Real-Time SOC Estimation of Li-ion Battery Pack Using Improved Kalman Filter

• Published in: 2024 Third International Conference on Power, Control and Computing Technologies (ICPC2T) pp. 151 - 156
• Main Authors: Singh, Rishita, Das, Sourabh, Samanta, Susovon
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.01.2024
• Subjects: Arduino; Batteries; Estimation; Integrated circuit modeling; Kalman filter; Kalman filters; Mathematical models; Real-time systems; RLS with forgetting factor; State of charge; Thevenin's model
• DOI: 10.1109/ICPC2T60072.2024.10474776

A battery Management System (BMS) is an electronic system that enables real-time monitoring and controlling of the battery pack voltage and current for safe and efficient operation. State of charge (SOC) is the crucial factor for BMS. For accurate battery SOC estimation, battery pack modeling is required. In this work, the battery pack is modeled by Electrical Equivalent Circuit Model (1-RC model). The parameters of battery pack are estimated by Fixed Forgetting Factor Recursive Least Squares (RLS-FF) based on terminal voltage and load current data. There are number of methods that can be used for the estimation of SOC. Here, Kalman filter is used for SOC estimation. However, the drawback of Kalman filter is that it cannot estimate the initial SOC of the battery pack. This draw-back can be improved by using the relationship between SOC and Open circuit voltage (OCV) which has been implemented in this paper. The experiment is implemented in Arduino Mega 2560 microcontroller to make it cost-efficient, and the results are validated using Kalman filter block in MATLAB/Simulink.