Research on NARX wind speed prediction network based on Copula

Abstract

Accurate wind speed forecasting is the key to ensure the safe and stable access of wind energy to the grid system. Accurate wind speed forecasting can avoid the unnecessary loss of wind turbines due to high wind speed, and it is also helpful for wind power forecasting. However, the wind speed data is different from the general time series data, and the intermittency of the wind brings great volatility due to the weak continuity.There are a large number of peak points within the range of high wind speed. These peak points make it difficult to achieve accurate wind speed forecasting, no improvement method has been proposed for these wind speed characteristics in existing studies. This work proves the inaccuracy of maximum wind speed prediction through mathematical modeling, analyzes the correlation between meteorological data and wind speed in the data set, and proposes to take the most dependent meteorological data as an additional input to train the network, and through the copula connection function, it is proved that there is a tail dependence between the new parameters and the maximum wind speed. It can be used as a new method to improve the prediction accuracy of high wind speed peak.