在燃料电池混合动力汽车的实际运行中，空调系统为驾驶员和乘客提供舒适的环境，然而空调系统的运行效果与汽车实际运行的能量分配相互影响，因此需要将空调系统考虑进能量管理策略，设计出在满足舱内温度舒适性要求的情况下，兼顾整车氢耗经济性的能量管理策略。首先在建立整车动力学模型的基础上，利用热平衡方程建立热泵空调系统模型和热负荷模型；然后采用结合了双Q网络和深度确定性策略梯度的优先经验采样的双延迟深度确定性策略梯度（TD3-PER）算法，建立考虑空调系统能耗与车辆运行需求的能量管理策略。在NEDC典型工况下进行仿真得出：TD3-PER能量管理策略下的空调系统能够使舱温在100 s内迅速达到并维持在22~26 ℃的舒适范围内，满足制冷/制热的同时又保证车舱温度舒适，验证了考虑空调系统时TD3-PER能量管理策略的可行性；在空调系统制冷/制热时，相比传统的深度确定性策略梯度（DDPG）算法策略，基于TD3-PER算法策略的功率分配情况能够延长燃料电池和蓄电池使用寿命，且在制冷/制热时根据氢耗量分别可提高2.59和3.58个百分点的经济性，验证了基于TD3-PER算法能量管理策略在降低氢耗量、提高整车经济性方面相较于传统算法更具优势。

In the actual operation of fuel cell hybrid electric vehicles, the air conditioning system provides a comfortable environment for drivers and passengers. However, the performance of the air conditioning system interacts with the vehicle’s energy distribution during operation. Therefore, it is necessary to integrate the air conditioning system into the energy management strategy, and design an energy management strategy that ensures the cabin temperature comfort requirements while also considering the overall hydrogen consumption efficiency of the vehicle. Firstly, based on the vehicle dynamics model, the heat balance equation was used to establish the heat pump air-conditioning system model and heat load model. Then, the dual delay depth deterministic strategy gradient (TD3-PER) algorithm combining the double Q network and the depth deterministic strategy gradient was used to establish the energy management strategy considering the energy consumption of the air conditioning system and the vehicle operation demand. Simulation under the typical NEDC driving cycle shows that with the TD3-PER energy management strategy, the air conditioning system can rapidly bring the cabin temperature to and maintain it within the comfortable range of 22 ℃ to 26 ℃ in 100 seconds, ensuring cooling/heating performance to maintain cabin comfort. This validates the feasibility of the TD3-PER energy management strategy when considering the air conditioning system. During cooling/heating operation, compared to the traditional Deep Deterministic Policy Gradient (DDPG) algorithm, the power distribution strategy based on the TD3-PER algorithm can extend the lifespan of both the fuel cell and the battery. Additionally, in terms of hydrogen consumption, the TD3-PER-based strategy can improve fuel economy by 2.59 percentage points during cooling and 3.58 percentage points during heating. This demonstrates that the TD3-PER algorithm-based energy management strategy offers significant advantages over traditional algorithms in terms of reducing hydrogen consumption and improving overall vehicle efficiency.