Enhancement of Charging Efficiency of Batteries for Electric Vehicles: Review

Abstract

To minimize emissions of greenhouse gases, the electric vehicle industry has grown dramatically in recent years. Traditional vehicles are improved by a variety of technologies, including regenerative braking, auto-start and shut-off, and even more advanced technology. Affordable charging stations with state-of-the-art control algorithms are required to boost the charging efficiency of electric vehicles (EVs). An electronic regulator known as a “Battery Management System” (BMS) controls battery monitoring and protection, calculation of battery status, and potential user notification of that condition. Due to many faults in battery packs, EVs encounter a number of issues; it is essential to assess the battery's level of charge (SOC). Measurement of battery performance and health are part of battery cell monitoring. The ideas of artificial intelligence (AI) and machine learning (ML) for battery management in new energy vehicles were initially presented to address the issue of protecting battery cells from over-current and over-voltage. Data is collected by applying multiline signals to one-of-a-kind dynamic charge and discharge test. Machine learning (ML) algorithms are helpful in state of charge (SOC) prediction challenges because they can record cell dynamics and store past data, both of which are necessary for predicting future charge levels. Along with different charging methods, energy storage technologies, BMS topologies, multiple ML algorithms and Battery Management System (BMS) used in an electric car arealso explored in this review.