EUVIMEDEUVIMED entdecken
Crossref

Microstrip patch antenna modeling using neural networks with knowledge-based regularization

Ekin Su Saçın, Ahmet Cemal Durgun

Neural Computing and Applications · 2025-02 · 37 (5) · 3827-3837

DOI: 10.1007/s00521-024-10860-5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Neural Computing and Applications, online erschienen am 16.12.2024. Verfasst von Ekin Su Saçın und 1 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: Türkiye Bilimsel ve Teknolojik Arastirma Kurumu.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neural networks (NNs) have proven a useful surrogate model for the design and optimization of high frequency structures including antennas. Black-box NNs are known to have scalability and accuracy problems as the dimension of the problem increases. This study proposes knowledge-based regularization methods, referred to as derivative, spectral, and magnitude regularization to address these issues. The proposed methods utilize the functional properties of S-parameters to improve the accuracy and prevent unphysical predictions. The NNs are trained and tested using a data set of 5000 samples generated by Latin Hypercube Sampling and simulated by Ansys HFSS. The goodness of fit is evaluated using Relative Squared Error. Derivative and spectral regularizations reduce the RSE loss from 0.052 to 0.046 and 0.043, respectively. When combined with magnitude regularization, up to 17%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} and 88%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} reduction in loss and passivity violations can be achieved, at the expense of a 37%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} increase in training time. Moreover, 25%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} less data is required, to maintain a similar loss to the reference NN.

Förderung

  • Türkiye Bilimsel ve Teknolojik Arastirma Kurumu

Zitieren

APA

Saçın, E. S., & Durgun, A. C. (2025). Microstrip patch antenna modeling using neural networks with knowledge-based regularization. Neural Computing and Applications, 37(5), 3827-3837. https://doi.org/10.1007/s00521-024-10860-5

Vancouver

Saçın ES, Durgun AC. Microstrip patch antenna modeling using neural networks with knowledge-based regularization. Neural Computing and Applications. 2025;37(5):3827-3837. doi:10.1007/s00521-024-10860-5

BibTeX
@article{San2025,
  author = {Saçın, Ekin Su and Durgun, Ahmet Cemal},
  title = {Microstrip patch antenna modeling using neural networks with knowledge-based regularization},
  journal = {Neural Computing and Applications},
  year = {2025},
  volume = {37},
  number = {5},
  pages = {3827--3837},
  doi = {10.1007/s00521-024-10860-5}
}
Kostenlos starten →