An electrothermal coupling model for estimating the internal temperature of lithium-ion battery based on microthermal resistance method

In the current study, the CN-CD30V10A battery tester, K-type temperature sensor, and RS485 are utilized to perform charging and discharging tests on the Panasonic 21,700 lithium-ion battery at various rates. The voltage and temperature changes are measured in real-time, with the data transmitted to the computer via RS485. Then a micro-thermal resistance method is proposed to calculate the equivalent thermal resistance of the cell core along the axial positive and negative poles, as well as along the radial outer circumference surface under anisotropic conditions. Based on the experimental results and the equivalent thermal resistance, an electrothermal coupling model is established for battery internal temperature estimation, taking into account anisotropy and temperature effects on heat transfer. Through the integration of the model and externally measured temperature, the internal temperature is accurately estimated. Moreover, the proposed model (Root-Mean-Square Error, RMSE=0.329) exhibits a significant improvement in accuracy compared to equivalent circuit model (RMSE=1.08), extended Kalman filter (RMSE=0.95), network extended Kalman filter (RMSE=0.58), and neural network unscented Kalman filter (RMSE=0.61). Specifically, the model achieves an approximately 76 % to 228 % reduction in RMSE compared to these methods.