A Physics‐Informed Neural Network for Predicting Two‐Dimensional Species Concentrations in a Magnetic Field‐Stabilized Atmospheric Pressure Glow Discharge
Plasma Processes and Polymers · 2026-07 · 23 (7)
DOI: 10.1002/ppap.70223 ↗Kurz gefasst
Fachartikel in der Zeitschrift Plasma Processes and Polymers, online erschienen am 20.07.2026. Verfasst von Xueqing Zhao und 7 weiteren Autor:innen. Die Arbeit nennt 59 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Atmospheric‐pressure non‐equilibrium plasma offers a promising pathway for sustainable nitrogen fixation. In this work, we employ a physics‐informed neural network model (PINN) to predict the two‐dimensional spatial distributions of 29 species in a magnetic field‐stabilized atmospheric pressure glow discharge system. The model combines ZDPlasKin simulations with experimentally measured 2D laser‐induced fluorescence data for N, O, and NO, while enforcing particle kinetics, charge conservation, current continuity, and mass conservation. The PINN model eliminates mesh dependence, reduces computational cost, and incorporates complex plasma chemistry. The results reveal detailed distributions of key nitrogen and oxygen species and their dominant reaction pathways. Model predictions agree well with theoretical estimates and experiments, demonstrating the potential of PINN for plasma nitrogen fixation diagnostics and optimization.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhao, X., Teng, X., Wu, E., Li, Z., Liu, J., Nie, L., Mesbah, A., & Lu, X. (2026). A Physics‐Informed Neural Network for Predicting Two‐Dimensional Species Concentrations in a Magnetic Field‐Stabilized Atmospheric Pressure Glow Discharge. Plasma Processes and Polymers, 23(7). https://doi.org/10.1002/ppap.70223
Vancouver
Zhao X, Teng X, Wu E, Li Z, Liu J, Nie L, et al. A Physics‐Informed Neural Network for Predicting Two‐Dimensional Species Concentrations in a Magnetic Field‐Stabilized Atmospheric Pressure Glow Discharge. Plasma Processes and Polymers. 2026;23(7). doi:10.1002/ppap.70223
BibTeX
@article{Zhao2026,
author = {Zhao, Xueqing and Teng, Xiaohan and Wu, Erqi and Li, Zhiyu and Liu, Jialin and Nie, Lanlan and Mesbah, Ali and Lu, Xinpei},
title = {A Physics‐Informed Neural Network for Predicting Two‐Dimensional Species Concentrations in a Magnetic Field‐Stabilized Atmospheric Pressure Glow Discharge},
journal = {Plasma Processes and Polymers},
year = {2026},
volume = {23},
number = {7},
doi = {10.1002/ppap.70223}
}
EUVIMED