Development and Analysis of Equivalent Circuit Models and Effect of Battery Parameter Variations on State of Charge Estimation Algorithm

Lithium-Ion batteries are popular for use in Electric vehicle (EV) applications. To improve and understand the use of Lithium-Ion batteries (LIBs) in EV application, present study focused and utilized equivalent circuit models (ECMs). Model parameters are identified using pulse charge and discharge test carried on 20Ah Lithium Iron Phosphate cell. Curve-fitting technique is utilized and detailed procedure to extract model parameters is presented. Models are validated with experimental data of pulse discharge test. Accuracy obtained using 1-RC, 2-RC, 3-RC circuit models is verified and high accuracy of 3RC circuit model can make it act as a battery emulator. Extended Kalman Filter (EKF) is utilized for estimation of State of Charge (SOC) of Lithium Iron Phosphate cell. As per our observation, a good accuracy with low computational burden can be achieved with 1RC model parameters. Therefore, accuracy of SOC estimation using 1RC model parameters is analyzed and effect of initial error in SOC is also observed. Moreover, battery model parameters are function of Temperature, C-rate as well as Ageing. As battery degrades, battery model parameters change especially internal resistance (IR) and capacity degradation happens. IR plays a vital role in estimation of State of Health (SOH) of battery in model based estimation methods. However, carrying out ageing study is difficult and very much time consuming process. So, an attempt is made to develop a framework which integrates 3-RC cell model with EKF based SOC estimator. The framework thus enable us to study effect of IR and battery ageing on estimation of EKF based SOC estimator. The limitations and future scope of work is described in summary section. The paper will help Industrial R&Ds and academic researchers to widen the scope of implementation of EKF.