Energy management of hybrid storage in an electric bus via optimized dual-level control

This paper proposes a dual-level control strategy for a hybrid energy storage system (HESS) designed for electric vehicles. The HESS integrates batteries and supercapacitors (SCs), where the SCs handle rapid transients while the battery provides long-term energy. The fast-level control distributes power through a low-pass filter with a fixed cutoff frequency, ensuring decoupling of high- and low-frequency load components. Complementing this, a slow-level velocity-based control regulates the charging and discharging of SCs using a PI controller whose voltage reference is dynamically regulated based on the vehicle's speed. This slow-level control discharges the SCs during high-speed operation to assist acceleration, and recharges them at lower speeds to recover regenerative energy. Both the cutoff frequency of the fast-level layer and the settling time of the PI controller in the slow-level layer are optimized using a Genetic Algorithm (GA), achieving up to a 20% reduction in battery degradation compared with LPF- and EWMA-based HESS control strategies, while maximizing system efficiency. The proposed approach improves battery lifetime, enhances energy efficiency, and can be adapted flexibly to different HESS sizing configurations. The methodology is validated using an Energetic Macroscopic Representation (EMR) model of an electric bus and an experimental HESS test platform, and the simulation and experimental results are shown to exhibit consistent dynamic behavior. Finally, the HESS control strategy shows a reduction in stress and battery ripple and offers robust adaptability in real driving situations.