A Review on Different State of Battery Charge Estimation Techniques and Management Systems for EV Applications

Abstract

Electric vehicles (EVs) have acquired significant popularity in recent decades due to their performance and efficiency. EVs are already largely acknowledged as the most promising solutions to global environmental challenges and CO2 emissions. Li-ion batteries are most frequently employed in EVs due to their various benefits. An effective Battery Management System (BMS) is essential to improve the battery performance, including charging–discharging control, precise monitoring, heat management, battery safety, and protection, and also an accurate estimation of the State of Charge (SOC). The SOC is required to provide the driver with a precise indication of the remaining range. At present, different types of estimation algorithms are available, but they still have several challenges due to their performance degradation, complex electrochemical reactions, and inaccuracy. The estimating techniques, average error, advantages, and disadvantages were examined methodically and independently for this paper. The article presents advanced SOC estimating techniques, such as LSTM, GRU, and CNN-LSMT, and hybrid techniques to estimate the average error of the SOC. A detailed comparison is presented with merits and demerits, which helped the researchers in the implementation of EV applications. This research also identified several factors, challenges, and potential recommendations for an enhanced BMS and efficient estimating approaches for future sustainable EV applications.