Capacity attenuation mechanism modeling and health assessment of lithium-ion batteries

Lithium-ion battery is a complex thermoelectric coupling system, which has complicated internal reactions. It is difficult to investigate the aging mechanism due to the lack of direct observation of side reaction. In response, a method of aging mode identification based on open-circuit voltage matching analysis is proposed in this work. Firstly, the LiCoO2 and graphite half cells are made to measure the open-circuit voltage for electrodes. The open-circuit voltage model of the full cell is established based on the state of charge matching relationship between the full cell and electrodes. Then, a non-destructive aging mechanism identification method is developed, which can quantify the loss of lithium inventory, the loss of active materials of electrodes. Whereafter, the aging semi-empirical models of the three aging modes are established respectively, and the mapping models with state of health, ohmic resistance and polarization resistance evolution are developed. Besides, the short-term state of health and remaining useful life prediction method is proposed based on the particle filter algorithm and established models. Finally, the developed models and methods are validated by the battery data. The experimental results show that the root mean square error and mean absolute error of the calculated voltage are kept within 38 mV and 51 mV. The root mean square error of RUL and short-term SOH prediction are maintained within 5.549 and 1.31%, respectively. And the predicted RUL remains within the 95% confidence interval. The results further prove that the established models and methods have high accuracy. (c) 2020 Elsevier Ltd. All rights reserved.

Automated summary

This work proposed an aging mode identification method based on open-circuit voltage matching analysis, established the open-circuit voltage model of the full cell, and developed a non-destructive aging mechanism identification method to quantify the loss of lithium inventory and active materials of electrodes. The developed models and methods showed high accuracy in predicting remaining useful life and short-term state of health.