Temperature Dependence of Li-ion Battery Hysteresis for Battery Modeling Purposes

Battery modeling is becoming more and more important in advanced battery management systems (BMS). Equivalent circuit models are widely used for their low computational requirements but they struggle with lower accuracy, which remains a challenge for the future. In this paper, a special focus is dedicated to the battery hysteresis effect, which significantly complicates the whole battery modeling and estimation processes. Measurement of the open circuit voltage (OCV) and parameter identification of the equivalent circuit model for the most used Li-ion battery technologies nickel-manganese-cobalt (NMC) and lithium-iron-phosphate (LFP) were performed. Experimental results showed that the most common approaches of battery modeling are not accurate enough and can cause problems, especially in the state of charge (SoC) estimation. Besides that, the lower the temperature, the bigger impact of the hysteresis effect, especially during discharging. That indicates this effect is closely linked to diffusion processes inside the electrode material, which are slower at lower temperatures. These findings have a great potential to help in the modeling of the hysteresis effect in the future.