The invention provides a hybrid power tracked vehicle energy management method capable of moving across motion dimensions, which adopts a more advanced deep reinforcement learning algorithm SAC in the current energy management technology, and combines with transfer learning on the basis to construct a novel transferable energy management strategy. And good fuel economy and stable state of charge of the battery can be balanced, and the method has a good application prospect. According to the method, an energy management strategy based on deep reinforcement learning is migrated from single straight line driving to three-dimensional motion including straight line driving, steering and climbing, and the defect that the prior art can only be carried out in the same motion dimension generally is overcome. According to the method, the effect of deep migration reinforcement learning can be effectively improved, and a typical and effective reference can be provided for developing energy management capable of realizing cross-motion dimension migration for hybrid vehicles of other vehicle types.
本发明提供了一种可跨运动维度迁移的混合动力履带车辆能量管理方法,其采用了目前能量管理技术中较为先进的深度强化学习算法SAC,并在此基础上和迁移学习相结合来构建新型可迁移能量管理策略,能够实现良好的燃料经济性与稳定的电池荷电状态之间的兼顾平衡,具有良好的应用前景。该方法通过将基于深度强化学习的能量管理策略从单一直线行驶迁移到包含直线行驶、转向和爬坡的三维运动中,克服了现有技术普遍只能在相同运动维度间开展的缺陷。利用经验回放池实现经验样本迁移,本发明能够有效提升深度迁移强化学习的效果,也可为其他车型混合动力车辆开发可跨运动维度迁移的能量管理提供了一个典型有效的参考。
Hybrid power tracked vehicle energy management method capable of moving across motion dimensions
一种可跨运动维度迁移的混合动力履带车辆能量管理方法
28.11.2023
Patent
Elektronische Ressource
Chinesisch
Energy management method and system for series hybrid power unmanned tracked vehicle
Europäisches Patentamt | 2023
|Hybrid power tracked vehicle energy management method and device based on evolutionary algorithm
Europäisches Patentamt | 2024
|