Implementation and evaluation of a practical electrochemical- thermal model of lithium-ion batteries for EV battery management system

Abstract

To realize the electrochemical-thermal model in the battery management system (BMS), a control-oriented electrochemical model of lithium-ion batteries and its real-time implementation in the BMS of an electric vehicle (EV) are presented in this paper. Firstly, the improved pseudo-two-dimensional electrochemical model is derived to obtain the transfer function between the cell voltage and current by simplifying the partial differential equations with the Laplace transform, Pade approximation, etc. The temperature-sensitive parameters are identified at different temperatures to obtain the model temperature dependency. Then these temperature-dependent parameters are extracted for model decoupling. The presented model is validated to achieve high accuracy over a wide range of state-of-charges and temperatures with offline data. The root-mean-square (RMS) error in the prediction of terminal voltage could be limited within 28 mV when temperature decreases to −10 °C. Finally, the model is discretized and embedded into the targeted algorithm framework in the BMS of an EV to verify its effectiveness in real-time. The RMS error of the online estimated voltage is less than 15.8 mV under realistic city driving conditions at −13 °C, indicating the feasibility of the developed electrochemical model for broader EV applications.