Parameterization of prismatic lithium-iron-phosphate cells through a streamlined thermal/electrochemical model

A model is proposed and used to parameterize the surface temperatures and electrical responses of A123 20 Ah LiFePO4 prismatic cells. The cell interior is described by a porous-electrode charge-transport model based on Newman-Tobias theory, which is coupled to a local heat balance. Dimensional analysis suggests that a multilayer electrode sandwich can be approximated as a single layer with appropriate rescalings of the model parameters, dramatically speeding computation. The simulation output depends on only a few observable dimensionless quantities, allowing parameter estimation through an iterative optimization scheme that directly compares computed results with measurements that track the cell voltage, while simultaneously recording infrared thermograms of the surface-temperature distribution. Despite the neglect of mass-transport limitations within Newman-Tobias theory, the model accurately predicts the dynamic terminal voltage, as well as the minimum, maximum, and surface-averaged temperature on the cell exterior. The electrochemical and thermal properties extracted from square-wave cycling data with various excitation amplitudes (2 C and 4 C) and short charge/discharge periods (50 s and 100 s) compare well with literature values, showing that it is possible to infer internal material properties by fitting external measurements.