Fast sorting method for lithium-ion batteries based on partial frequency bands of electrochemical impedance spectroscopy

Abstract

Efficient sorting of lithium-ion batteries (LIBs) is essential to reduce inconsistency within battery packs and improve overall performance. At the same time, with the increasing number of retired batteries, sorting and recycling are critical for energy savings. Traditional sorting methods based on time domain characteristics are time-consuming, energy-intensive, and inefficient for large-scale applications. This study proposes a fast LIBs sorting method utilizing partial frequency bands of electrochemical impedance spectroscopy (EIS). The method generates the required EIS frequency bands within 5s, reducing testing time by 96 %. To address challenges such as real part drift and the impact of full-spectrum use on sorting accuracy, the method employs dispersion analysis to select the imaginary part of specific EIS frequency bands as the sorting feature, enhancing algorithm stability and accuracy. The sorting process is optimized using the elbow method and K-means++ algorithm. Experimental results show that the proposed method reduces sorting time from several hours to under 5min and significantly lowers energy consumption compared to traditional time domain methods. A comparative analysis with three other sorting methods demonstrates that the proposed method maintains high accuracy while ensuring speed and efficiency. This approach demonstrates exceptional suitability for large-scale battery sorting applications, delivering substantial economic advantages and efficiency improvements.