Enhanced Lithium-Ion Battery Model for estimation of Degraded Capacity and SoC Using Sigma Point Kalman Filter

Abstract

For an effective battery management system (BMS), accurate estimation of the state of charge (SoC) is essential, which signifies the residual charge in the battery. In addition, SoC estimation relies on aspects such as appropriate battery modelling, battery age, ambient temperature, and many unknown parameters. Thus, the research focuses on developing an accurate battery cell model which includes these non-linearities and ageing effects. Existing mathematical models emphasize reflecting non-linearities such as diffusion and hysteresis effect, but they fail to incorporate the capacity fading effect model. Since the total capacity of the battery degrades concerning ageing. Including the capacity fading model in the battery cell model is critical. This work is on developing a mathematical model for the capacity fading effect. The capacity degradation model has been developed based on the temperature rate dependency and the number of cycles utilized for SoC estimation. The proposed model has been employed and given as input for the state estimation technique to obtain accurate SoC. Capacity loss for the sample battery cell is modelled up to 1000 cycles.Further, the effectiveness of the proposed model is validated and simulated using the SPKF algorithm in MATLAB/Octave environment. Throughout the evaluation procedure, SPKF achieved an estimation error of less than 1%. The proposed capacity fading model and estimation approach based on SPKF may thus provide high robustness and accurate SoC estimation.