SOC Estimation Using Extended Kalman Filter in Electric Vehicle Batteries

This paper delves into the pivotal role of State of Charge (SOC) estimation in electric vehicle (EV) battery management, employing the Extended Kalman Filter (EKF) to enhance accuracy and reliability using Custom Neural Network(NNET) using MATLAB. Amidst growing environmental concerns and the shift towards renewable energy sources, lithium-ion batteries emerge as essential for their high energy density and longevity. However, precise SOC estimation remains a challenge, influencing both vehicle performance and battery lifespan. Traditional methods like Coulomb counting and voltage-based estimation fall short in dynamic conditions. The EKF, known for its adeptness in handling non-linear systems, offers a promising solution. By comparing predicted and measured voltages, the EKF adjusts internal states, refining SOC estimations. This research presents a comprehensive implementation and evaluation of the EKF in SOC estimation, showcasing its superiority over conventional methods through simulation results through MATLAB using real time data.The real SOC and estimated SOC is compared with ANN for error correction.The findings not only underscore the EKF’s potential in improving battery management systems but also highlight its adaptability to real-world conditions, marking a significant step towards efficient and sustainable EV technologies.