Identification of the parameters of the aluminum-air battery with regard to temperature

Abstract

Aluminum-air batteries have the advantages of clean raw materials, high theoretical energy density, and easy storage and transportation. It has been applied in the fields of communication base station backup power, underwater equipment driving energy and electric vehicle energy. However, aluminum-air batteries have the problems that the remaining power cannot be expressed visually and is greatly affected by external factors. Therefore, this paper establishes an equivalent circuit model of aluminum-air batteries affected by ambient temperature to obtain the real-time output voltage and SOC of aluminum-air batteries more accurately. Firstly, it is found that the performance of the battery is optimized with the increase of temperature between 10 °C and 30 °C by conducting open-circuit voltage (OCV) and linear scanning voltammetry (LSV) tests at different temperatures. The OCV increases from 1.6675 V to 1.728 V and the peak power density increased from 46.269 mW cm⁻² to 123.95 mW cm⁻². Thus, it is shown that the discharge performance of aluminum-air batteries at different temperatures varies greatly and there is a relationship. The electrochemical impedance spectroscopy (EIS) testing of the aluminum-air battery was then continued at different temperatures. And several equivalent circuit models were fitted using ZSimpWin. By comparison, the second-order RC model had the smallest fitting error of 0.502 % at the three temperatures. It was therefore determined that the second-order RC model would be used in subsequent work. Finally, the second-order RC model based on the variation with ambient temperature was established by identifying the battery parameters in the second-order RC equivalent circuit model using the method of least squares with forgetting factor (FFRLS) based on the hybrid pulse characteristic (HPPC) experiment. The simulated voltage is also compared and verified with the measured voltage, and the results show that the maximum absolute error is 0.0182 V and the maximum relative error is only 1.4 %. The high accuracy of the developed model can be effectively demonstrated by the voltage comparison results. The model provides an effective simulation for the parameter identification of aluminum-air batteries, and also paves the way for the subsequent estimation of the SOC of aluminum-air batteries at different temperatures.