AI in real-time will be able to look for solar flares. Artificial intelligence from the University of Colorado in boulder will be able to improve predictions of space weather and to increase the understanding of the solar data, causing the data collected from different satellites on one card, writes “EurekAlert!”.
The change of space weather and emissions from the Sun can affect not only the various technologies on Earth, but on the state space objects: satellites and the ISS. The technology is forecasting Sunny weather is not accurate.
Technology of machine learning, developed by engineers from the National center NOAA environmental information (NCEI), together with scientists from the University of Colorado, analyzes the arrays of satellite data to identify patterns of space weather.
The ability to process solar data in real time is important because the flash appearing on the Sun affect the Earth for several minutes. These methods provide fast, constantly updated overview of the characteristics of the Sun, and can point us to areas requiring more careful study. Rob Steenberg, a forecaster at the prediction Center space weather NOAA
Now scientists for the analysis of solar activity twice a day studying the Sun. Often they use hand-drawn map marked with regions of active zones, filaments and coronal boundaries of the plots. However, solar thermal imagers transmit data to the Ground every few minutes, but scientists simply do not have time to watch them because it takes a long time.
Artificial intelligence will analyze the data sets derived from satellites to study the Sun, and will be able to tell astronomers a ready-made forecasts of space weather. At the same time for the algorithm will be able to explore hundreds of different variations of the development of space weather based on each report received from the satellites.
Officially, the instrument will enter into operation before the end of 2019, according to the developers.