Analysis and advancements of the state of charge estimation methods in smart battery management system supported by lithium-ion battery operated electric vehicles

Abstract

This study presents the comprehensive analysis and investigation of State of Charge (SOC) estimation methods used for Lithium-Ion batteries in electric vehicles. The lithium-ion batteries are elegantly used in the electric vehicle due to their advantages like light in weight, economic, higher specific and energy density, long service life, but they must be operated in the safe operating area. Since it is the most precious part of electric vehicle it needs continuous observations, this task is completed by the Battery Management System (BMS) in Electric Vehicles. SOC of lithium-Ion battery plays a crucial role in the smart BMS for the safe and reliable operation of batteries, hence the advancement in the fast-growing battery technology needs an efficient BMS, also SOH, SOF, SOAP, RUL, Charging/discharging as well as almost all the safety operational parameters are largely depends on SOC, which leads to enhance robustness, accuracy and nonlinear handling of SOC estimation methods. This study incorporates the thorough discussion on the estimation methods utilized in realizing the SOC, generally these methods are grouped as conventional or book keeping methods, equivalent model-based methods, filtering algorithm methods, data driven methods and hybrid methods. As its well defined that every technique has its pros and cons with challenges are thoroughly described in this study. The primary objective of this study to give up to date summary of all latest advancement in SOC estimation methods, offering insights to innovative approaches and emphasis on developing new learning algorithms and hybrid methods to enhance accuracy and adaptability. This study systematically presents analysis of the advancement of the SOC estimation methods the study mainly focuses on enhancement of estimation accuracy, robustness and reliability of the methods used in realizing the SOC with the changing temperature and nonlinear behavior of the LIBs. In real time the accuracy and the nonlinear behavior of energy storage systems becomes the limitations at the extreme temperature. But, in a contrast with that the data driven techniques are having the ability to cope up with this behavior of temperature fluctuation and with different operating conditions from the historical data because these methods basically extract the nonlinear relationships form their history. In addition, with this the hybrid approach of integrating AI methods with the available SOC estimation methods gives the fruitful result with lightweight algorithm nature with improved accuracy and robustness this enables the hybrid approach making it suitable for deployment in real-world for easy implanting in BMS to estimating SOC where embedded constraints are significant. This study also suggests that future research should focused on furnishing the existing algorithms, investigation of new learning algorithm methods and integrating the latest and advance sensor technologies to realize real time, accurate and reliable SOC estimation for Electric Vehicles.