Fuel optimal control of parallel hybrid electric vehicles

Abstract

A mathematical model for fuel optimal control and its corresponding dynamic programming (DP) recursive equation were established for an existing parallel hybrid electric vehicle (HEV). Two augmented cost functions for gear shifting and engine stop-starting were designed to limit their frequency. To overcome the problem of numerical DP dimensionality, an algorithm to restrict the exploring region was proposed. The algorithm significantly reduced the computational complexity. The system model was converted into real-time simulation code by using MATLAB/RTW to improve computation efficiency. Comparison between the results of a chassis dynamometer test, simulation, and DP proves that the proposed method can compute the performance limitation of the HEV within an acceptable time period and can be used to evaluate and optimize the control strategy.

References

1. Tate E D Boyd S P Finding ultimate limits ofperformance for hybrid electric vehiclesSAE Paper 00FTT-50 1998
2. Peng Tao Chen Quanshi Tian Guangyu Li Haichen Parametermatching of parallel hybrid electric vehicle powertrainChinese Journal of Mechanical Engineering 2003 39(2)6973
3. Pu Jinhuan Yin Chengliang Zhang Jianwu Ma Dengze Modelingand development of the control strategy for a hybrid carJournal of Shanghai Jiaotong University 2004 38(11)19171921
4. Rimaux S Delhom M Combes E Hybrid vehicle powertrain: modeling and controlIn: Proceedings of the 16th International ElectricVehicle SymposiumBeijingEVAAP 1999
5. Lin C C Kang J M Grizzle J W Peng H Energy managementstrategy for a parallel hybrid electric truckIn: Proceedings of American Control ConferenceArlington, VA, USAIEEE 2001 28782883
6. Lin C C Peng H Grizzle J W et al.Power management strategy for a parallel hybridelectric truckIEEE Transactions on ControlSystems Technology 2003 11(6)839849. doi:10.1109/TCST.2003.815606
7. Bertsekas D P DynamicProgramming and Optimal Control2nd edition. Vol.1 Belmont, MassAthena Scientific 2000
8. De Madrid A P Dormido S Morilla F Reduction of the dimensionality of dynamic programming:a case studyIn: Proceedings of AmericanControl ConferenceSan Diego, California,USAIEEE 1999 28522856