Electrochemical model of a lithium-ion battery implemented into an automotive battery management system

This paper presents the development of an electrochemical model that can be implemented into automotive battery management systems (BMSs). Compared with empirical models, the electrochemical model features more accurate state estimates over a broader and longer use of the battery. In this work, model implementation schemes are devised to make the electrochemical model uncomplicated enough to be embedded into the BMS. A nonlinear system of partial differential equations in the model is discretized into a linearized system of algebraic equations (AEs). A solver selected to evaluate the resulting system of AEs is modified for its application to the BMS. As the BMS is preoccupied by its existing tasks, the reformulated equations and optimized solver are reorganized such that the limited computational resources of the BMS are appropriately exploited. The electrochemical model is consequently implemented into the BMS, predicting battery behaviors in 1 s intervals while occupying a 14 kB RAM.