The invention provides a battery equalization method based on reinforcement learning, and the method comprises the steps: building a physical field model of a battery control board as a simulation environment of battery equalization, and the physical field model comprises to-be-identified system parameters; acquiring experimental data of the battery control board, and identifying and determining system parameters in the physical field model based on the experimental data; and based on a reinforcement learning algorithm, combining a greedy rule and a preset constraint condition, training a neural network through interaction with a simulation environment, and realizing battery equalization control through an equalization control signal output by the neural network.
本发明提供了一种基于强化学习的电池均衡方法,包括:建立电池控制板的物理场模型以作为电池均衡的仿真环境,物理场模型中包括待辨识的系统参数;采集电池控制板的实验数据,基于实验数据辨识确定物理场模型中的系统参数;以及基于强化学习算法,结合贪婪规则与预设约束条件通过与仿真环境的交互训练神经网络,通过神经网络输出的均衡控制信号实现电池均衡控制。
Battery equalization method and device based on reinforcement learning
一种基于强化学习的电池均衡方法及装置
2023-09-19
Patent
Elektronische Ressource
Chinesisch
IPC: | B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen |
Battery pack equalization method based on reinforcement learning
Europäisches Patentamt | 2023
|Battery equalization method and battery equalization device
Europäisches Patentamt | 2020
|Battery equalization method and battery equalization system
Europäisches Patentamt | 2023
|Equalization method and equalization device for battery module
Europäisches Patentamt | 2021
|