Charging demand forecasting of electric vehicles considering uncertainties in a microgrid. (15th May 2022)

• Record Type: Journal Article
• Authors: Wu, Chuanshen; Jiang, Sufan; Gao, Shan; Liu, Yu; Han, Haiteng
• Is Part Of: Energy. Volume 247(2022)
• Publication Date: 2022-05-15
• Subjects: Charging demand -- Electric vehicle -- Forecasting -- Microgrid -- Uncertainty; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2022.123475
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The currently increasing integration of electric vehicles (EVs) in microgrids (MGs) has gained significant attention. However, affected by the high uncertainties of weather, traffic, and driver behavior, the charging demand of EVs is difficult to forecast accurately. In this study, an optimal parameter forecasting method is presented to improve the forecasting accuracy of charging demand of EVs in an MG. For the methods of forecasting of EV status by sampling from probability distributions, this study modifies the optimal parameter values of probability distributions within fuzzy sets based on the feedback of EVs that have arrived in an MG. Fuzzy sets are utilized to limit the modification ranges of parameter values for the consideration of robustness. Moreover, the average values of multiple sampling results are calculated to improve the stability of forecasting results. Combined with the forecasted results, this study is executed over a rolling time horizon for energy management of EVs, ensuring that acceptable charge levels are reached at the disconnection times. Simulation results show that, compared with other state-of-the-art forecasting methods, the proposed forecasting method is highly effective in reducing forecasting errors of EVs and, hence, has better performance in regulating the charging of EVs in an MG. Highlights: A method for improving the forecasting accuracy of electric vehicles is proposed. The feedback of electric vehicles is used to modify the optimal parameter values. The fuzzy sets are utilized to limit the modification ranges of parameter values. The stability of forecasting results of electric vehicles is improved. The better energy management effects of electric vehicles are obtained.