AUTHOR=Wang Jingjing , Luo Bingxian , Liu Siqing , Shi Liqin 

TITLE=A machine learning-based model for the next 3-day geomagnetic index (Kp) forecast

JOURNAL=Frontiers in Astronomy and Space Sciences

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/astronomy-and-space-sciences/articles/10.3389/fspas.2023.1082737

DOI=10.3389/fspas.2023.1082737

ISSN=2296-987X

ABSTRACT=The 3-day Kp forecast product is important and necessary for space weather forecast. There is some essential information that can be obtained from the 3-day Kp forecast product, such as \textcolor{red}{the start time of the geomagnetic storm}, the maximum storm level, the storm duration. In this study, we aimed to predict the next 3-day Kp index based on the previous Kp time-series and SDO/AIA 193 \r{A} images. We prepared dataset from May 2010 to Dec 2019 for training and dataset from Jan 2020 to Oct 2022 for testing. The similarity parameters of the previous and current geomagnetic conditions between the samples are calculated and analyzed. We assumed that the paired samples with high similarity of the previous and current geomagnetic conditions will also have high similarity of the next 3-day geomagnetic conditions. Based on the assumption, we selected the three best similarity parameters by feature selection process and adopted Scalable Tree Boosting System (XGBoost) to develop a prediction model. It took the similarity parameters of the previous and current geomagnetic conditions as input, and provided the best match sample from the training subset as a forecast. For the next 3-day period non-storm (maximum Kp $<$ 5) prediction, our model reached a F1-score of 0.96. For the next 3-day period storm (maximum Kp $\ge$ 5) prediction, our model reached a F1-score of 0.82, a recall of 0.70, and a precision of 0.98.