The invention discloses a fuel cell hybrid electric vehicle energy management method based on deep reinforcement learning, and belongs to the field of new energy automobiles. According to the fuel cell hybrid electric vehicle energy management method based on deep reinforcement learning, a target network and an evaluation network are set by using a deep deterministic strategy gradient algorithm (DDPG), a neural network is trained by adopting a deep learning method, and the whole vehicle energy management problem is solved. The method comprises the following steps: acquiring related state information of the fuel cell vehicle, establishing an energy management system model of the fuel cell vehicle, initializing an energy management strategy model of the fuel cell vehicle based on a DDPG algorithm, and designing an energy management strategy of the fuel cell vehicle by using the trained energy management strategy model of the fuel cell hybrid electric vehicle based on the DDPG algorithm. According to the invention, the cost loss of the fuel cell and the lithium battery is reduced, the hydrogen consumption is reduced, the SOC is maintained, and the fuel economy is optimized on the premise of ensuring the performance of the fuel cell hybrid electric vehicle.
一种基于深度强化学习的燃料电池混合动力汽车能量管理方法,属于新能源汽车领域。本发明的目的是使用深度确定性策略梯度算法(DDPG)算法,设置目标网络以及评价网络,并采用深度学习的方法训练神经网络,实现了整车能量管理问题的基于深度强化学习的燃料电池混合动力汽车能量管理方法。本发明步骤:获取燃料电池汽车的相关状态信息,建立燃料电池汽车能量管理系统模型,初始化DDPG算法的燃料电池汽车的能量管理策略模型,利用训练后的DDPG算法的燃料电池混合动力汽车的能量管理策略模型设计燃料电池汽车能量管理策略。本发明降低了燃料电池和锂电池的成本损耗,同时降低了耗氢量,保持SOC,在保证燃料电池混合动力汽车性能的前提下,优化了燃料经济性。
Fuel cell hybrid electric vehicle energy management method based on deep reinforcement learning
基于深度强化学习的燃料电池混合动力汽车能量管理方法
2024-01-02
Patent
Elektronische Ressource
Chinesisch
Europäisches Patentamt | 2022
|Europäisches Patentamt | 2022
|Hybrid electric vehicle energy management method based on reinforcement learning
Europäisches Patentamt | 2021
|Europäisches Patentamt | 2023
|