Energy Management Strategy for Hybrid Electric Vehicles Based on Reinforcement Learning

For the purpose of enhance the fuel consumption rate together with stabilize the control strategies concerning hybrid electric vehicles. This work introduces a layered energy management strategy combining the Equivalent Consumption Minimization Strategy (ECMS) and Deep Reinforcement Learning (DRL). Experimentation outcomes confirm which this layered control approach in addition to allows the intelligent agent to achieve model-free adjustable consumption-reducing control along with ensures which the hybrid vehicle's State of Charge (SOC) meets the constraints under all operating conditions. This hierarchical control strategy can greatly enhance fuel efficiency compared to directive-based energy management methods and non-hierarchical deep reinforcement learning strategies.