Identifying Flux Rope Signatures Using a Deep Neural Network

Santos, L. F. Guedes dos; Narock, Ayris; Nieves‐Chinchilla, Teresa; Núñez, Marlon; Kirk, Michael

doi:10.1007/s11207-020-01697-x

Cited by 18 publications

(28 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This raises several interesting questions: How much data is necessary for a PINN to accurately reconstruct the true environment, and how far out (in space and time) from the observed data can the PINN output be reasonably trusted? These have implications for, e.g., reconstructions of environments or magnetic structures around spacecraft observations (e.g., flux ropes (Slavin et al, 2003;Hasegawa et al, 2006;DiBraccio et al, 2015) and coronal mass ejections (Nieves-Chinchilla et al, 2018;dos Santos et al, 2020)).…”

Section: Sod Shock Tubementioning

confidence: 99%

“…This transformer network + collocation point approach does not seem to be the best option for replacing global simulations in extracting the full temporal and spatial global environment information from satellite observations. But, this technique may work well for reconstructing simpler cases at smaller scales, e.g., equilibrium or force-free flux ropes (Slavin et al, 2003;DiBraccio et al, 2015), current sheets (Hasegawa et al, 2019), magnetopause structures (Hasegawa et al, 2006;Chen and Hau, 2018), coronal mass ejections (e.g., dos Santos et al (2020)), or other geospace magnetic structures.…”

Section: Extension To Higher Dimensionsmentioning

confidence: 99%

See 1 more Smart Citation

Neural Network Reconstruction of Plasma Space-Time

Bard

Dorelli

2021

Front. Astron. Space Sci.

View full text Add to dashboard Cite

We explore the use of Physics-Informed Neural Networks (PINNs) for reconstructing full magnetohydrodynamic solutions from partial samples, mimicking the recreation of space-time environments around spacecraft observations. We use one-dimensional magneto- and hydrodynamic benchmarks, namely the Sod, Ryu-Jones, and Brio-Wu shock tubes, to obtain the plasma state variables along linear trajectories in space-time. These simulated spacecraft measurements are used as constraining boundary data for a PINN which incorporates the full set of one-dimensional (magneto) hydrodynamics equations in its loss function. We find that the PINN is able to reconstruct the full 1D solution of these shock tubes even in the presence of Gaussian noise. However, our chosen PINN transformer architecture does not appear to scale well to higher dimensions. Nonetheless, PINNs in general could turn out to be a promising mechanism for reconstructing simple magnetic structures and dynamics from satellite observations in geospace.

show abstract

Section: Sod Shock Tubementioning

confidence: 99%

Section: Extension To Higher Dimensionsmentioning

confidence: 99%

Neural Network Reconstruction of Plasma Space-Time

Bard

Dorelli

2021

Front. Astron. Space Sci.

View full text Add to dashboard Cite

show abstract

“…Meanwhile, an increase of space-and groundbased data availability has led to more interest in applications of machine learning within the space weather community [see (Camporeale, 2019), and references therein]. Nguyen et al (2018) have explored machine learning techniques for automated identification of ICMEs and dos Santos et al (2020) used a deep neural network to create a binary classifier for flux ropes in the solar wind, determining whether a flux rope was or was not present in a given interval. Recently, Reiss et al (2021) use machine learning to predict the minimum Bz value as a magnetic cloud was sweeping past a spacecraft.…”

Section: Introductionmentioning

confidence: 99%

“…Here, we extend the binary classifier work of dos Santos et al (2020) and explore a neural network's ability to predict the orientation, impact parameter, and chirality of an already identified flux rope. We extend the capabilities presented in Reiss et al (2021) by reconstructing the entire three dimensional magnetic field profile.…”

Section: Introductionmentioning

confidence: 99%

Identification of Flux Rope Orientation via Neural Networks

Narock

Santos

et al. 2022

Front. Astron. Space Sci.

Self Cite

View full text Add to dashboard Cite

Geomagnetic disturbance forecasting is based on the identification of solar wind structures and accurate determination of their magnetic field orientation. For nowcasting activities, this is currently a tedious and manual process. Focusing on the main driver of geomagnetic disturbances, the twisted internal magnetic field of interplanetary coronal mass ejections (ICMEs), we explore a convolutional neural network’s (CNN) ability to predict the embedded magnetic flux rope’s orientation once it has been identified from in situ solar wind observations. Our work uses CNNs trained with magnetic field vectors from analytical flux rope data. The simulated flux ropes span many possible spacecraft trajectories and flux rope orientations. We train CNNs first with full duration flux ropes and then again with partial duration flux ropes. The former provides us with a baseline of how well CNNs can predict flux rope orientation while the latter provides insights into real-time forecasting by exploring how accuracy is affected by percentage of flux rope observed. The process of casting the physics problem as a machine learning problem is discussed as well as the impacts of different factors on prediction accuracy such as flux rope fluctuations and different neural network topologies. Finally, results from evaluating the trained network against observed ICMEs from Wind during 1995–2015 are presented.

show abstract

“…CME evolution and propagation in the heliosphere is still one of the critical areas of research. In this collection, there are several studies that span from a case event analysis, data analysis, 3D simulation, theoretical understanding of the physical processes associated with the evolution, and identification of the internal structure based on artificial intelligence techniques (Davies et al, 2020;Balmaceda et al, 2020;Desai et al, 2020;Florido-Llinas, Nieves-Chinchilla, and Linton, 2020;dos Santos et al, 2020). As stated by Balmaceda et al (2020) (Editor's choice), a proper characterization of the kinematics of CMEs is important not only for practical purposes, i.e.…”

mentioning

confidence: 99%