An integration and selection scheme for capacity estimation of Li-ion battery based on different state-of-charge intervals

Much research has been done to predict batteries' cyclic aging state-of-health (SOH) over the entire state-ofcharge (SOC) range in constant-voltage and constant-current charge-discharge mode. Considering that the battery cycle is usually carried out in part of the SOC range in the actual use process, this paper further studies the cycle degradation characteristics of Lithium-ion (Li-ion) batteries in the partial SOC ranges. First, to clarify the factors that affect battery cycle aging, the changes in battery degradation characteristics caused by various stresses are analyzed. Then, a scheme to construct the state of charge - voltage (SOC-V) curve based on the firstorder equivalent circuit and lumped thermal model is introduced for accurately extracting health indexes (HIs) on incremental capacity (IC) and differential voltage (DV) curves. Meanwhile, the least-squares method is used to identify the offline parameters of the proposed model. Perform a sensitivity analysis on the equation complexity and fitting effect, which can determine the exact model order. Multiple regression models are employed, which map the relationship between HIs fluctuations and capacity degradation for real-time estimation of SOH in different SOC intervals. Finally, different SOC intervals screen out the most accurate prediction models to realize the SOH prediction in all SOC cycle ranges. In this framework, the results show that SOH prediction based on SOC interval is desirable. Analysis results reveal the above measures' trustworthiness, with the average predicted error being within 1 % under the cross-validation test. What's more, a set of different types of batteries are used to verify the robustness of this method.