Reliable Probability Forecast of Solar Flares: Deep Flare Net-Reliable (DeFN-R)

• Published in: The Astrophysical journal Vol. 899; no. 2; pp. 150 - 157
• Main Authors: Nishizuka, Naoto, Kubo, Yuki, Sugiura, Komei, Den, Mitsue, Ishii, Mamoru
• Format: Journal Article
• Language: English
• Published: Philadelphia The American Astronomical Society 01.08.2020; IOP Publishing
• Subjects: Artificial neural networks; Astrophysics; Astrostatistics techniques; Brightening; Classification; Corona; Emissivity; Feature extraction; Magnetic fields; Mathematical models; Multilayer perceptrons; Neural networks; Optimization; Photosphere; Solar active regions; Solar activity; Solar flares; Solar images; Solar magnetic fields; Solar x-ray flares; Space weather; Statistical analysis; Time series analysis; Training
• ISSN: 0004-637X; 1538-4357
• DOI: 10.3847/1538-4357/aba2f2

We developed a reliable probabilistic solar-flare forecasting model using a deep neural network, named Deep Flare Net-Reliable (DeFN-R). The model can predict the maximum classes of flares that occur in the following 24 hr after observing images, along with the event occurrence probability. We detected active regions from 3 × 105 solar images taken during 2010-2015 by Solar Dynamic Observatory and extracted 79 features for each region, which we annotated with flare occurrence labels of X-, M-, and C-classes. The extracted features are the same as used by Nishizuka et al.; for example, line-of-sight/vector magnetograms in the photosphere, brightening in the corona, and the X-ray emissivity 1 and 2 hr before an image. We adopted a chronological split of the database into two for training and testing in an operational setting: the data set in 2010-2014 for training and the one in 2015 for testing. DeFN-R is composed of multilayer perceptrons formed by batch normalizations and skip connections. By tuning optimization methods, DeFN-R was trained to optimize the Brier skill score (BSS). As a result, we achieved BSS = 0.41 for ≥C-class flare predictions and 0.30 for ≥M-class flare predictions by improving the reliability diagram while keeping the relative operating characteristic curve almost the same. Note that DeFN is optimized for deterministic prediction, which is determined with a normalized threshold of 50%. On the other hand, DeFN-R is optimized for a probability forecast based on the observation event rate, whose probability threshold can be selected according to users' purposes.