The invention relates to a reinforcement learning-based battery pack equalization method. The method comprises the steps of determining an equalization target and constraint conditions of a battery pack equalization process according to rated capacity of single batteries in a battery pack and equalization topological parameters in an equalization system; establishing an action space of an equalization system intelligent agent according to the equalization current control quantity of the battery pack equalizer, and establishing a state space of the equalization system intelligent agent according to the inconsistent state information of the battery pack and the equalization current control quantity generated by the intelligent agent under the state information; establishing a deep learning network of an Actor-Critic architecture, and constructing a deep reinforcement learning equalization strategy based on a double-delay depth deterministic strategy gradient algorithm; designing a reward function of a battery equalization system, training a deep reinforcement learning equalization strategy, and randomly initializing the SOC state of a single battery in each training round; and performing battery pack equalization control by using the trained reinforcement learning equalization strategy. The method is beneficial to shortening the battery pack equalization time and reducing the energy waste in the battery pack equalization process.

    本发明涉及一种基于强化学习的电池组均衡方法,包括:根据电池组中单体电池的额定容量及均衡系统中均衡拓扑参数确定电池组均衡过程的均衡目标和约束条件;以电池组均衡器的均衡电流控制量建立均衡系统智能体的动作空间,以电池组的不一致性状态信息和该状态信息下智能体产生的均衡电流控制量建立均衡系统智能体的状态空间;建立Actor‑Critic架构的深度学习网络,并构建基于双延迟深度确定性策略梯度算法的深度强化学习均衡策略;设计电池均衡系统奖励函数,训练深度强化学习均衡策略,并在每个训练回合随机初始化单体电池的SOC状态;利用训练好的强化学习均衡策略进行电池组均衡控制。该方法有利于缩短电池组均衡时间,减少电池组均衡过程中的能量浪费。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Battery pack equalization method based on reinforcement learning


    Weitere Titelangaben:

    一种基于强化学习的电池组均衡方法


    Beteiligte:
    WANG YAXIONG (Autor:in) / YANG QINGWEI (Autor:in) / LIANG FEIFAN (Autor:in) / OU KAI (Autor:in)

    Erscheinungsdatum :

    2023-09-01


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    BATTERY CELL EQUALIZATION METHOD FOR BATTERY PACK

    LI QIANG / TANG YINXIA / HOU SEN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Battery system and battery pack equalization method

    YOSHIDA HIROSHI / NAKAGAWA TAKAHIKO / UWAI KENTA | Europäisches Patentamt | 2022

    Freier Zugriff

    Battery equalization method and device based on reinforcement learning

    YANG ZHIFEI / KE XICHUN / LIU CHUANG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Battery pack equalization method and device

    LI YIPING / WU JUN / PENG SHIMING | Europäisches Patentamt | 2020

    Freier Zugriff

    Composite battery pack equalization circuit

    XUAN DONGJI / WANG BIAO / CHENG TAIHONG et al. | Europäisches Patentamt | 2020

    Freier Zugriff