Real-time adaptive hybrid energy management for hybrid-electric UAVs: Offline–online DP-ECMS collaborative strategy in low-altitude inspection

This paper proposes an energy management strategy that integrates dynamic programming (DP) offline global optimization and equivalent consumption minimization strategy (ECMS) online dynamic optimization. First, based on a given route of the UAV, using the predicted flight data of the power system, the demanded power is input into the DP algorithm, and a multi-objective cost function considering the efficiency and loss is established to realize the global offline energy allocation strategy optimization and derive the reference value of the SOC of the lithium battery. Second, the ECMS uses the DP result as the initial reference to design the equivalent factor based on the reference SOC. Third, the ECMS collects flight information in real time and corrects the power output online when the actual demand power deviates from the predicted power; when the SOC of the lithium battery is close to the lower limit due to a strong sudden perturbation, the weight of lithium battery protection is increased through the optimization cost function. Finally, the hardware-in-the-loop simulation platform is constructed to verify the effectiveness of the proposed algorithm. Compared with the traditional ECMS, this strategy reduces hydrogen consumption by 7.2% and improves overall system efficiency by 5.7%, thereby demonstrating its superiority in optimizing energy utilization and ensuring safe system operation. • For fixed-route inspection, we propose an energy management framework of “offline global optimization+online tracking&correction”. • ECMS dynamically corrects offline DP results, breaking traditional single-strategy limitations. • Offline, a new efficiency cost function balances battery life and system efficiency; a dynamic adaptive equivalent factor adjusts offline energy allocation trajectory in real time via actual data to maintain optimality.