The invention relates to a vehicle-station-network interactive decision-making method based on multi-agent deep reinforcement learning. The method comprises the following steps: collecting data of a charging station and data of a user new energy vehicle; constructing a distributed partially observable Markov decision model, and setting the sum of penalty caused by user energy consumption cost and charging station peak-valley difference and penalty caused by unsatisfied constraint conditions in the distributed partially observable Markov decision model as a target function; and performing centralized training and distributed execution by using a distributed partially observable Markov decision model constructed by a multi-agent reinforcement learning BiCNet algorithm to obtain an actual decision method. According to the invention, the matching of electric vehicle charging and renewable energy sources can be effectively realized, the load power consumption peak is reduced to the maximum extent, and the load peak-valley difference of the charging station is reduced.

    本发明涉及基于多智能体深度强化学习的车‑站‑网互动决策方法,通过采集充电站数据以及用户新能源汽车的数据;构建分布式部分可观测马尔科夫决策模型,并在分布式部分可观测马尔科夫决策模型中设置以用户用能成本、充电站峰谷差造成的惩罚和约束条件不满足造成的惩罚之和作为目标函数;使用多智能体强化学习BiCNet算法构建的分布式部分可观测马尔科夫决策模型进行集中训练和分布式执行,得到实际决策方法。本发明能够有效地使实现电动汽车充电与可再生能源的匹配,最大化的减少了负荷用电高峰,降低了充电站负荷峰谷差。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Vehicle-station-network interaction decision-making method based on multi-agent deep reinforcement learning


    Weitere Titelangaben:

    基于多智能体深度强化学习的车-站-网互动决策方法


    Beteiligte:
    ZHANG JIAN (Autor:in) / XI YANJUN (Autor:in) / WANG QIANG (Autor:in) / JI YANG (Autor:in) / ZHANG XUZE (Autor:in) / MENG FANJIE (Autor:in) / SHI HANG (Autor:in) / YANG JINGLU (Autor:in)

    Erscheinungsdatum :

    2023-03-31


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / H02J CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER , Schaltungsanordnungen oder Systeme für die Abgabe oder Verteilung elektrischer Leistung



    Intersection decision-making method based on multi-agent deep reinforcement learning

    DU YU / JIANG ANNI / ZHAO SHIXIN et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Deep reinforcement learning-based ethical-driven multi-modal decision-making method

    LI XUEYUAN / GAO XIN / LUAN TIAN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Automatic driving vehicle overtaking decision-making method based on deep reinforcement learning

    LIU JIA / LI HUIYUN / CUI YUNDUAN | Europäisches Patentamt | 2022

    Freier Zugriff


    Automatic driving decision-making method based on deep reinforcement learning

    LIU CHENGQI / LIU SHAOWEIHUA / ZHANG YUJIE et al. | Europäisches Patentamt | 2024

    Freier Zugriff