Order reduction electrochemical mechanism model of lithium-ion battery based on variable parameters

Abstract

Electrochemical model is of great significance for the research of state estimation and aging of lithium-ion batteries (LIBs). However, the mechanism model has some problems, such as difficult parameter acquisition, time-consuming online identification and poor adaptability, which limit the accuracy of the model in practical application. In this paper, a novel order reduction mechanism model of LIB based on variable parameters is proposed to overcome the above problems. First, a spectral method is used to reduce the partial differential equation (PDE) of solid-phase diffusion process, and a lithium-ion concentration model in the radial direction can be obtained. Next, a terminal voltage model of battery including liquid potential, electrochemical polarization potential and ohmic polarization potential is established. This modeling process considers the double electric layer (DEL) effect on liquid potential. Finally, a variable parameter model based on temperature and SOC is developed to improve the adaptability of the mechanism model, and a novel adaptive multi-innovation unscented Kalman filtering (AMUKF) is proposed to estimation the SOC of LIB. A various of experiments and verifications show that the presented model is effective, and has better modeling ability in the full SOC stage and changeable environment compared with other several common models.