Experimental estimation of core temperature and directional thermophysical properties for cylindrical lithium-ion battery utilizing an innovative thermal interrogation method

Abstract

The research focuses on the importance of estimating the core temperature and directional thermophysical properties of lithium-ion batteries to enhance performance and safety, as the choice and design of proper thermal management systems depend on these factors. Therefore, the aim of this research is to develop a new experimental noninvasive battery thermal interrogation model based on the lumped capacitance method to estimate the battery core temperature, specific heat capacity, and directional thermal conductivities without environmental interference. This method utilizes a combination of directional (radial and axial) experimental real-time transient surface heat flux and temperature measurements on the outside of the cylindrical lithium-ion (18650) battery cell surface. Unlike the literature thermal methods, this method does not depend on the external environmental effects because the heat flux sensor is mounted underneath the thin-film polyimide electric heater and it measures the real-time transient heat flux to the battery cell. Moreover, the heat flux measurements are used as input in the transient thermal model rather than the temperature measurements to provide a good estimation of the battery core temperature and the battery directional thermophysical properties. A simple parameter estimation algorithm is used to estimate the optimal parameters by using the minimum root mean square error between the analytical and experimental difference in directional surface temperature values. The results are compared with other different methods, such as two-state thermal method, inverse heat conduction thermal method, and Kalman filter method. The results demonstrate that the proposed method accurately captured the battery core temperature, the specific heat capacity, and the directional thermal conductivity values with small standard deviation and minimum 95 % confidence interval. This proposed method can be applied to different geometries and chemistries of lithium-ion batteries.