State-of-health estimation of LiFePO4/graphite batteries based on a model using differential capacity

•A model for SOH estimation was based on the deformed pseudo-Voigt peak function.•The parameters were associated with the realistic phase transition behavior.•The SOH estimation was performed with different charge/discharge histories.•The estimated SOH errors were within ±3% using a part of charged curve.

A model for expressing the differential capacity characteristics of the LiFePO4 (LFP)/graphite battery for the state-of-health (SOH) estimation was proposed. Our model was based on the deformed pseudo-Voigt peak function with several parameters which are directly associated with the phase transition behavior of the active LFP and graphite materials. Charge/discharge cycle tests for accelerated battery fading were performed under a constant high-temperature condition (40 and 45 °C). The SOH estimation was carried out at different fading point of the battery using a part of the responses for the differential capacity versus voltage (dQ/dV vs. V) against the charging process at the rate of C/5 under constant temperature of 25 °C. The changes in the variables of the model with cycling were correlated to the generally mentioned phenomena that the main factors determining the capacity fading of the LFP/graphite battery are the loss of Li+ by a side reaction and that of the active electrode materials. In addition, the robustness related to the charge/discharge history was confirmed, since the memory effect of the LFP/graphite battery, being induced by the previous condition for use, has an influence on the dQ/dV vs. V. The evaluated SOH errors were within ±3%.