Lithium-ion-battery state of health estimation based on coefficient of variation

Abstract The state of health (SOH) is an important indicator of the reliable and safe working of lithium-ion battery systems in new energy vehicles and energy storage power stations. In order to find a fast, accurate, concise and understandable approach for estimating the SOH of lithium-ion batteries, a health estimation method for lithium-ion batteries is proposed based on the coefficient of variation (COV) as the health factor. The battery SOH is estimated by deriving the COV based on the charge and discharge voltage series of li-ion batteries over a certain voltage window or a certain state of charge (SOC) range, calculating its linear relationship with the SOH, and finally building a COV-SOH model. The effects of different voltage windows on the health estimation model of nickel-cobalt-aluminum ternary batteries (NCA) and ferrous lithium phosphate (LFP) batteries are also investigated at different multiplications of the charging and discharging process. The results illustrate that for NCA batteries, the COV is linearly positively correlated with SOH for both voltage series based on 0.25C, 0.5C, and 1C charging multiplicity as well as 1C discharging, with a coefficient of determination (R 2 ) of up to 0.98 after linear fitting; for the LFP batteries, the COV is linearly negatively correlated with SOH, whether based on partial voltage sequences of 5.2C and 4.2C charging or 4C discharging, with a R 2 of up to 0.99 after linear fitting. The COV-SOH model demonstrates a strong linear correlation for both NCA and LFP batteries, under both charging and discharging conditions, and across a range of current multiplication rates, indicating the model’s high robustness.