Short-term wind speed forecasting using recurrent neural networks with error correction

As a type of clean energy, wind energy has been effectively used in power systems. However, due to the influence of the atmospheric boundary layer, wind speed exhibits strong nonlinearity and nonstationarity. Therefore, the accurate and stable prediction of wind speed is highly important for the security of the power grid. To improve the forecasting accuracy, a novel hybrid forecasting system is proposed in this paper that includes effective data decomposition techniques, recurrent neural network prediction algorithms and error decomposition correction methods. In this system, a novel decomposition approach is used to first decompose the original wind speed series into a set of subseries, then it predicts the wind speed by recurrent neural network, and finally, it decomposes the error to correct the previously predicted wind speed. The effectiveness of the proposed model is verified using data from four different wind farms in China. The results show that the proposed hybrid system is superior to other single models and traditional models and realizes highly accurate prediction of wind speed. The proposed system may be a useful tool for smart grid operation and management. (C) 2020 Elsevier Ltd. All rights reserved.

Automated summary

This study presents a novel hybrid forecasting system that significantly enhances the accuracy of wind speed prediction, achieving superior prediction results compared to other models through a process of decomposition, prediction, and error correction.