A deep learning model with dynamic masks and multi-parameter constraints predicts solar vector magnetic fields over 12 hours, achieving SSIM 0.912 and CC 0.998 for the radial component with 7.82% unsigned flux error.
Prediction of the next solar rotation synoptic maps using an artificial intelligence–based surface flux transport model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
astro-ph.SR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Deep Learning with Magnetic Parameter Constraints for Short-Term Prediction of Solar Active Region Vector Magnetic Fields
A deep learning model with dynamic masks and multi-parameter constraints predicts solar vector magnetic fields over 12 hours, achieving SSIM 0.912 and CC 0.998 for the radial component with 7.82% unsigned flux error.