Aiming at the problem of how to extract more effective characteristic parameters from the capacity increment (IC) curve for state of health (SOH) estimation of lithium batteries

a modified Lorentz voltage-capacity (RL-VC) based model is proposed. The capacity increment analysis (ICA) of lithium batteries is first performed using the traditional filtering method. Then the RL-VC model is used for comparison to obtain the corresponding feature parameters and calculate the capacity modeling error. The experimental data obtained based on the self-constructed experimental platform and the dynamic dataset NCM from the open-source dataset NASA are carried out separately. The errors of VC capacity modeling are within 0.23% and 0.16%

respectively. The feature parameters extracted from the IC curves fitted by the RL-VC model are highly linearly correlated with the capacity of Li-ion batteries

which lays the foundation for the subsequent SOH work. The IC analysis method based on the RL-VC model proposed in this paper not only has higher robustness in battery aging compared with the traditional filtering method

but also avoids subjectivity and uncertainty in feature parameter extraction.