Order Reduction of Lithium-Ion Battery Model Based on Solid State Diffusion Dynamics via Large Scale Systems Theory

A simple and accurate model is essential for the model-based control design. The 1D electrode-averaged battery model maintains the prediction accuracy and depth of physical interpretations. The most critical and complicated dynamics in the system is solid phase concentration diffusion, which has a direct impact on battery status. The finite difference method based on uneven discretization benefits to reduce the complexity. Then, the simplification theory of large scale systems is applied to retain the dominant variable and make the model more compact. The simplification process meets the requirements of fast computation and retention of electrochemical diffusion dynamics. This approach was experimentally verified under 0.2C to 2Cdischarge tests. The simulation output converges to the real terminal voltage within an error of ±0.1V. The simplified model enjoys advantages including high accuracy, low computational cost, and therefore is suitable for deployment and use in real-time embedded battery management systems (BMSs).