Wind power prediction based on multipositon NWP with rough set theory

Abstract

Wind power prediction is critical to power balance and economic operation of power system when connected to the grid. In order to improve prediction accuracy, NWP information of different positions and height are taken into consideration to predict wind power in wind farms. In this paper, similar day as the prediction day was searched as training sample at first. The key factors of multiposition NWP that affect the wind power prediction are identified by rough set theory. Then the rough set neural network prediction model is built by treating the key factors as the inputs to the model. To test the approach, the NWP data and actual wind power data from a wind farm are used for this study. The prediction results are presented and compared to the single position wind power calculation model, the single position NWP neural network model and persistence model. The results show that rough set method is a useful tool in short term multistep wind power prediction.