Analysis and prediction of chaotic time series

Abstract

Prediction of time-series has a growing interest in many real-world applications such as prediction of solar radiation for effective use of photovoltaic systems, prediction of electric power demand, weather forecasting, business and financial planning. This contribution deals with analysis and prediction of chaotic time series generated from logistic map using feed-forward back-propagation Neural Network. Simulation results, confirm the effectiveness of this model for predicting chaotic time series.