FPGA Based Improved State-of-Charge Estimation Using Modified Coulomb Counting Method for Lithium-Ion Battery

A key feature for the protection and durability of electric vehicles is the State of Charge (SOC) monitoring. But the SOC can not be estimated directly. Inaccuracies in the battery model and parametric uncertainties are the issues that might result in inaccurate SOC prediction. Therefore, precise parameter estimation yields accurate SOC estimation. The Recursive Least Squares (RLS) estimation method has been employed to estimate the parameters of the battery, which has been used to model equivalent circuit model (ECM). An improved Coulomb counting method is used for SOC estimation of the battery. The Coulomb counting method has been improved by accurately estimating the initial SOC using the open-circuit voltage (OCV)-SOC relationship, and piece-wise linearizing the relationship increases estimation accuracy. The aims to develop and implement the proposed algorithm on a PYNQ-Z2 board for accurate and fast battery SOC estimation. The effectiveness of the proposed model in precisely estimating the SOC of lithium-ion batteries has been verified through simulations and real-time testing.