Dynamic Energy Management Strategy of Hybrid Electric Vehicles based on Velocity Prediction. Issue 10 (2021)

• Record Type: Journal Article
• Authors: Li, Xiangcheng; Lin, Weipeng; Jin, Yuanyang; Chen, Daxin; chen, Tao
• Is Part Of: IFAC-PapersOnLine. Volume 54:Issue 10(2021)
• Page Start: 363
• Page End: 369
• Publication Date: 2021
• Subjects: Energy management strategy -- Short-term velocity prediction -- Dynamic thresholds; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2021.10.189
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Energy management strategy is used to realize dynamic mode switching of hybrid electric vehicles, which can achieve good energy saving effect. However the vehicle future speed, road traffic flow information and other factors are unknown, excellent energy saving effect can't be realised by the existing energy management strategy. Owe to the development of V2X, more information will be provided, a dynamic energy management strategy of hybrid electric vehicle (HEV) combining velocity prediction with dynamic optimization is proposed to deal with the complex and changeable problems of vehicle running process. The speed prediction layer, threshold optimization layer and strategy execution layer are included in this strategy. The LSTM is used to predict the vehicle speed based on V2X information in the speed prediction layer, so as to predict the future short-term torque demand of the vehicle. Based on the predicted vehicle speed curve, the LS-PSO algorithm is used in the threshold optimization layer to optimize the mode switching thresholds of the energy management strategy and the thresholds are transmited to the executive layer for vehicle energy allocation. After simulation verification, the results show that compared with the strategy of fixed rule, energy consumption of the dynamic energy management strategy is reduced. The average energy consumption in the ideal scenario can be reduced averagely by 22.1%.