The invention relates to an HEV energy management method based on a deep reinforcement learning A3C algorithm, and belongs to the field of new energy vehicles. The method comprises the steps: S1, dividing vehicle driving standard working conditions according to characteristic parameters of the working conditions; S2, calculating the required power of the whole vehicle; S3, determining a required state variable, an action variable and a reward function; S4, establishing an A3C algorithm model, and setting an environment-agent module; S5, designing and adding a rule-based engine start-stop strategy according to the SOC of the battery and the required power; S6, loading different types of standard working conditions in the environment-agent module, training a deep neural network in the algorithm model in a continuous iterative trial-and-error learning mode, ending the training process after the total rewards are in a stable convergence state, and storing a persistence model of the globalneural network. According to the method, the self-adaptive capacity to all random working conditions is achieved under the condition that the fuel economy is guaranteed.

    本发明涉及一种基于深度强化学习A3C算法的HEV能量管理方法,属于新能源汽车领域。该方法包括:S1:根据工况的特征参数划分车辆行驶标准工况;S2:计算整车的需求功率;S3:确定需要的状态变量、动作变量以及奖励函数;S4:建立A3C算法模型,设定环境‑智能体模块;S5:根据电池SOC以及需求功率的大小,设计并加入基于规则的发动机启停策略;S6:环境‑智能体模块中分别加载不同类型的标准工况,通过不断迭代试错的学习方式训练算法模型中的深度神经网络,当总奖励处于稳定收敛状态后结束训练过程,并且保存全局神经网络的持久化模型。本发明在保证燃油经济性的条件下实现对所有随机工况的自适应能力。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    HEV energy management method based on deep reinforcement learning A3C algorithm


    Weitere Titelangaben:

    一种基于深度强化学习A3C算法的HEV能量管理方法


    Beteiligte:
    TANG XIAOLIN (Autor:in) / CHEN JIAXIN (Autor:in) / PU HUAYAN (Autor:in) / ZHANG ZHIQIANG (Autor:in) / YANG XIN (Autor:in) / HU XIAOSONG (Autor:in) / LI JIACHENG (Autor:in)

    Erscheinungsdatum :

    02.10.2020


    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 / G06K Erkennen von Daten , RECOGNITION OF DATA / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Hierarchical energy management method for hybrid electric vehicle based on deep reinforcement learning algorithm

    FU ZHUMU / GONG HUIXIAN / TAO FAZHAN et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Deep reinforcement learning-based antilock braking algorithm

    Mantripragada, V. Krishna Teja / Kumar, R. Krishna | Taylor & Francis Verlag | 2023


    Near space aircraft energy system management method based on deep reinforcement learning

    ZHONG HAO / ZHANG ZHICHENG / ZHANG HONGXI et al. | Europäisches Patentamt | 2024

    Freier Zugriff


    Dual-motor energy management method and system based on deep reinforcement learning

    LIANG ZHAOWEN / WANG ZHENPO / LI TONGYANG et al. | Europäisches Patentamt | 2024

    Freier Zugriff