Dynamical modeling of Li-ion batteries for electric vehicle applications based on hybrid Particle Swarm–Nelder–Mead (PSO–NM) optimization algorithm

Abstract In recent years, Li-ion batteries are widely used in various applications, such as electric and hybrid vehicles application. Their higher specific power and energy density, high cycle lifetime and decreasing costs have made them an attractive and alternative energy storage technology to lead-acid or nickel-metal hydride batteries in embedded power supplies. In the present work, the electric modeling of a Li-ion battery cell in real operation conditions imposed by an electric vehicle application is carried out. A dynamic equivalent circuit model has been used to simulate several electrochemical processes occurring in a commercially available 40 Ah Li-ion battery cell with NMC cathode material and graphitic anode. The model is parameterized with measurement data in time-domain using a hybrid Particle Swarm–Nelder–Mead (PSO–NM) optimization algorithm. This last one is used to solve the parameters identification problem of Li-ion battery model. The developed model of Li-ion battery cell has been validated on real driving cycle provided by an urban electric vehicle. Obtained results show that there is a good match between experiment and simulation results with a maximum modeling error less than 0.5%, which proves the well performance of our model and confirms the interest of a hybrid (PSO–NM) optimization algorithm for battery identification parameters.