The invention provides a hybrid power vehicle energy management method and system based on global speed planning and reinforcement learning, and the method comprises the following steps: A, carrying out the equal-time-interval discretization of a whole driving route, discretizing the whole route into equal-time route segments, setting a reference route segment, constructing a target function, and carrying out the calculation of the target function; forming a global speed plan v; b, establishing a series hybrid electric vehicle (SHEV) model; c, adopting an MPC method to establish a vehicle prediction control model; d, setting a speed sequence in a prediction domain at each moment according to a global speed planning curve; and E, adopting the constructed MPC model, adopting a reinforcement learning algorithm in a prediction domain to solve an optimal control sequence, outputting a first optimal control value, realizing rolling optimization, and enabling a vehicle engine to run on an economic curve. According to the invention, the vehicle energy consumption can be locally optimal, the fuel efficiency is improved, and the energy conservation and efficiency improvement of the vehicle are realized.
本发明提供一种基于全局速度规划及强化学习的混合动力车辆能量管理方法及系统,其方法包括以下步骤:A、对行车路线全程进行等时间间距离散,将整个路程离散为等时间的路程片段,同时设定一个参考路程片段,构建目标函数,形成全局速度规划v;B、建立串联混合动力车辆SHEV模型;C、采用MPC方法,建立车辆预测控制模型;D、根据全局速度规划曲线,设定各时刻预测域内的速度序列;E、采用构建的MPC模型,在预测域内采用强化学习算法求解最优控制序列,并将第一个最优控制值输出,实现滚动优化,使车辆发动机运行在经济曲线上。本发明能够使车辆能量消耗局部最优,提高燃油效率,实现车辆节能增效。
Hybrid vehicle energy management method and system based on global speed planning and reinforcement learning
一种基于全局速度规划及强化学习的混合动力车辆能量管理方法及系统
2024-07-23
Patent
Electronic Resource
Chinese
IPC: | G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / 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 |
Hybrid electric vehicle energy management method based on reinforcement learning
European Patent Office | 2021
|Hybrid vehicle energy management method based on reinforcement learning training network model
European Patent Office | 2022
|European Patent Office | 2022
|Fuel cell hybrid electric vehicle energy management method based on deep reinforcement learning
European Patent Office | 2024
|