Wind Power Prediction Based on Principal Component Analysis and Long Short-Term Memory Networks

In recent years, the scale of new energy represented by wind power and photovoltaic has been continuously developed at a high speed. Due to the strong randomness and volatility of wind energy, it poses a severe challenge for the safety, reliability, stability and economy of grid operation. The effective improvement of wind power prediction accuracy is of great significance to power grid operation. Along with the continuous development of artificial intelligence methods, the emerging wind power prediction methods have been greatly improved in terms of prediction accuracy and calculation rate. As a current predictive model for deep learning, Long Short-Term Memory (LSTM) is excellent for predicting long-term sequences. This paper proposes a wind power prediction method based on LSTM. In addition, Principal Component Analysis (PCA) method is used to achieve the dimension reduction of data. The simulation results demonstrate the effectiveness and advantages of the proposed PCA-LSTM method.