Ultra-short term prediction of wind power based on multiples model extreme leaning machine

Abstract

Wind energy is considered one of the most rapidly growing energy resources all over the world. The uncertainty in wind generation is very large due to the stochastic in wind, and this needs a good algorithm applying to forecast wind speed and power for grid operator rapidly adjusting management planning. In this paper, Multiple Models Extreme Learning Machine (MMELM) is proposed for the feature of wind. A suspending criterion is designed to separate the whole models into two parts: the suspending models and the updating models. For the suspending models with minor error, the models needn't update online. For the updating models with major error, it must utilize the random selection method to update the model parameters. Finally, the final output value should be the sum of all models outputs multiple their corresponding weight number. The simulation result shows that the fitting accuracy meets the requirement. © 2011 Springer-Verlag.