Determining the parameter for the linear force-free magnetic field model with multi-dipolar configurations using deep neural networks

Benson, Bernard; Pan, W. David; Gary, G. A.; Hu, Qiang; Staudinger, Tyler

doi:10.1016/j.ascom.2018.11.002

Cited by 3 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cost/loss function is calculated by "backpropagation" of the cost function gradients (Rumelhart et al 1986). ML techniques, including the NNs, have been successfully applied to prediction and forecasting problems in heliophysics (Fernandez Borda et al 2002;Qu et al 2003;Li et al 2007;Qahwaji & Colak 2007;Wang et al 2008;Yuan et al 2010;Ahmed et al 2013;Bobra & Couvidat 2015;Jonas et al 2018;Benson et al 2019Benson et al , 2020Benson et al , 2021. Camporeale (2019) provides an extensive review of the methods and applications of ML research in heliophysics.…”

Section: Improving Arrival Times With MLmentioning

confidence: 99%

Improving the Arrival Time Estimates of Coronal Mass Ejections by Using Magnetohydrodynamic Ensemble Modeling, Heliospheric Imager Data, and Machine Learning

Singh

Benson²,

Raza

et al. 2023

ApJ

Self Cite

View full text Add to dashboard Cite

The arrival time prediction of coronal mass ejections (CMEs) is an area of active research. Many methods with varying levels of complexity have been developed to predict CME arrival. However, the mean absolute error (MAE) of predictions remains above 12 hr, even with the increasing complexity of methods. In this work we develop a new method for CME arrival time prediction that uses magnetohydrodynamic simulations involving data-constrained flux-rope-based CMEs, which are introduced in a data-driven solar wind background. We found that for six CMEs studied in this work the MAE in arrival time was ∼8 hr. We further improved our arrival time predictions by using ensemble modeling and comparing the ensemble solutions with STEREO-A and STEREO-B heliospheric imager data. This was done by using our simulations to create synthetic J-maps. A machine-learning (ML) method called the lasso regression was used for this comparison. Using this approach, we could reduce the MAE to ∼4 hr. Another ML method based on the neural networks (NNs) made it possible to reduce the MAE to ∼5 hr for the cases when HI data from both STEREO-A and STEREO-B were available. NNs are capable of providing similar MAE when only the STEREO-A data are used. Our methods also resulted in very encouraging values of standard deviation (precision) of arrival time. The methods discussed in this paper demonstrate significant improvements in the CME arrival time predictions. Our work highlights the importance of using ML techniques in combination with data-constrained magnetohydrodynamic modeling to improve space weather predictions.

show abstract

Section: Improving Arrival Times With MLmentioning

confidence: 99%

Improving the Arrival Time Estimates of Coronal Mass Ejections by Using Magnetohydrodynamic Ensemble Modeling, Heliospheric Imager Data, and Machine Learning

Singh

Benson²,

Raza

et al. 2023

ApJ

Self Cite

View full text Add to dashboard Cite

show abstract

On the Estimation of the SHARP Parameter MEANALP from AIA Images Using Deep Neural Networks

et al. 2021

View full text Add to dashboard Cite

Solar force-free magnetic fields

Wiegelmann

Sakurai

2021

Living Rev Sol Phys

View full text Add to dashboard Cite

The structure and dynamics of the solar corona is dominated by the magnetic field. In most areas in the corona magnetic forces are so dominant that all non-magnetic forces such as plasma pressure gradients and gravity can be neglected in the lowest order. This model assumption is called the force-free field assumption, as the Lorentz force vanishes. This can be obtained by either vanishing electric currents (leading to potential fields) or the currents are co-aligned with the magnetic field lines. First we discuss a mathematically simpler approach that the magnetic field and currents are proportional with one global constant, the so-called linear force-free field approximation. In the generic case, however, the relationship between magnetic fields and electric currents is nonlinear and analytic solutions have been only found for special cases, like 1D or 2D configurations. For constructing realistic nonlinear force-free coronal magnetic field models in 3D, sophisticated numerical computations are required and boundary conditions must be obtained from measurements of the magnetic field vector in the solar photosphere. This approach is currently a large area of research, as accurate measurements of the photospheric field are available from ground-based observatories such as the Synoptic Optical Long-term Investigations of the Sun and the Daniel K. Inouye Solar Telescope (DKIST) and space-born, e.g., from Hinode and the Solar Dynamics Observatory. If we can obtain accurate force-free coronal magnetic field models we can calculate the free magnetic energy in the corona, a quantity which is important for the prediction of flares and coronal mass ejections. Knowledge of the 3D structure of magnetic field lines also help us to interpret other coronal observations, e.g., EUV images of the radiating coronal plasma.

show abstract

Determining the parameter for the linear force-free magnetic field model with multi-dipolar configurations using deep neural networks

Cited by 3 publications

References 19 publications

Improving the Arrival Time Estimates of Coronal Mass Ejections by Using Magnetohydrodynamic Ensemble Modeling, Heliospheric Imager Data, and Machine Learning

Improving the Arrival Time Estimates of Coronal Mass Ejections by Using Magnetohydrodynamic Ensemble Modeling, Heliospheric Imager Data, and Machine Learning

On the Estimation of the SHARP Parameter MEANALP from AIA Images Using Deep Neural Networks

Solar force-free magnetic fields

Contact Info

Product

Resources

About