Adaptive real-time ECMS based on optimal SOC path following for flywheel hybrid electric vehicle

Flywheel hybrid electric vehicle (FHEV) demonstrates significant advantages in energy utilization and recovery. To further improve energy-saving performance and maintain battery state of charge (SOC) stability, an adaptive equivalent consumption minimization strategy (ECMS) integrating driving condition recognition and target battery SOC path tracking, referred to as DA-ECMS is proposed. Firstly, an improved particle swarm optimization based support vector machine model is established to identify driving conditions. Then, the dynamic programming algorithm is used to determine the optimal battery SOC sequence as the target SOC path of ECMS to adjust the equivalent factor in real time, controlling the actual battery SOC path accurately. Finally, the simulation is carried out under CLTC-C and Zibo conditions. The results show that DA-ECMS improves internal combustion engine thermal efficiency and motor/generator efficiency, reduces energy consumption compared to rule-based energy management strategy and traditional ECMS, and stabilizes the SOC. Meanwhile, braking energy recovery increases by 9.74% and 6.18%, with 37.35% recovered directly by the flywheel under CLTC-C condition. Moreover, when the flywheel is unlocked, both fuel and electricity consumption are significantly reduced compared to the locked case.