Result Details

Machine Learning for Antenna Design: Combining CST Studio Suite and Python

BEDNARSKÝ, V.; RAIDA, Z. Machine Learning for Antenna Design: Combining CST Studio Suite and Python. In PROCEEDINGS I OF THE 29TH STUDENT EEICT 2023. 1. Brno: BRNO UNIVERSITY OF TECHNOLOGY, FACULTY OF ELECTRICAL ENGINEERING AND COMMUNICATION, 2023. p. 352-356. ISBN: 978-80-214-6153-6.
Type
conference paper
Language
English
Authors
Bednarský Vojtěch, Ing.
Raida Zbyněk, prof. Dr. Ing., UREL (FEEC)
Abstract

The design and optimization of antennas is a
complex and time-consuming process which combines an
electromagnetic analysis to evaluate cost functions and a machine
learning to consequently improve designs. In this paper, CST
Studio Suite performs the numerical analysis, and Python scripts
implement other steps. Python executes numerical operations,
automatically generates models, and supports the CST analyses
without requiring user’s interaction. Ultimately, the approach is
aimed to utilize Python’s libraries PyTochr and TensorFlow to
automate antenna designs, which can be leveraged by artificial
intelligence, at a later stage.

Keywords

CST Studio Suite, Python, PyTochr, TensorFlow,
particle swarm optimization (PSO), canonical antenna

URL
Published
2023
Pages
352–356
Proceedings
PROCEEDINGS I OF THE 29TH STUDENT EEICT 2023
Series
1
Edition
1
Conference
STUDENT EEICT 2023
ISBN
978-80-214-6153-6
Publisher
BRNO UNIVERSITY OF TECHNOLOGY, FACULTY OF ELECTRICAL ENGINEERING AND COMMUNICATION
Place
Brno
BibTeX
@inproceedings{BUT188126,
  author="Vojtěch {Bednarský} and Zbyněk {Raida}",
  title="Machine Learning for Antenna Design: Combining CST Studio Suite and Python",
  booktitle="PROCEEDINGS I OF THE 29TH STUDENT EEICT 2023",
  year="2023",
  series="1",
  number="1",
  pages="352--356",
  publisher="BRNO UNIVERSITY OF TECHNOLOGY, FACULTY OF ELECTRICAL ENGINEERING AND COMMUNICATION",
  address="Brno",
  isbn="978-80-214-6153-6",
  url="https://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_1.pdf"
}
Departments
Back to top