Contribution to the modeling of Li-ion batteries

Abstract

A battery model that predicts the electrical performance with high accuracy is vital for circuit designing to control the battery usage and improve its runtime and safety in electrical systems such as electric vehicle (EV). In this Thesis, a relatively simple equivalent circuit based model is developed for modeling the performance of lithium ion batteries used in automotive applications. The model's strateg y adopted for characterization of the internal parameters is based on two main techniques: one is the electrochemical impedance spectroscopy (EIS) tests, where comparable circuit is set up as per EIS test perceptions and the other is the current pulses technique, this latter is carried out using a proposed current pulses profile. The extracted parameters are scheduled on the state-of-charge, temperature, and current direction. The comparison between experiment and simulation results demonstrates that with the optimally extracted performance parameter s, the proposed approach can accurately predict the performance of differents batteries of various sizes, capacities, and materials. Eventually, this simple battery model can serve as a robust and reliable tool for predicting the battery's I-V performance in electrified vehicle battery management systems.