A fast method for estimating the state of charge and state of health of lithium-ion batteries based on electrochemical impedance spectroscopy at multiple frequency

Abstract Electrochemical impedance spectroscopy (EIS) is widely used for state estimation of energy storage batteries. Currently, EIS is typically obtained via frequency sweep methods using electrochemical workstations, which incur high detection time costs. Existing methods also show limited accuracy in battery state estimation. To reduce measurement time and improve estimation accuracy, a transformer-based estimation method utilizing electrochemical impedance at multiple frequency points is proposed. Specifically, by conducting EIS experiments on 18 650 NCM lithium batteries manufactured by Tianjin Lishen under various states of charge (SOC) (0%–100%), temperatures (15 °C, 25 °C, 35 °C, 45 °C), and aging levels (80%–100%). Characteristic frequencies such as 794.3 Hz, 50.1 Hz, 10 Hz, and 0.158 Hz are selected to reconstruct the full impedance spectrum. Key parameters needed for battery state estimation are identified to enable accurate state assessment of the battery. The results indicate that the reconstruction method proposed in this paper exhibits excellent reproducibility of the original data, and the selected characteristic parameters enable accurate estimation of the battery’s state of health (SOH) and SOC. Compared with the frequency sweep method, the measurement time is reduced by 42%, with maximum estimation errors of 5% for SOH and 3% for SOC. The estimation performance of the proposed method surpasses that of several existing methods.