Energy Management on Battery/Ultracapacitor Hybrid Energy Storage System based on Adjustable Bandwidth Filter and Sliding-mode Control

• Published in: Journal of energy storage Vol. 30; p. 101569
• Main Authors: Asensio, E. Maximiliano, Magallán, Guillermo A., De Angelo, Cristian H., Serra, Federico M.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2020
• Subjects: batteries; current prediction; dc/dc converter; Electric vehicles; sliding mode control; ultracapacitor; Electric vehicles; dc/dc converter; batteries; sliding mode control; ultracapacitor; current prediction
• ISSN: 2352-152X; 2352-1538
• DOI: 10.1016/j.est.2020.101569

•Real-time power-split strategies are needed for Hybrid Energy Storage System in EV•Effective high-level power split is achieved with an adjustable-bandwidth filter•Precise low-level sliding-mode control ensures stability under constant power loads•Current prediction and sliding mode control maintains the current ripple bounded A real-time power-split control strategy for a hybrid energy storage system (HESS) used in electric vehicles is proposed in this work. The HESS topology corresponds to a semi-active ultracapacitor (UC) configuration. The HESS goals are to prevent battery degradation and to preserve its lifetime while improving the system efficiency by supplying the fast dynamics power demands through the UC pack. In order to generate the UC power reference, a digital low-pass filter whose bandwidth is adjusted according to the UC SOC is proposed. This allows a better usage of the UC available energy, while reducing frequent activation of controller protections and avoiding the calculation of the filter cut-off frequency for a certain driving cycle. The low-level control strategy is based on a sliding mode controller combined with a closed-loop current observer, which allows to improve the implementation performance while maintaining the current ripple bounded. Simulation and experimental results were presented showing that the proposed strategy preserves battery health in a more effective way than filter-based strategies with fixed bandwidth, for cases where no future information about the required power is available.