This paper presents an application of the neuro-fuzzy modeling to analyze the time series of solar activity, as measured through the relative Wolf number. The neuro-fuzzy structure is optimized based on the linear adapted genetic algorithm with controlling population size (LAGA-POP). Initially, the dimension of the time series characteristic attractor is obtained based on the smallest regularity criterion (RC) and the neuro-fuzzy model. Then the performance of the proposed approach, in forecasting yearly sunspot numbers, is favorably compared to that of other published methods. Finally, a comparison predictions for the remaining part of the 22nd and the whole 23rd cycle of the solar activity are presented.
Forecasting solar and geomagnetic levels of activity is essential to help plan missions and to design satellites that will survive for their useful lifetimes. Therefore, amplitudes of the upcoming solar cycles and the geomagnetic activity were forecasted using the neuro-fuzzy approach. Results of this work allow us to draw the following conclusions: Two moderate cycles are estimated to approach their maximum sunspot numbers, 110 and 116 in 2011 and 2021, respectively. However, the predicted geomagnetic activity shown to be in phase with the peak of the 24th sunspot cycle will reach its minimum three years earlier, then it will rise sharply to reach the 25th maximum a year earlier (i.e., 2020). Our analysis of the three-century long sunspot number data-set suggests that the quasi-periodic variation of the long-term evolution of solar activity could explain the irregularity of the short-term cycles seen during the past decades.
This paper presents an application of the neuro-fuzzy modeling to analyze the time series of solar activity, as measured through the relative Wolf number. The neuro-fuzzy structure will be optimized based on the linear adapted genetic algorithm with controlling population size (LAGA-POP). First, the dimension of the time series characteristic attractor is obtained based on the smallest Regularity Criterion (RC) and the neuro-fuzzy modeling. Second, after describing the neuro-fuzzy structure and optimizing its parameters based on LAGA-POP, the performance of the present approach in forecasting yearly sunspot numbers is favorably compared to that of other published methods. Finally, the comparison predictions for the remaining part of the 22 nd and the whole 23 rd cycle of solar activity are presented.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.