A Two-Stage Multi-Agent System to Predict the Unsmoothed Monthly Sunspot Numbers

摘要: A multi-agent system is developed here to predict monthly details of the upcoming peak 24 th solar magnetic cycle. While studies typically timing and magnitude cycle peaks using annual data, this one utilizes unsmoothed sunspot number instead. Monthly numbers display more pronounced fluctuations during periods strong activity than numbers. Because activities may cause significant economic damages, predicting variations should provide different perhaps helpful information for decision-making purposes. The operates in two stages. In first, it produces twelve predictions second, uses those deliver a final forecast. Acting as expert agents, genetic programming neural networks produce fits forecasts well According results obtained, next predicted be 156 expected occur October 2011- with an average 136 that year. Keywords—Computational techniques, discrete wavelet transformations, prediction,