Wind Power Forecasting by Using Artificial Neural Networks and Grubbs Criterion

The new energy power generation is a problem which deserves increasing attention from the power industry in the future to meet the growing demand for electricity and reduce global warming. The new energy power generation technologies including wind power generation have the advantages of clean, low carbon, and renewable, and provides strong support for countries to carry out energy conservation, emission reduction, and environmental protection. The effective operation of wind farms and the optimal management of yield risks are inseparable from wind power forecasting. With the consideration that wind speed and wind power contain singular points and large fluctuation, this paper proposes an improved neural network model for wind power forecasting. First, the singular points of wind speed and wind power are eliminated by using the Grubbs criterion. Then the wind speed and yaw angle are used as parameters to be incorporated into the artificial neural networks for modeling and analysis. Finally, the measured data of a single wind turbine in a wind farm are used for verification analysis. Experimental results show that the model proposed in this paper can provide more accurate and stable forecasting.