DECISION MAKING SUPPORT SYSTEM WITH ADAPTIVE UNITS OF RESTORING AND FORECASTING OF SOLAR RADIO FLUXES

Abstract

Context. Creation of new methods for solar data restoring and forecasting, new decision making support systems for processing of solar radio fluxes at a wavelength of 10.7 cm are topical problems, because it will give an opportunity to perform preliminary data processing and further forecasting correctly and automated. Objective. To develop moving two-sided exponential smoothing method for restoring of missed values and forecasting of time series; to create integral criterion of model adequacy and the proximity criterion for using for restoring of the true regularities of time series evolution; to develop decision making support system for analysis and forecasting of solar radio fluxes at a wavelength of 10.7 cm with using of developed methods; to apply developed methods for real data and to compare with traditional methods. Method. For achieving of formulated goal the following methods were used: developed moving two-sided exponential smoothing method; exponential smoothing method; 13-month running mean method; exponential approach, which was suggested by Hathaway, Wilson and Reichmann. Results. Moving two-sided exponential smoothing method for restoring of missed values of time series and for forecasting of time series was developed. The integral criterion of model adequacy and the proximity criterion were created. Decision making support system for analysis and forecasting of solar data was developed. Practical application of developed methods for real data and comparison with traditional methods were performed. Conclusions. Developed moving two-sided exponential smoothing method is shown superiority in comparison with all traditional methods in the restoring of missed values, true regularities and forecasting of solar radio fluxes.