Application of convolution neural networks for critical frequency fₒF2 prediction
Journal Title: Solar-Terrestrial Physics - Year 2023, Vol 9, Issue 1
Abstract
Ionosphere has an important impact on the quality of radio communication, radar, and global positioning. One of the essential characteristics describing the state of the ionosphere is its critical frequency fₒF2. Its prediction provides effective modes of operation of technical radio equipment as well as enables calculation of the corrections needed to improve the accuracy of its functioning. Different physical and empirical models are generally used for fₒF2 prediction. This paper proposes an empirical prediction technique based on machine learning methods and observational history. It relies on a regression approach to the prediction based on the known daily quasi-periodicity of ionospheric parameters related to solar illumination. Algorithmically, this approach is implemented in the form of convolutional neural networks with two-dimensional convolution. The input data for the analysis is presented as two-dimensional solar time — date matrices. The model was trained on data from the mid-latitude ionosonde in Irkutsk (RF) and tested using data from several mid-latitude ionosondes: Arti (RF), Warsaw (Poland), Mohe (China). It is shown that the main contribution to the prediction value of fₒF2 is made by the data on the nearest few days before the prediction; the contribution of the remaining days strongly decreases. This model has the following forecast quality metrics (Pearson correlation coefficient 0.928, root mean square error 0.598 MHz, mean absolute error in percent 10.45 %, coefficient of determination 0.861) and can be applied to fₒF2 forecast in middle latitudes.
Authors and Affiliations
Salimov B. G. , Berngardt O. I. , Hmelnov A. E. , Ratovsky K. G. , Kusonsky O. A.
Keywords
Related Articles
- EP ID EP714643
- DOI 10.12737/stp-91202307
- Views 48
- Downloads 0
COMPARISON BETWEEN PROBABILITY DENSITY FUNCTIONS OF VERTICAL ELECTRIC CURRENT IN SOLAR ACTIVE REGIONS BASED ON HMI/SDO AND SOT/HINODE DATA
Studying electric currents in solar active regions (AR) is an essential step in understanding solar activity in general and solar flares in particular. In this paper, we compare probability density functions of vertical...
SPATIAL AND TEMPORAL VARIATIONS OF K CA II LINE PROFILE SHAPES IN DIFFERENT STRUCTURES OF THE SOLAR CHROMOSPHERE. II. DETERMINATION TECHNIQUE AND CORRELATION RELATIONSHIPS BETWEEN THE K CA II LINE PARAMETERS FOR K₁ AND K₂ FEATURES
We have studied two regions located at the base of a coronal hole. For the K₁ intensity minima and K₂ peaks, which form between the upper photosphere and the lower chromosphere and in the lower chromosphere respectively,...
DATABASE OF GEOMAGNETIC OBSERVATIONS IN RUSSIAN ARCTIC AND ITS APPLICATION FOR ESTIMATES OF THE SPACE WEATHER IMPACT ON TECHNOLOGICAL SYSTEMS
An archive of digital 1-min data from Soviet/Russian Arctic magnetic stations has been created, starting from 1983 to the present. The archive includes data from stations deployed along the Arctic coast by various USSR/R...
RELATIONSHIP BETWEEN GEOMAGNETIC STORM DEVELOPMENT AND THE SOLAR WIND PARAMETER ß
We have analyzed the dynamics of solar wind and interplanetary magnetic field (IMF) parameters during the development of 933 isolated geomagnetic storms, observed over the period from 1964 to 2010. The analysis was carri...
SPACE WEATHER: RISK FACTORS FOR GLOBAL NAVIGATION SATELLITE SYSTEMS
Extreme space weather events affect the stability and quality of the global navigation satellite systems (GNSS) of the second generation (GPS, GLONASS, Galileo, BeiDou/Compass) and GNSS augmentation. We review the theory...