Open AccessShort-term load forecasting algorithm based on LSTM-DBN considering the flexibility of electric vehicle
Author(s) -
Zhang Na,
Tao HanZhen,
Yutong Liu,
Jia Cui,
Junyou Yang,
Gang Wang
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/546/4/042001
Subject(s) - flexibility (engineering) , randomness , computer science , term (time) , electric vehicle , monte carlo method , electrical load , algorithm , artificial intelligence , engineering , power (physics) , voltage , mathematics , statistics , physics , quantum mechanics , electrical engineering
Due to the rapid development of electric vehicles, the randomness of charging and discharging modes brings huge challenges to load forecasting. Considering the large-scale gridding of electric vehicles, a load combination forecasting algorithm is proposed in this paper which considers the flexibility of electric vehicles. Firstly, the user behaviors of electric vehicle are modeled based on Monte Carlo method. Secondly, different from traditional single forecasting models, the combination of deep belief network and long short-term memory net-work is used in order to improve the forecasting accuracy and calculation speed for the complex case. Thirdly, a dynamic weight distribution method is proposed to combine the two forecasting results. Finally, Numerical examples have been performed with the results proving that the proposed method outperforms the state-of-art significantly in load forecasting, and the combined model has a higher forecasting accuracy than traditional methods.