Parameter Estimation of an Electrochemical Li-ion Battery Model Using Innovative Global Harmony Search

The electrochemical model is one of the important model of battery, since it represents the real dynamics in the cell. The parameters of the model have physical meanings, so estimating the parameters, especially aging parameters might be helpful to efficient cell operation. There are many parameter estimation algorithms now on. Among the algorithms, harmony search is better than other algorithms like genetic algorithm, particle swarm optimization or gradient-descent methods, since harmony search has fast process characteristic which is suitable to electrochemical model. The 5 algorithms are applied to same parameter estimating problem to compare their performance in the aspect of objective values and parameter errors. The results show that innovative global harmony search is effective to estimate the better solutions with small parameter errors.