Reduced Model for Fast Simulation of a Lithium-Ion Battery Pack Taking Into Account Current and State of Charge Dispersion

Abstract

Modeling the electrical behavior of a battery is essential for obtaining information that cannot be directly measured, such as the state of charge (SOC), or that needs a complex instrumentation, such as the individual cell current. Each cell in a battery pack behaves dierently, creating dispersion of current, SOC, voltage and temperature. In this paper, we propose a method to estimate the current and the SOC of each cell in the pack, taking into account dierences in temperature and connection resistances from one cell to another. To reduce computation time, what makes the algorithm more suitable for real-time applications, we simulate only a few cells in detail. For all the other cells, we use a simplied calculation of their resistances based on their temperatures to estimate current and SOC dispersions. Our method reduces the computation time by 60%, while preserving the accuracy of the model.

Field programmable gate arrays implementation of a Kalman filter based state of charge observer of a lithium ion battery pack

“…The battery pack is designed to operate at a total nominal voltage of 18 V, with each individual cell having a nominal voltage of 3.6 V, and an overall capacity of 2.5 Ah. In previous research, the parameters of the battery model were identified as a function of temperature, SOC, and current using the Galvanostatic Intermittent Titration Technique (GITT) [53,54]. Figure 4 For the implementation, the battery's temperature was assumed at 25 °C and that its parameters are constant with temperature, SOC, and current.…”