Dynamical modeling and Identification of Li-ion batteries based on Squirrel Search Optimization Algorithm for Electric Vehicle Applications

Abstract This paper presents an advanced method for modeling and parameter identification of the lithium-ion (Li-ion) battery using an experimental characterization. A comparison study between two Li-ion models and a proposed one are performed where each model parameters are identified using three different algorithms. The obtained models are then tested and validated using experimental data obtained from a real test bench. The identification methods are implemented using a precise nonlinear model based on electric equivalent circuit of Li-ion battery and the parameter identification process formulated as a nonlinear optimization problem. In order to compare the capability of each model to represent effectively the Li-ion battery regardless of the optimization method, three optimization algorithms are involved: Squirrel Search Algorithm (SS), Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). Numerical simulations along with experimental validation are performed on 20 Ah Li-ion battery. The models are compared according to the lowest fitness function levels of each algorithm and their limitation were also discussed in this paper. The obtained results show that the proposed models are able to simulate the dynamic behavior of Li-ion battery with good performances.