WCSE 2021
ISBN: 978-981-18-1791-5 DOI: 10.18178/wcse.2021.06.055

Electric Load Combination Forecast Method Based on EEMD

Yunfeng Shao, Yajing Wang, Yuanming Sun, Zhongjing Ma, Yang Zhao, Yongqiang Liu

Abstract— Load forecasting is of great significance to improve power system safety and reliability. Aiming at the problems of low electric load forecast accuracy and strong randomness, a combined load forecast method based on ensemble empirical mode decomposition is proposed. First, ensemble empirical mode decomposition is used to decompose the load data into intrinsic mode functions with different frequencies, and the sample matrix is formed according to decomposed components. Then, principal component analysis is used to construct a transformation matrix which is used to reduce the noise of the sample matrix, unit root test is used to judge the stability of each component of the sample matrix after noise reduction. If the component is judged to be stationary, multiple linear regression is used to forecast. If the component is judged to be non-stationary, long short term memory is used to forecast. Superimpose the results of each component to get the final load forecast result. Based on the proposed method, the load of a certain area in Shanxi is forecasted and compared with other methods. The results show that this method can forecast the load more effectively while reducing the noise of the load

Index Terms— load forecasting, EEMD, PCA, LSTM

Yunfeng Shao
Lvliang Power Supply Company, State Grid Shanxi Electric Power Company, CHINA
Yajing Wang
Beijing Institute of Technology, CHINA
Yuanming Sun
Beijing Institute of Technology, CHINA
Zhongjing Ma
Beijing Institute of Technology, CHINA
Yang Zhao
Lvliang Power Supply Company, State Grid Shanxi Electric Power Company, CHINA
Yongqiang Liu
Lvliang Power Supply Company, State Grid Shanxi Electric Power Company, CHINA

[Download]


Cite: Yunfeng Shao, Yajing Wang, Yuanming Sun, Zhongjing Ma, Yang Zhao, Yongqiang Liu, "Electric Load Combination Forecast Method Based on EEMD ," 2021 The 11th International Workshop on Computer Science and Engineering (WCSE 2021), pp. 389-396, Shanghai, China, June 19-21, 2021.