A learning‐based hierarchical energy management control strategy for hybrid electric vehicles

• Published in: IET control theory & applications 2024-12, Vol.18 (18), p.2725-2741
• Main Authors: Chen, Yanfang, Li, Xuefang
• Format: Article
• Language: English
• Subjects: energy management; hybrid electric vehicles; iterative learning model predictive control; speed tracking
• ISSN: 1751-8644; 1751-8652
• DOI: 10.1049/cth2.12749

In this work, a novel energy management control framework is developed for hybrid electric vehicles (HEVs) driving in car‐following scenarios. In order to enhance the energy efficiency while maintaining the driving safety, a hierarchical control approach consisting of an upper level speed tracking control scheme and a lower level energy management control strategy is proposed. For the upper level tracking control system, an iterative learning model predictive control (ILMPC) scheme is developed to guarantee the tracking performance and the driving safety simultaneously. Additionally, a model predictive control (MPC) algorithm is adopted at the lower level to optimize the torque distribution in real‐time based on the driving cycles generated by the upper level control system. With the proposed hierarchical control framework, HEVs are able to improve the energy efficiency significantly by taking the advantages of the operational repeatability. The convergence of the proposed control strategy is analyzed rigorously, and its effectiveness is illustrated through numerical simulations. A hierarchical learning EMS is developed for hybrid electric vehicles (HEVs) in car‐following scenarios. A tube‐based iterative learning model predictive control scheme is proposed for speed tracking under uncertainties. The operation repetition is helpful to improve fuel efficiency of HEVs.