Kangwoo Yi

Kangwoo Yi

Auto-generated profile, based on publicly available information

Kangwoo Yi

New Jersey Institute of Technology Newark NJ USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2021 in The Astrophysical Journal
Verified
Journal Article
Closed access
KYKangwoo Yi
Kangwoo Yi
YMYong-Jae Moon
Yong-Jae Moon
DLDaye Lim
Daye Lim
EPEunsu Park
Eunsu Park
HLHarim Lee
Harim Lee
In this study, we present a visual explanation of a deep learning solar flare forecast model and its relationship to physical parameters of solar active regions (ARs). For this, we use full-disk magnetograms at 00:00 UT from the Solar and Heliospheric Observatory/Michelson Doppler Imager and the Solar Dynamics Observatory/Helioseismic and Magnetic Imager, physical parameters from the Space-weather HMI Active Region Patch (SHARP), and Geostationary Operational Environmental Satellite X-ray flare data. Our deep learning flare forecast model based on the Convolutional Neural Network (CNN) predicts “Yes” or “No” for the daily occurrence of C-, M-, and X-class flares. We interpret the model using two CNN attribution methods (guided backpropagation and Gradient-weighted Class Activation Mapping [Grad-CAM]) that provide quantitative information on explaining the model. We find that our deep learning flare forecasting model is intimately related to AR physical properties that have also been distinguished in previous studies as holding significant predictive ability. Major results of this study are as follows. First, we successfully apply our deep learning models to the forecast of daily solar flare occurrence with TSS = 0.65, without any preprocessing to extract features from data. Second, using the attribution methods, we find that the polarity inversion line is an important feature for the deep learning flare forecasting model. Third, the ARs with high Grad-CAM values produce more flares than those with low Grad-CAM values. Fourth, nine SHARP parameters such as total unsigned vertical current, total unsigned current helicity, total unsigned flux, and total photospheric magnetic free energy density are well correlated with Grad-CAM values.

7 February 2020 in The Astrophysical Journal Letters
Verified
Journal Article
Closed access
KYKangwoo Yi
Kangwoo Yi
YMYong-Jae Moon
Yong-Jae Moon
GSGyungin Shin
Gyungin Shin
DLDaye Lim
Daye Lim
In this Letter, we present the application of a couple of novel deep learning models to the forecast of major solar X-ray flare flux profiles. These models are based on a sequence-to-sequence framework using long short-term memory cell and an attention mechanism. For this, we use Geostationary Operational Environmental Satellite 10 X-ray flux data from 1998 August to 2006 April. Seven hundred sixty events are used for training and 85 for testing. The models forecast 30 minutes of X-ray flux profiles during the rise phase of the solar flare with a minute time cadence. We evaluate the models using the 10-fold cross-validation and rms error (RMSE) based on flux profiles and RMSE based on its peak flux. For comparison we consider two simple deep learning models and four conventional regression models. Major results of this study are as follows. First, we successfully apply our deep learning models to the forecast of solar flare X-ray flux profiles, without any preprocessing to extract features from data. Second, our proposed models outperform the other models. Third, our models achieve better performance for forecasting X-ray flux profiles with low-peak fluxes than those with high-peak fluxes. Fourth, our models successfully predict flare duration with high correlations for both all cases and cases at peak times. Our study indicates that our deep learning models can be useful for forecasting time-series data in astronomy and space weather, even for impulsive events such as major flares.

15 March 2023 in The Astrophysical Journal Supplement Series
Verified
Journal Article
Open access
KYKangwoo Yi
Kangwoo Yi
YMYong-Jae Moon
Yong-Jae Moon
HJHyun-Jin Jeong
Hyun-Jin Jeong
In this study, we present the application of deep reinforcement learning to the forecasting of major solar flares. For this, we consider full-disk magnetograms at 00:00 UT from the Solar and Heliospheric Observatory/Michelson Doppler Imager (1996–2010) and the Solar Dynamics Observatory/Helioseismic and Magnetic Imager (2011–2019), as well as Geostationary Operational Environmental Satellite X-ray flare data. We apply Deep Q-Network (DQN) and Double DQN, which are popular deep reinforcement learning methods, to predict “Yes or No” for daily M- and X-class flare occurrence. The reward functions, consisting of four rewards for true positive, false positive, false negative, and true negative, are used for our models. The major results of this study are as follows. First, our deep-learning models successfully predict major solar flares with good skill scores, such as HSS, F1, TSS, and ApSS. Second, the performance of our models depends on the reward function, learning method, and target agent update time. Third, the performance of our deep-learning models is noticeably better than that of a convolutional neural network (CNN) model with the same structure: 0.38 (CNN) to 0.44 (ours) for HSS, 0.47 to 0.52 for F1, 0.53 to 0.59 for TSS, and 0.09 to 0.12 for ApSS.

1 August 2024 in The Astrophysical Journal Letters
Verified
Journal Article
Open access
PDF file
JLJaewon Lee
Jaewon Lee
YMYong-Jae Moon
Yong-Jae Moon
HJHyun-Jin Jeong
Hyun-Jin Jeong
KYKangwoo Yi
Kangwoo Yi
HLHarim Lee
Harim Lee
We address the question of whether the solar limb flare prediction can be properly made by EUV intensity, which has less projection effects than solar white light and magnetogram data. We develop empirical and multilayer perceptron (MLP) models to forecast the probability of a major solar limb flare within a day. We use Solar Dynamics Observatory (SDO)/Atmospheric Imaging Assembly (AIA) 94 and 131 Å that have high correlations and large slopes with X-ray flare fluxes from 2010 to 2022. We select 240 flares stronger than or equal to the M1.0 class and located near the limb region (60° or more in heliographic longitude). For input data, we use the limb intensity as the sum of SDO/AIA intensities in the limb region and the total intensity of the whole image. We compare the model performances using metrics such as the receiver operating characteristic—area under the curve. Our major results are as follows. First, we can forecast major solar limb flare occurrences with only SDO/AIA 94 and/or 131 Å intensities. Second, our models show better probability prediction than the climatological model. Third, both empirical (AUC = 0.85) and MLP (AUC = 0.84) models have similar performances, which are much better than a random forecast (AUC = 0.50). Finally, it is interesting to note that our model can forecast the flaring probability of all 52 events during the test period, while the models in the NASA/CCMC flare scoreboard can forecast only 22 events. From the above results, we can answer that the solar limb flare prediction using EUV intensity can be properly made.

11 February 2026 in Space Weather
Verified
Journal Article
Open access
EJEun‐Young Ji
Eun‐Young Ji
YMYong-Jae Moon
Yong-Jae Moon
YKYoungsil Kwak
Youngsil Kwak
KYKangwoo Yi
Kangwoo Yi
JKJeong‐Heon Kim
Jeong‐Heon Kim
WJWoong Jeon
Woong Jeon
In this study, we propose an AI‐based method to improve the electron density profiles generated by the International Reference Ionosphere (IRI)‐2020 model using the Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) radio occultation (RO) data. Specifically, we employ a Multi‐Layer Perceptron (MLP), a type of artificial neural network (ANN), which learns to transform IRI‐2020 profiles into COSMIC‐like profiles using paired data collected between 2007 and 2019. The data set is divided into training (2007–2013), validation (2014, 2019), and test (2015–2018) subsets. Using the test set, we evaluate the performance of our model by calculating the correlation coefficient (CC) and root mean square error (RMSE) between the model outputs and COSMIC electron density profiles. The results show that our model outperforms the IRI model, yielding higher CCs and lower RMSEs. The model demonstrates consistent improvements across various geomagnetic conditions and geographic regions, particularly in low‐ and mid‐latitudes. Further evaluation using incoherent scatter radar (ISR) data from two stations indicates that both our model and the IRI model show comparable performance in capturing the vertical structure of the ionosphere. These findings demonstrate that machine learning techniques, such as MLPs, provide an effective means of leveraging satellite‐based observational data to improve the performance of empirical models such as the IRI model.