IFAC PapersOnLine 51-31 (2018) 612–617 ScienceDirect Available online at www.sciencedirect.com 2405-8963 © 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2018.10.146 © 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. 1. INTRODUCTION In recent years, automobile as a resource consumption product has been paid much more attention due to natural resource consumption and people’s increasing awareness of environment (Guo et al. (2017)) and it is still necessary to develop hybrid electric vehicles to reduce emission and energy consumption at current stage because of limitations of technical bottlenecks. Optimization is an indispensable part of hybrid electric vehicle, which mainly includes topology optimization, component size optimization and energy management strategy optimization (Jager et al. (2013)). The optimization of the three aspects mentioned above is interconnected as expounded in (Moura et al. (2010)) in which the relationship between battery capacity and optimal control algorithm applied is analyzed. For energy management strategy, many researches have been done on this topic. The energy management strategy generally consists of rule-based control strategy based on human experience and control strategy based on optimal control theory. The former is composed of determined rule-based control strategy (Tribioli et al. (2014)) and fuzzy rule-based control (Montazeri-Gh and Mahmoodi-K (2016)) which does not depend on accurate mathematical models, the latter is more rich in content such as model predictive control (Guo et al. (2017)), dynamic program- ⋆ This work is supported by the National Key R&D Program of China (2018YFB0104805), National Nature Science Foundation of China (61520106008, 61522307), China Automobile Industry Inno- vation and Development Joint Fund (U1664257) and Program for JLU Science and Technology Innovative Research Team (2017TD- 20). ming (Gao et al. (2015)) and equivalent consumption mini- mization strategy (ECMS) (Guzzella et al. (2007)). Among them, ECMS is one of the most promising strategies that can be applied into practice. The equivalence factor (EF) plays a key role in ECMS. The advantage and limitation of ECMS is also inherited by it. Driving cycle information can be reflected by equivalence factor and at the same time the equivalence factor of a certain driving cycle is not suitable for another driving cycle (Guzzella et al. (2007)). To overcome the drawback of ECMS, different solutions are put forward by many scholars. Within the framework of intelligent transporta- tion, the driving cycle information ahead is predicted using predictive control (Payri et al. (2014)) or neural network (Tianheng et al. (2015)), the power demand of predictive horizon is then transformed into a referential state of charge (SoC ) by means of mathematical transformation. The equivalence factor is adjusted based on the difference between the actual SoC and the referential SoC . The equivalence factor is altered by feedback of SoC using P or PI controller (Sciarretta et al. (2014)). Dynamic programming and Pontryagin’s minimimum principle are different methods for solving optimal control problems, the information concerning equivalence factor can be extracted from the optimal results of dynamic programming and based on the information the optimal equivalence factor is determined by Pontryagin’s minimimum principle (Han et al. (2014))(Kim et al. (2012)). For the literatures (Zeng and Wang (2016)) and (Yang et al. (2017)), the path trav- eled by an automobile is divided into multiple segments, each segment corresponds to a constant optimal equiva- lence factor and meanwhile the disturbance of SoC is also Keywords: Hybrid Electric Vehicle, Energy Management, Calculation of Equivalence Factor, Dynamic Programming. Abstract: Equivalence factor plays a key role in equivalent consumption minimization strategy and due to its strong relevance with driving cycle, the change of driving cycle can be reflected by the equivalence factor and at the same time different driving cycle has different equivalence factor. In this paper, the optimal equivalence factor as one of the control variables is calculated by dynamic programming. The three initial maps of the optimal equivalence factor concerning state of charge and power demand are formulated over NEDC, UDDS, 10-15 mode cycle respectively. These maps are fused together to obtain the final one for simulation analysis. Compared with the rule-based control strategy, the proposed method has a better fuel economy over NEDC, UDDS, 10-15 mode cycle. * State Key Laboratory of Automotive Simulation and Control, Jilin University, PR China (e-mail: gaobz@jlu.edu.cn, Tel: +86-431-85095198). ** College of Automotive Engineering, Jilin University,Changchun 130022 Qing Zheng * Haorui Yuan *,** Jinzhu Wu * Bingzhao Gao * Equivalent Consumption Minimization Strategy Based on Dynamic Programming for Plug-in Hybrid Electric Vehicle ⋆