group of cylindrical objects. Computer Optics 2025; 49(5): 909-915. DOI: 10.18287/2412-6179-CO-1640.

Acknowledgements:
This work was funded under a government project of the NRC “Kurchatov Institute”.

References:
  1. Taflove A. Computational electrodynamics: the finite-difference time-domain method. Norwood: Artech House; 2005.
  2. Moharam M, Gaylord T. Rigorous coupled-wave analysis of planar-grating diffraction. J Opt Soc Am 1981; 71(7): 811-818. DOI: 10.1364/JOSA.71.000811.
  3. Born M, Wolf E. Principles of optics. Cambridge University Press; 2019. ISBN: 978-1-108-47743-7.
  4. Jiang J, Chen M, Fan JA. Deep neural networks for the evaluation and design of photonic devices. Nat Rev Mater 2021; 6(8): 679-700. DOI: 10.1038/s41578-020-00260-1.
  5. Noakoasteen O, Wang S, Peng Z, Christodoulou C. Physics-informed deep neural networks for transient electromagnetic analysis. IEEE Open J Antennas Propag 2020; 1(1): 404-412. DOI: 10.1109/OJAP.2020.3013830.
  6. Qi S, Wang Y, Li Y, Wu X, Ren Q, Ren Y. Two-dimensional electromagnetic solver based on deep learning technique. IEEE J Multiscale Multiphysics Comput Tech 2020; 5: 83-88. DOI: 10.1109/JMMCT.2020.2995811.
  7. Kuhn L, Repan T, Rockstuhl C. Exploiting graph neural networks to perform finite-difference time-domain based optical simulations. APL Photonics 2023; 8(3): 036109. DOI: 10.1063/5.0139004.
  8. Ye Z, Cheng X, Ding X, Ou H, Peng L, Shao W. A new unconditionally stable FDTD method based on artificial neural network. IEEE Trans Microw Theory Tech 2024; 72(10): 5759-5767. DOI: 10.1109/TMTT.2024.3390174.
  9. Kim W, Seok J. Simulation acceleration for transmittance of electromagnetic waves in 2D slit arrays using deep learning. Sci Rep 2020; 10(1): 10535. DOI: 10.1038/s41598-020-67545-x.
  10. Li R, Gu X, Li K, Huang Y, Li Zh, Zhang Zh. Deep learning-based modeling of photonic crystal nanocavities. Opt Mater Express 2021; 11(7): 2122-2133. DOI: 10.1364/OME.425196.
  11. Kaiming H, Xiangyu Zh, Shaoqing R, Jian S. Deep residual learning for image recognition. 2016 IEEE Conf on Computer Vision and Pattern Recognition 2016: 770-778. DOI: 10.1109/CVPR.2016.90.
  12. Szegedy C, Ioffe S, Vanhoucke V, Alemi A. Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning. Proc AAAI Conf on Artificial Intelligence 2017; 31(1): 4278-4284. DOI: 10.1609/aaai.v31i1.11231.
  13. GOST 13659-78. Colourless optical glass. Physical and chemical properties. Basic parameters [In Russian]. Moscow: "Izdateljstvo standartov" Publisher; 1980.
  14. Hawkins S. Algorithm 1009: MieSolver – An object-oriented mie series software for wave scattering by cylinders. ACM Trans Math Softw 2020; 46(2): 19. DOI: 10.1145/3381537.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20