Universal Adaptive Stabilizer Based Optimization for Li-Ion Battery Model Parameters Estimation: An Experimental Study

This paper proposes a two-stage universal adaptive stabilizer (UAS)-based optimization technique for an accurate and efficient estimation of Li-ion battery model parameters. The first stage utilizes an UAS-based adaptive parameters estimation (APE) technique to acquire an initial estimate of battery model parameters. The second stage utilizes one of the three different optimization techniques, i.e.,
fmincon
, particle swarm optimization (PSO), and hybrid PSO to improve the accuracy of battery model parameters obtained by the APE. The parameters estimated by the APE help in reducing the search space interval required by the optimization technique, thus reducing the computation time of the optimization process. Intensive computer simulation and experimentation are performed to estimate the battery terminal voltage using the estimated parameters. The accuracy of estimated battery parameters is evaluated by comparing the estimated and measured battery terminal voltage. The results show that the accuracy of the battery model parameters obtained by the optimization techniques alone is poor, and the required computation time is high. The accuracy of parameters obtained by UAS-based APE is good with very low computation time, while it is best when UAS-based APE is used in combination with the PSO, or hybrid PSO optimization techniques while requiring an intermediate amount of computation time.