Skip to content
DraconiticResearch
Search Sources API
194,161 studies31 sourcesUpdated 26 min ago
PreprintCC BY 4.0arXiv

Reconstruction of Magnetic Fields with a Physics-Informed Neural Network

  • P. Adhikari
  • R. Bhattarai
  • J. Brewington
  • D. Buskirk
  • F. B. Hills
  • Md F. Hossain
  • R. McDonald
  • P. A. Palamure
  • A. Reid
  • S. Sharmin
  • A. Tewsley-Booth
  • B. Plaster
Open at sourcePDFDOICite

Abstract

Accurate reconstruction of magnetic fields and their gradients is essential for assessing and controlling systematic effects in precision measurement experiments. We have developed a physics-informed neural network (PINN) method which reconstructs three-dimensional magnetic fields subject to the source-free Maxwell equations, ∇·B=0 and ∇×B=0. In an extension of our previous work on PINN methods for reconstruction of magnetic fields in inaccessible interior regions via measurements at exterior locations, we now demonstrate the feasibility of applying PINN methods to the extraction of magnetic field gradient terms from discrete measurement points subject to constraints from Maxwell's equations.

Related

  1. Physics-informed, boundary-constrained Gaussian process regression for the reconstruction of fluid flow fields
    Adrian Padilla-Segarra, Pascal Noble, Olivier Roustant +1
    International Journal for Uncertainty QuantificationPeer-reviewedLicence not statedFull text2 citedStatistical & nonlinear physicsHAL
  2. Reconstruction of Molten Pool Flow Fields from High-Speed Video Using Physics-Informed Neural Networks
    Yue Cao, Qin Su, Dung Hoang Tien +1
    arXivPreprintarXiv licenceFull textFluid dynamicsarXiv
  3. Cosmo-PINN: A Physics-Informed Neural Network for Cosmological Reconstruction
    Andronikos Paliathanasis
    arXivPreprintCC BY 4.0Full textCosmologyGravitation & general relativityarXiv
  4. Reconstruction and Synthesis of Arbitrary Optical Intensity Patterns in Extremely Multimode Waveguide Cavities via Physics-Informed Optimized Modal Superposition
    Javad Babaki, Hadi Ahmadi, Sanchita Sarker +2
    Optics ExpressPeer-reviewedLicence not statedOptics & photonicsCrossref
  5. A Physics-Informed, PIV-Based Method for Surface Pressure Reconstruction
    Abbas Daliri, Fulvio Scarano
    arXivPreprintCC BY-SA 4.0Full textFluid dynamicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{adhikari2026reconstruction,
  title = {{Reconstruction of Magnetic Fields with a Physics-Informed Neural Network}},
  author = {P. Adhikari and R. Bhattarai and J. Brewington and D. Buskirk and F. B. Hills and Md F. Hossain and R. McDonald and P. A. Palamure and A. Reid and S. Sharmin and A. Tewsley-Booth and B. Plaster},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.23691},
  eprint = {2610.23691},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.23691},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Reconstruction of Magnetic Fields with a Physics-Informed Neural Network
AU  - P. Adhikari
AU  - R. Bhattarai
AU  - J. Brewington
AU  - D. Buskirk
AU  - F. B. Hills
AU  - Md F. Hossain
AU  - R. McDonald
AU  - P. A. Palamure
AU  - A. Reid
AU  - S. Sharmin
AU  - A. Tewsley-Booth
AU  - B. Plaster
PY  - 2026
DA  - 2026/10/16
PB  - arXiv
DO  - 10.48550/arxiv.2610.23691
UR  - https://arxiv.org/abs/2610.23691
AB  - Accurate reconstruction of magnetic fields and their gradients is essential for assessing and controlling systematic effects in precision measurement experiments. We have developed a physics-informed neural network (PINN) method which reconstructs three-dimensional magnetic fields subject to the source-free Maxwell equations, ∇·B=0 and ∇×B=0. In an extension of our previous work on PINN methods for reconstruction of magnetic fields in inaccessible interior regions via measurements at exterior locations, we now demonstrate the feasibility of applying PINN methods to the extraction of magnetic field gradient terms from discrete measurement points subject to constraints from Maxwell's equations.
ER  -