Sunspot prediction by a Time Delay line Recurrent Fuzzy Neural Network using emotional learning

Sunspot time series have a nonlinear, chaotic and complex behavior. Although different approaches have been established in order to characterize sunspot´s behavior, but it is an open problem yet and needs more accurate methods. In this article a Time Delay line Recurrent Fuzzy Neural Network (TDLRFNN) has been designed to model sunspot data and also to predict sunspot numbers for future periods. To improve precision and accuracy of prediction, emotional learning has been employed to train parameters of proposed network. Here, 195 years of sunspot numbers are used as train and test data sets. It will be shown that the proposed approach represents better precision in comparison to other approaches.