Solar Activity Prediction for Alarming Human’s Artifacts via Neurofuzzy Modeling and Singular Spectrum Analysis

solar activity modeling in recent years becomes a major area of investigation, especially due to the advent of satellite technology. Design of reliable alerting and warning systems is of utmost importance and international collaboration is needed to develop accurate prediction methodologies before the next strikes. This paper introduces an alarm system for long term prediction of solar activity indices which is based on combination of spectral analysis and neurofuzzy technique. A combination of singular spectrum analysis and locally linear neurofuzzy modeling technique is used to make accurate long-term prediction of solar extreme events. The principal components obtained from spectral analysis have narrow band frequency spectra and definite linear or nonlinear trends and periodic patterns, hence they are predictable in large prediction horizon. The incremental learning algorithm initiates a model for each of the components as an optimal linear least squares estimation, and adds the nonlinear neurons if they help to reduce error indices over training and validation sets. Simulation results depict the power of proposed method in long-term prediction of some important solar activity indices.

Keywords:

solar activity , geomagnetic disturbances , neurofuzzy modeling , SSA , LoLiMoT