Predicting Solar Flares Using a Long Short-term Memory Network

Hao Liu1,2, Chang Liu1,3,4, Jason T. L. Wang1,2, and Haimin Wang1,3,4

1. Institute for Space Weather Sciences, New Jersey Institute of Technology
2. Department of Computer Science, New Jersey Institute of Technology
3. Big Bear Solar Observatory, New Jersey Institute of Technology
4. Center for Solar-Terrestrial Research, New Jersey Institute of Technology


Abstract

We present a long short-term memory (LSTM) network for predicting whether an active region (AR) would produce a ϒ-class flare within the next 24 hours. We consider three ϒ classes, namely ≥M5.0 class, ≥M class, and ≥C class, and build three LSTM models separately, each corresponding to a ϒ class. Each LSTM model is used to make predictions of its corresponding ϒ-class flares. The essence of our approach is to model data samples in an AR as time series and use LSTMs to capture temporal information of the data samples. Each data sample has 40 features including 25 magnetic parameters obtained from the Space-weather HMI Active Region Patches (SHARP) and related data products as well as 15 flare history parameters. We survey the flare events that occurred from 2010 May to 2018 May, using the GOES X-ray flare catalogs provided by the National Centers for Environmental Information (NCEI), and select flares with identified ARs in the NCEI flare catalogs. These flare events are used to build the labels (positive vs. negative) of the data samples. Experimental results show that (i) using only 14-22 most important features including both flare history and magnetic parameters can achieve better performance than using all the 40 features together; (ii) our LSTM network outperforms related machine learning methods in predicting the labels of the data samples. To our knowledge, this is the first time that LSTMs have been used for solar flare prediction.


Datasets and Source Code

» Click here to download the collection of 4,203 B-class flares, 6,768 C-class flares, 704 M-class flares, and 49 X-class flares used to build the data samples studied in the paper.
» Click M5, M, C to download the data samples in the ≥M5.0 class, ≥M class, and ≥C class respectively described in Table 2 in the paper. Click here to see the explanation of the downloaded files. Click here to download the raw data used to build the data samples described in Table 2.
» Click here to download our Python program, run on GPUs, and a sample dataset used to generate the performance metric values for the ≥C class in Table 5 in the paper.
» Click here to download our DEMONSTRATION Python programs, run on GPUs, and sample datasets where you can run the programs on the datasets and see how our programs make predictions.


Reference

Predicting Solar Flares Using a Long Short-term Memory Network. Liu, H., Liu, C., Wang, J. T. L., Wang, H., ApJ., 877:121, 2019   [GitHub] [MyGitHub]