An analysis and evaluation of state estimation methods for lithium-ion batteries regarding requirements of 48 V HEV applications

To cope with increasingly stricter CO2 emission regulations and to improve comfort features for their customers, the automotive industry introduced 48V systems in a broad range of their new vehicles powered by a conventional internal combustion engine (ICE). With the addition of a 48V lithium-ion battery, it becomes crucial to accurately estimate its State of Charge (SoC) and State of Health (SoH) to ensure a safe operating window, high performance and a long lifetime. Several research studies examined a variety of methods to estimate the SoC and SoH. However, self-sufficient 48V systems challenge state-of-the-art battery state estimation methods due to high C-rates and high SoC variation. Only few methods meet the automotive requirements for real-time applicability on embedded systems and online estimation. This study analyses typical battery usage patterns in 48V systems and compares it to the ones of battery electric vehicles (BEVs). Various state estimation methods presented in literature including direct measurement, model-based, and data-driven methods are discussed to determine whether the implementation and validation of the proposed method fit the 48V application scenario. This study shall form the basis for consecutive papers which focus on the implementation and validation of promising methods such as Kalman filters under various operating conditions and real driving data of 48V hybrid electric vehicles (HEVs). •A comparison of BEV and 48V HEV systems based on real field data is carried out.•The special requirements of 48V HEV systems are analysed and characterised.•An overview of battery state estimation methods targeting HEV and BEV application.•Literature for state estimation methods is assessed regarding suitability to 48V HEV.