Yongqing Zhu, Qingsheng Li, Zhen Li, and Zhaofeng Zhang
Inverse cloud grey correlation similar days, complementary ensemble empirical mode decomposition (CEEMD), fuzzy entropy, improved sparrow search algorithm (ISSA)-long short-term memory network (LSTM) neural network, short-term wind power forecasting
A short-term prediction method for wind and solar power is proposed based on complementary ensemble empirical mode decomposition (CEEMD), fuzzy entropy (FE), improved sparrow search algorithm (ISSA), and long short-term memory network (LSTM). First, the uncertainty of wind and solar power output is represented by a cloud model, and the impact of diļ¬erent meteorological characteristics on output power is analysed by combining reverse cloud with a grey correlation degree. Then, selection criteria and comprehensive scoring indicators are established. Complementary set empirical mode decomposition (CEEMD) is used to decompose the power data of similar days into subsequences. After CEEMD processing, FE is used to restructure the entropy values of each subsequence. Finally, in the improved LSTM model optimised by ISSA, the subsequence and meteorological data are used as prediction inputs for training, thereby predicting wind and solar power generation. The simulation results show that the proposed method takes into account weather factors and has high prediction accuracy.
Important Links:
Go Back
