Predicting Solar Flares Using a Long Short-term Memory Network

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.

Parameters Derived from the SDO/HMI Vector Magnetic Field Data: Potential to Improve Machine-learning-based Solar Flare Prediction Models

“…With quickly increasing interest in and success of pilot studies in the solar-physics community, machinelearning technology has opened a new window into the space weather forecast. Various models have been investigated and developed (e.g., Ahmed et al 2013;Bobra & Couvidat 2015;Liu et al 2017aLiu et al , 2019Nishizuka et al 2017Nishizuka et al , 2018Benvenuto et al 2018;Florios et al 2018;Huang et al 2018;Jonas et al 2018).…”

Generative Adversarial Networks for Stochastic Video Prediction With Action Control

“…In future work we plan to extend the VPGAN framework for video processing in scientific domains (e.g., solar physics). In solar physics, deep learning has drawn a lot of interest due to its effectiveness in processing big and complex observational data gathered from diverse instruments [43]. Video is the most common form of observational data.…”

Space Weather Related to Solar Eruptions With the ASO-S Mission

“…Because of the random nature of solar flares, it is difficult to find an effective precursor. In the recent development of solar flare forecast, deep learning methods [76][77][78][79] are used to automatically extract forecasting patterns from the observational data and finally build a forecasting model. Depending on the big observational solar data, deep learning methods may be one of ways to improve Figure 4 illustrates the structure of the CNN.…”