The invention relates to the field of fuel cell hybrid electric vehicle design, in particular to a vehicle-following energy management method based on safety reinforcement learning, which comprises the following steps of: firstly, establishing a controlled vehicle dynamic model and a fuel cell and lithium battery experience aging model in a vehicle-following scene; a hierarchical reinforcement learning vehicle-following energy management framework based on multiple agents is designed by using a double-delay depth determination strategy gradient algorithm, and collaborative optimization of vehicle external motion control and internal power system energy management is realized. According to the method, a controlled vehicle dynamic model and a fuel cell and lithium battery experience aging model in a vehicle following scene are established, actual vehicle distances, vehicle speeds, a fuel cell SoH and a lithium battery SoH at different sampling moments are obtained, and according to the obtained state information, the controlled vehicle is taken as the environment, and the real-time performance of the controlled vehicle is improved. And designing a multi-agent-based layered reinforcement learning vehicle-following energy management framework by using the TD3, and realizing collaborative optimization of external motion control and internal power system energy management of the vehicle.

    本发明涉及燃料电池混合动力汽车设计领域,尤其涉及一种基于安全强化学习的跟车能量管理方法,首先建立跟车场景下被控车辆动力学模型、燃料电池和锂电池经验老化模型,其次以被控车辆本身为环境,利用双延迟深度确定策略梯度算法设计基于多智能体的分层强化学习跟车能量管理框架,实现车辆外部运动控制和内部动力系统能量管理的协同优化。本发明建立跟车场景下被控车辆动力学模型、燃料电池和锂电池经验老化模型,获取不同采样时刻下实际车间距、车辆速度、燃料电池SoH和锂电池SoH,根据所获状态信息,以被控车辆本身为环境,利用TD3设计基于多智能体的分层强化学习跟车能量管理框架,实现车辆外部运动控制和内部动力系统能量管理的协同优化。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Car following energy management method based on safety reinforcement learning


    Weitere Titelangaben:

    一种基于安全强化学习的跟车能量管理方法


    Beteiligte:
    FU ZHUMU (Autor:in) / ZHU LONGLONG (Autor:in) / TAO FAZHAN (Autor:in) / WANG NAN (Autor:in) / WANG JUN (Autor:in) / GAO SONG (Autor:in) / YANG YI (Autor:in) / CHEN LINGFENG (Autor:in) / GAO XIANGQIAN (Autor:in) / MA HAOXIANG (Autor:in)

    Erscheinungsdatum :

    12.09.2023


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    HEV energy management method based on deep reinforcement learning in vehicle following environment

    TANG XIAOLIN / CHEN JIAXIN / YANG KAI et al. | Europäisches Patentamt | 2020

    Freier Zugriff


    Hybrid electric vehicle energy management method based on reinforcement learning

    LOU DIMING / ZHAO YINGHUA | Europäisches Patentamt | 2021

    Freier Zugriff


    HEV energy management method based on deep reinforcement learning A3C algorithm

    TANG XIAOLIN / CHEN JIAXIN / PU HUAYAN et al. | Europäisches Patentamt | 2020

    Freier Zugriff