Lithium-ion batteries (LiBs) are increasingly being used in EV/HEV applications due to their virtue of high energy density, power density, low weight and long life. Operation of Li-Ion batteries beyond their safe zone may lead to hazardous thermal runaway events. It is difficult to estimate the remaining usable capacity left in the battery during operation. Thus, accurate state of charge (SOC) and state of available power estimation is crucial to ensure the safe and reliable operation of lithium-ion batteries used in EV/HEVs. In this paper, a model based framework for real time estimation of battery state of charge and available power estimation is discussed. A detailed equivalent circuit model capturing battery dynamics is developed and parameterized for a battery cell. The model achieves >98% accuracy in full SOC window for dynamic profiles Adaptive Extended Kalman Filter based SOC estimation algorithm is proposed and validation for drive cycle data. Tuning of the Kalman Filter for improving state estimation is discussed. The Kalman filter based approach is compared with coulomb counting. Finally, a multi constrained algorithm for real time estimation of available power in the battery is developed and integrated with the SOC estimation framework. Later, a comprehensive evaluation of the models is carried out for model complexity and accuracy and the results are compared against the experimental data. The results demonstrate that model based SOC and available power estimation provide a significant advantage over traditional coulomb counting based methods and actionable insights for optimal control of EV batteries.
A Model based Framework for Online State of Charge and Available Power Estimation in Li-ion Batteries
Sae Technical Papers
WCX SAE World Congress Experience ; 2023
2023-04-11
Aufsatz (Konferenz)
Englisch
A practical framework of electrical based online state-of-charge estimation of lithium ion batteries
Online Contents | 2014
|