Automatic System Identification of Equivalent Circuit Parameters for Lithium Ion Batteries

Abstract

Lithium-ion batteries are a critical component of electric vehicles (EVs), directly influencing their performance, range, and longevity. Battery management algorithms often rely on equivalent circuit models (ECMs) to describe the voltage response of cells to applied currents. However, accurately identifying ECM parameters is challenging due to their dependency on factors such as temperature, state of charge (SOC), and hysteresis. These challenges are further compounded by the timeconsuming nature of the parameter identification process.

This thesis presents a novel system for identifying ECM parameters by executing realistic drive cycle current profiles on lithium-ion cells. The method accounts for temperature effects, SOC variations, and the hysteresis effect to improve modeling accuracy. The identified parameters were validated and stored in lookup tables for integration into battery management systems (BMS). The results demonstrated consistent parameter identification across various SOCs and temperatures, with a notable enhancement in ECM accuracy. Despite some limitations, such as challenges in parameter estimation at low temperatures, this work provides a robust foundation for more accurate and efficient BMS algorithms, particularly in automotive applications.