CME Arrival Time Prediction Using Convolutional Neural Network

Wang, Yimin; Liu, Jiajia; Jiang, Ye; Erdélyi, R.

doi:10.3847/1538-4357/ab2b3e

Cited by 34 publications

(39 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even if only low corona imaging observations are available, prediction of CME arrival at L1 is still expected as demonstrated by e.g. Wang et al (2019).…”

Section: Expanding the Capabilities Of Sammmentioning

confidence: 99%

“…Some recent, pioneering approaches with varying degrees of success attempt already to incorporate solar atmospheric extreme ultraviolet (EUV) data and/or use Machine Learning in order to improve forecasting accuracy (see e.g. Qahwaji and Colak, 2007;Bobra and Couvidat, 2015;Anastasiadis et al, 2017;Florios et al, 2018;Kim et al, 2019;Wang et al, 2019;Campi et al, 2019;Camporeale, 2019). Caveats of applying ML techniques is also addressed in Liu et al (2021).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Solar Activity Monitor Network – SAMNet

Erdélyi

Korsós

Huang

et al. 2022

J. Space Weather Space Clim.

Self Cite

View full text Add to dashboard Cite

The Solar Activity Magnetic Monitor (SAMM) Network (SAMNet) is a future UK-led international network of ground-based solar telescope stations. SAMNet, at its full capacity, will continuously monitor the Sun’s intensity, magnetic and Doppler velocity fields at multiple heights in the solar atmosphere (from photosphere to upper chromosphere). Each SAMM sentinel will be equipped with a cluster of identical telescopes each with different magneto-optical filter (MOFs) to take observations in K~I, Na~D and Ca~I spectral bands. A subset of SAMM stations will have white-light coronagraphs and emission line coronal spectropolarimeters. The objectives of SAMNet are to provide observational data for the space weather research and forecast. The goal is to achieve an operationally sufficient lead time of e.g. flare warning of 2-8 hours, and provide much sought-after continuous synoptic maps (e.g., LoS magnetic and velocity fields, intensity) of the lower solar atmosphere with a spatial resolution limited only by seeing or diffraction limit, and with a cadence of 10 minutes. The individual SAMM sentinels will be connected into their master HQ hub where data received from all the slave stations will be automatically processed and flare warning issued up to 26 hrs in advance.

show abstract

“…Even if only low corona imaging observations are available, prediction of CME arrival at L1 is still expected as demonstrated by e.g. Wang et al (2019).…”

Section: Expanding the Capabilities Of Sammmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Solar Activity Monitor Network – SAMNet

Erdélyi

Korsós

Huang

et al. 2022

J. Space Weather Space Clim.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most flare and CME forecast methods apply photospheric magnetic and Doppler data of ARs for forecasting. Some recent, pioneering approaches with various degrees of success attempt to incorporate solar atmospheric extreme ultraviolet (EUV) data and/or use machine learning in order to improve forecasting accuracy (see, e.g., Qahwaji & Colak 2007;Bobra & Couvidat 2015;Florios et al 2018;Campi et al 2019;Kim et al 2019;Wang et al 2019). Detailed information on measuring, and the consequent modeling, of the 3D magnetic field structure of an AR would be important to obtain more accurate insight into the preflare evolution locally in the solar atmosphere.…”

Section: Introductionmentioning

confidence: 99%

Solar Flare Prediction Using Magnetic Field Diagnostics above the Photosphere

Korsós

Georgoulis

Gyenge

et al. 2020

ApJ

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, Liu et al [23] introduced RNN to analysis the relation between an active region and CMEs. Wang et al [37] and Liu et al [24] introduced CNN and CAT-PUMA to predict CMEs arrival time, respectively. However, to the best of our knowledge, discovering the important factors of CME arrival time has not been documented.…”

Section: Coronal Mass Ejection Analysismentioning

confidence: 99%

Distribution-dependent feature selection for deep neural networks

Zhao

Chen

et al. 2021

Appl Intell

View full text Add to dashboard Cite

While deep neural networks (DNNs) have achieved impressive performance on a wide variety of tasks, the black-box nature hinders their applicability to high-risk, decision-making fields. In such fields, besides accurate prediction, it is also desired to provide interpretable insights into DNNs, e.g., screening important features based on their contributions to predictive accuracy. To improve the interpretability of DNNs, this paper originally proposes a new feature selection algorithm for DNNs by integrating the knockoff technique and the distribution information of irrelevant features. With the help of knockoff features and central limit theorem, we state that the irrelevant feature's statistic follows a known Gaussian distribution under mild conditions. This information is applied in hypothesis testing to discover key features associated with the DNNs. Empirical evaluations on simulated data demonstrate that the proposed method can select more true informative features with higher F 1 scores. Meanwhile, the Friedman test and the post-hoc Nemenyi test are employed to validate the superiority of the proposed method. Then we apply our method to Coronal Mass Ejections (CME) data and uncover the key features which contribute to the DNN-based CME arrival time.

show abstract

CME Arrival Time Prediction Using Convolutional Neural Network

Abstract: This is a repository copy of CME arrival time prediction using convolutional neural network.

Cited by 34 publications

References 45 publications

The Solar Activity Monitor Network – SAMNet

The Solar Activity Monitor Network – SAMNet

Solar Flare Prediction Using Magnetic Field Diagnostics above the Photosphere

Distribution-dependent feature selection for deep neural networks

Contact Info

Product

Resources

About