Estimation of li-ion battery state of charging and state of healthy based on unsented Kalman filtering

In order to obtain the accurate estimation of SOC (State of Charge) and predicted lithium battery SOH (State of Healthy), this article is based upon the internal resistance of the battery model. By using UKF method, the estimation of SOC and SOH can be carried out in the nonlinear conditions. The UKF algorithm considers the internal resistance of the model parameters and SOC as the state parameters. Depending on UKF, the SOC will be estimated and the resistance will be constantly adjusted to compensate for model inaccuracies. Due to the internal resistance has the relation with the state, thus the SOH indirectly would be estimated. The final simulations and the result of the experiments show that unscented Kalman filter can make the accuracy of SOC estimation within 4% while achieving an accurate prediction of the SOH.