Dimensioning and Power Management of Hybrid Energy Storage Systems for Electric Vehicles With Multiple Optimization Criteria
Huilong Yu,Francesco Castelli-Dezza,Federico Cheli,Xiaolin Tang,Xiaosong Hu,Xianke Lin
DOI: https://doi.org/10.1109/tpel.2020.3030822
IF: 5.967
2021-05-01
IEEE Transactions on Power Electronics
Abstract:Hybrid energy storage systems (HESS) that combine lithium-ion batteries and supercapacitors are considered as an attractive solution to overcome the drawbacks of battery-only energy storage systems, such as high cost, low power density, and short cycle life, which hinder the popularity of electric vehicles. A properly sized HESS and an implementable real-time power management system are of great importance to achieve satisfactory driving mileage and battery cycle life. However, dimensioning and power management problems are quite complicated and challenging in practice. To address these challenges, this article proposes a bilevel multiobjective design and control framework with the nondominated sorting genetic algorithm NSGA-II and fuzzy logic control (FLC) as key components, to obtain an optimal sized HESS and the corresponding optimal real-time power management system based on FLC simultaneously. In particular, a vectorized fuzzy inference system is devised, which allows large-scale fuzzy logic controllers to run in parallel, thereby improving optimization efficiency. Pareto optimal results of different HESSs incorporating both optimal design and control parameters are obtained efficiently thanks to the vectorization. An example solution chosen from the Pareto front shows that the proposed method can achieve a competitive number of covered laps while improving the battery cycle life significantly.
engineering, electrical & electronic