A Fast Impedance Measurement Method for Lithium-Ion Battery Using Power Spectrum Property

Electrochemical impedance spectroscopy (EIS) can provide fruitful information for Lithium-ion (Li-ion) battery modeling and diagnosis, yet EIS measurement is time-consuming with low-frequency signal injection. By stacking a group of broadband signals, pseudorandom sequence (PRS) makes it possible to obtain the battery EIS in a few seconds at the expense of measurement accuracy and signal-to-noise ratio (SNR). Thus, this article focuses on developing a highly effective signal processing procedure to extract useful information from the PRS for accurate EIS measurement. To enhance the ability of the data cleaning procedure, a three-dimensional cloud is first reconstructed for each impedance by integrating its power spectrum (PS). The impedance with lower PS can be easily removed through a statistical based multiple selection mechanism, which enables the extraction of the EIS without altering the original measurement. Experimental results on a 3000 mAh Li-ion battery prove the effectiveness of the proposed method.