State of health (SoH) monitoring is now an essential part of all advanced battery monitoring systems (BMS). Analysis of the health of the battery is essential for predicting the battery lifespan, ensuring efficient operation and protection against unsafe usage. In this paper, an accurate model for online capacity estimation of Li-ion batteries was developed using the linear relationship between the fade in capacity and the increase in impedance. Through the analysis of the effect of ageing on battery impedance characteristics, the relationship between the fade in battery capacity to the increase in battery impedance was used to estimate the current capacity of the battery. Firstly, analysis of the impedance at multiple frequencies was performed to ascertain which frequency’s displayed high correlation to the fade in battery capacity. Next, a calibration was performed for each battery to determine the capacity / impedance relationship. Finally, 6 Li-ion polymer batteries were tested over 300 cycles to validate the accuracy and reliability of the model. The capacity estimates indicated that the average RMSE (Ah) and percentage error (%) of estimates were very low at 0.0080 Ah and 0.241 %. All impedance measurements for the proposed method were performed online, indicating its suitability to be used in real-world applications.
