An on-line predictive energy management strategy for plug-in hybrid electric vehicles to counter the uncertain prediction of the driving cycle. (1st January 2017)

• Record Type: Journal Article
• Authors: Chen, Zeyu; Xiong, Rui; Wang, Chun; Cao, Jiayi
• Is Part Of: Applied energy. Volume 185:Part 2(2017)
• Page Start: 1663
• Page End: 1672
• Publication Date: 2017-01-01
• Subjects: Plug-in hybrid electric vehicles -- Power management -- Local optimal control -- Predictive control -- Particle swarm optimization; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2016.01.071
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: An online predictive energy management approach was proposed using a DPSO algorithm. Energy optimal problem with a dynamic cost function was formulated. Fuzzy logic-based correction method was proposed to counter the imprecise prediction. The effect of the presented strategy and correction algorithm was verified. Abstract: Predictive energy management could be implemented in real-time with a short period of future driving cycle prediction. However, the completely precise prediction of the future driving cycle remains quite difficult. Two areas of effort have been explored in this study. The first is the implementation of a dynamic-neighborhood particle swarm optimization algorithm in the local optimal energy management strategy of plug-in hybrid electric vehicles based on data from the prediction of the future driving cycle. Second, the influence of an imprecise driving cycle prediction is considered, and then an online correction algorithm is proposed based on the backup control strategy and a fuzzy logic controller. In addition to these efforts, a predictive energy management strategy with an online correction algorithm is finally proposed. Compared with the optimal heuristic method, the presented energy management strategy could reduce the energy by up to 9.7% if the prediction of the future driving cycle is precise. For the situation of imprecise prediction, the online correction algorithm could reduce the deviation from the actual optimal policy by up to 32.39%.