Skip to content
DraconiticResearch
Search Sources API
182,469 studies31 sourcesUpdated 43 min ago
Peer-reviewedCC BY 4.0The Astrophysical Journal Supplement Series

Identification of Magnetic Reconnection in Plasma Turbulence Using a Dual-branch Lightweight Hybrid Convolutional Neural Network

  • Y. Yin1
  • S. Y. Huang1
  • Q. Y. Xiong1
  • H. Zhang1
  • Z. G. Yuan1
  • K. Jiang1
  • H. H. Wu1
  • H. W. Zhan1
  • Y. T. Tang1
  • H. Bai1
  1. 1Wuhan University
Open at sourcePDFDOICite

Abstract

Magnetic reconnection is a fundamental plasma process that plays a crucial role in energy dissipation in plasma turbulence, in which magnetic energy is rapidly converted into particle energy. However, the identification of reconnection within turbulence remains a significant challenge. To automatically detect magnetic reconnection in plasma turbulence, we propose a novel and efficient classifier called a dual-branch lightweight hybrid convolutional neural network (DHCNN), which, for the first time, achieves automatic localization of reconnection sites by effectively leveraging negative-valued information in physical variables. We adopt a temporal partitioning scheme together with diverse data augmentation strategies to enhance the model’s generalization ability. Our results show that the DHCNN accurately identifies reconnection sites, achieving a Matthews correlation coefficient ( M ) of 0.67, a true skill statistic ( T ) of 0.67, a recall ( R ) of 0.83, a specificity ( S ) of 0.84, and an accuracy ( A ) of 0.84. Finally, by analyzing the impact of different physical variables, we find that the total current density (| J |), in-plane components of the magnetic field ( B x , B z ), and energy conversion term ( D e ) are the primary drivers of the model’s performance.

Related

  1. Particle Acceleration in Stratified 2D MRI Turbulence: From Reconnection to First- and Second-Order Fermi and Shear Acceleration
    Astor Sandoval, Mario Riquelme, Jorge Cuadra
    arXivPreprintCC BY 4.0Full textHigh-energy astrophysicsPlasma physicsarXiv
  2. Three-dimensional magnetic reconnection in unipolar regions. A case study of coronal bright points in the Sun
    Daniel Nóbrega-Siverio, Fernando Moreno-Insertis, Andy S. H. To
    Astronomy & AstrophysicsPeer-reviewedPublisher termsSolar & stellar physicsCrossref
  3. Current Sheets and Magnetic Reconnection statistics by Solar Orbiter at few to tens of Ion Inertial Lengths II: Conditions for Reconnection, and Current Sheet Planarity
    Inmaculada F. Albert, Sergio Toledo-Redondo, Víctor Montagud-Camps +6
    arXivPreprintarXiv licenceFull textPlanetary science & exoplanetsSolar & stellar physicsarXiv
  4. Guide field effects on particle acceleration in three-dimensional relativistic magnetic reconnection
    Despina Karavola, Lorenzo Sironi, Maria Petropoulou
    arXivPreprintarXiv licenceFull textHigh-energy astrophysicsPlasma physicsarXiv
  5. Nonuniform Particle Injection into Black Hole Jets by Radiative Magnetic Reconnection
    Rin Oikawa, Kenji Toma, Shigeo S. Kimura
    arXivPreprintCC BY 4.0Full textHigh-energy astrophysicsPlasma physicsarXiv

Cite

BibTeXRIS
BibTeX
@article{yin2026identification,
  title = {{Identification of Magnetic Reconnection in Plasma Turbulence Using a Dual-branch Lightweight Hybrid Convolutional Neural Network}},
  author = {Y. Yin and S. Y. Huang and Q. Y. Xiong and H. Zhang and Z. G. Yuan and K. Jiang and H. H. Wu and H. W. Zhan and Y. T. Tang and H. Bai},
  journal = {The Astrophysical Journal Supplement Series},
  year = {2026},
  month = oct,
  doi = {10.3847/1538-4365/aeaa32},
  url = {https://iopscience.iop.org/article/10.3847/1538-4365/aeaa32},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Identification of Magnetic Reconnection in Plasma Turbulence Using a Dual-branch Lightweight Hybrid Convolutional Neural Network
AU  - Y. Yin
AU  - S. Y. Huang
AU  - Q. Y. Xiong
AU  - H. Zhang
AU  - Z. G. Yuan
AU  - K. Jiang
AU  - H. H. Wu
AU  - H. W. Zhan
AU  - Y. T. Tang
AU  - H. Bai
PY  - 2026
DA  - 2026/10/29
JO  - The Astrophysical Journal Supplement Series
DO  - 10.3847/1538-4365/aeaa32
UR  - https://iopscience.iop.org/article/10.3847/1538-4365/aeaa32
AB  - Magnetic reconnection is a fundamental plasma process that plays a crucial role in energy dissipation in plasma turbulence, in which magnetic energy is rapidly converted into particle energy. However, the identification of reconnection within turbulence remains a significant challenge. To automatically detect magnetic reconnection in plasma turbulence, we propose a novel and efficient classifier called a dual-branch lightweight hybrid convolutional neural network (DHCNN), which, for the first time, achieves automatic localization of reconnection sites by effectively leveraging negative-valued information in physical variables. We adopt a temporal partitioning scheme together with diverse data augmentation strategies to enhance the model’s generalization ability. Our results show that the DHCNN accurately identifies reconnection sites, achieving a Matthews correlation coefficient ( M ) of 0.67, a true skill statistic ( T ) of 0.67, a recall ( R ) of 0.83, a specificity ( S ) of 0.84, and an accuracy ( A ) of 0.84. Finally, by analyzing the impact of different physical variables, we find that the total current density (| J |), in-plane components of the magnetic field ( B x , B z ), and energy conversion term ( D e ) are the primary drivers of the model’s performance.
SN  - 0067-0049
SN  - 1538-4365
ER  -