The invention relates to a hydrogen fuel unmanned aerial vehicle energy management method based on deep reinforcement learning, and the method comprises the steps that a speed model, an air density and mass loss model and a propulsion load model of an HEUA are built according to different flight profiles of an aircraft, and the load power in the flight state is obtained; then, an ECMS energy control model based on reinforcement learning is established, power distribution of the lithium battery and the hydrogen fuel battery in the state is obtained through load power, a power distribution curve under the flight profile is obtained, the variation of the SOC value is accurately calculated, and a descent curve of the power distribution curve is obtained; and autonomous management of energy in the flight process of the unmanned aerial vehicle is realized.
本发明涉及一种基于深度强化学习的氢燃料无人机能量管理方法,根据飞机的不同飞行剖面,建立HEUA的速度模型、空气密度和质量损失模型和推进载荷模型,得到在此飞行状态下的负载功率。接着建立基于强化学习的ECMS能量控制模型,由负载功率得到此状态下锂电池和氢燃料电池的功率分配,得到飞行剖面下的功率分配曲线并精确计算SOC值的变化量,得到其下降曲线。实现对无人机飞行过程中能量的自主管理。
Hydrogen fuel unmanned aerial vehicle energy management method based on deep reinforcement learning
一种基于深度强化学习的氢燃料无人机能量管理方法
2021-08-03
Patent
Elektronische Ressource
Chinesisch
IPC: | B64D Ausrüstung für Flugzeuge , EQUIPMENT FOR FITTING IN OR TO AIRCRAFT / B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Unmanned aerial vehicle hydrogen fuel cell system, control method and unmanned aerial vehicle
Europäisches Patentamt | 2023
|Europäisches Patentamt | 2024
|