A Physics-Informed Surrogate Model for Efficient Prediction of Electromagnetic Transmission in Nonuniform Plasmas
IEEE Transactions on Plasma Science · 2026-07 · 54 (7) · 3451-3458
DOI: 10.1109/tps.2026.3702102 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Plasma Science, 2026-07. Verfasst von Yuhang Shi und 4 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
In this article, a physics-informed surrogate model (PISM) is proposed that integrates a transfer matrix circuit topology model (TM-CTM) for the efficient prediction of electromagnetic (EM) transmission characteristics in complex plasmas. The reliance on EM software for simulated plasma is eliminated by modeling the plasma using the TM-CTM. The strength Pareto evolutionary algorithm 2 (SPEA2) is employed to rapidly determine circuit component values, thereby enhancing simulation efficiency and generating a comprehensive dataset. Subsequently, a PISM is trained to map plasma parameters directly to S-parameters. The PISM effectively replaces time-consuming full-wave simulations. Validation results demonstrate that the PISM enables the rapid prediction of $S_{21}$ amplitude and phase, serving as a robust framework for analyzing complex plasma sheaths.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Shi, Y., Wang, C., Peng, L., Jiang, X., & Mo, J. (2026). A Physics-Informed Surrogate Model for Efficient Prediction of Electromagnetic Transmission in Nonuniform Plasmas. IEEE Transactions on Plasma Science, 54(7), 3451-3458. https://doi.org/10.1109/tps.2026.3702102
Vancouver
Shi Y, Wang C, Peng L, Jiang X, Mo J. A Physics-Informed Surrogate Model for Efficient Prediction of Electromagnetic Transmission in Nonuniform Plasmas. IEEE Transactions on Plasma Science. 2026;54(7):3451-3458. doi:10.1109/tps.2026.3702102
BibTeX
@article{Shi2026,
author = {Shi, Yuhang and Wang, Chen and Peng, Lin and Jiang, Xing and Mo, Jinjun},
title = {A Physics-Informed Surrogate Model for Efficient Prediction of Electromagnetic Transmission in Nonuniform Plasmas},
journal = {IEEE Transactions on Plasma Science},
year = {2026},
volume = {54},
number = {7},
pages = {3451--3458},
doi = {10.1109/tps.2026.3702102}
}
EUVIMED