Combined Prediction of Wind Power with Chaotic Time Series Analysis

Wind power prediction is one of the most significant technologies to promote the capability of the whole power system that takes in wind electricity. A combined model for wind power forecasting is presented to decrease the influence of reconstructed parameters by chaotic time series analysis and the neural networks (NNs) in this work. The combined model respectively makes use of linear weighted model and NNs method to achieve combination of several neural networks models through phase space reconstruction after wind power series chaotic characteristics acquisition, which can integrate information and reduce prediction error in different embedding dimension, leading to higher forecast accuracy. Simulation is performed to the real power time series from Meijia wind farm. The results show that the proposed model is more effective than single embedding dimension model and linear weighted combination model, and the prediction error of neural network combination is less than 7%.