An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques

Details An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques

May 2009

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009spd....40.1611y&link_type=abstract

American Astronomical Society, SPD meeting #40, #16.11; Bulletin of the American Astronomical Society, Vol. 41, p.841

Mathematics

Probability

Scientific paper

Solar flares can have strong impact on the near earth space environment (socalled space weather). They produce streams of highly energetic particles in the solar wind that present radiation hazards to spacecraft and astronauts. X-rays and Ultraviolet radiation emitted by solar flares can affect Earth's ionosphere and disrupt long-range radio communications. Direct radio emission at decimetric wavelengths may disturb operation of radars and other devices operating at these frequencies. Our previous studies demonstrated strong correlation between flares and magnetic field topology in the solar photosphere, such as total unsigned magnetic flux, length of the strong-gradient neutral line, and total magnetic energy dissipation. Using these three parameters, in this paper we present a new method to predict flares: a combination of ordinal logistic regression and supporting vector machines. Our method predicts the probability of the occurrence of C-, M-, and X-class flares for a given active region in the 24 hours following the time of the magnetogram. In addition the method also outputs a classification of flare-active or flare-quiet for a given active region. The method presented can lead to fully automatic solar flare forecasting in real-time. As a major advantage, our method is quite scalable: there is no limit on predictive parameters; the codes can be directly used when new solar parameters are added.

Affiliated with

Also associated with

No associations

Rating

An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and An Automated Flare Forecasting System Using Statistical and Machine Learning Techniques will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1112598