Affiliations: [a] State Grid Hebei Electric Power Research Institute, Shijiazhuang, Hebei, China | [b] State Grid Hebei Electric Power Co., Ltd., Shijiazhuang, Hebei, China | [c] School of Electrical and Control Engineering, North China University of Technology, Beijing, China
Correspondence:
[*]
Corresponding author: Rui Ma, State Grid Hebei Electric Power Research Institute, Shijiazhuang, 050021, Hebei, China. E-mail: [email protected].
Abstract: In the era of increasingly prominent energy crisis and environmental protection issues, New Energy (NE) Power Generation (PG) has increasingly attracted people’s attention with its advantages of cleanness, environmental protection and renewable. However, the output of the generation power of the NE generation system has periodicity, volatility and randomness. After grid connection, it becomes an uncontrollable power source and may affect the security and stability of the grid. Therefore, it is of great significance to carry out in-depth discussion on the power prediction of NE generation. This paper explored the NE PG prediction method from the perspective of photovoltaic (PV) PG, and put forward the PV PG power prediction method: the prediction method combining meteorological data and gray correlation degree, and the prediction method combining PG data and wavelet Neural Network (NN). This paper analyzed the prediction effect after putting forward the generation power prediction method, and evaluated the prediction effect through the sum of squares, mean square error and mean absolute error. The following conclusions were drawn: From the perspective of error sum of squares and mean square error, the prediction method combined with PG data and wavelet NN had better prediction effect; the difference of the average absolute error of the PV power generated by the two prediction methods was 2. From the point of view of the average absolute error, the prediction method combined with PG data and wavelet NN had better prediction accuracy.
Keywords: Generation power prediction, new energy, photovoltaic power generation, neural network
