Review on technological advancement of lithium-ion battery states estimation methods for electric vehicle applications

Due to the dynamic and non-linear behavior of lithium-ion battery (LIB) states, the accuracy of state estimation proportionally impacts the performance of the battery management system (BMS) as well as the life cycle of LIB. Generally, four different battery states including the state of charge (SOC), state of energy (SOE), state of power (Power), and state of health (SOH) have been utilized to control and optimize the performance LIB used in electric vehicles (EV). Along with the SOH, the remaining useful life (RUL) is important to control the LIB performance and life. With technological advancement, there are several advanced battery state estimation algorithms have been developed for individual and combined states estimation methods. All the existing state estimation algorithms have their pros and cons. Therefore, there is a need of the state of art review and analyze the performance of existing advanced state estimation algorithms. In this paper, the existing individual, and combined states estimation algorithms suitable for SOC, SOE, SOP, and SOH are explored. Moreover, the mathematical formulas involved in state estimation are illustrated. Based on the critical findings from the literature review, a new combined states estimation method for SOC, SOE, SOH, and SOP is proposed to achieve a higher estimation accuracy and lower computational burden. The performance of the proposed combined states estimation algorithm is validated using a dynamic load profile under a wide range of operating temperature conditions. The experimental results show that the estimated SOC and SOE error is <2.5 % irrespective of the change in operating conditions. Further, the proposed method is capable of accurately estimating actual capacity and (dis)charge SOP, simultaneously. The estimated capacity converges to actual values within the first few seconds under considered operating conditions. Finally, the ongoing research comprising of advanced states estimation approaches are distinctly emphasized through reviewing various studies for future research.