Estimation of temperature field for blade battery based on frequency domain heat generation model

Abstract

Sinusoidal pulse excitation is widely used in the pre-heating process of commercial lithium batteries due to its fast heating rate and almost no capacity degradation under long cycles. For the emerging blade cell, its temperature distribution gradient is significant under the Alternating Current (AC) pulse, and it isn't easy to realize the all-around monitoring of its temperature by the lumped parameter model. So this paper proposes a Joint Kalman Filter (JKF) based on a distributed heat source, and it can realize the joint estimation of tab temperature and the time-varying thermal coefficient. Firstly, the impedance at different temperatures is analyzed by Electrochemical Impedance Spectroscopy (EIS), and the distributed heat model in the frequency domain is constructed. Then, the thermal resistance grid is used to divide the nodes, and the heat transfer equation is formed by using the multi-point temperatures and time-varying heat coefficients to estimate the temperature field. Simulation results show that the method can track the temperature field of the blade cell and realize the temperature estimation with a maximum error of less than 1.5 ℃ in a wide domain range, which is expected to realize the online temperature management and has great application potential in the design of the thermal management system.