Abstract：Accurately estimating the state of health（SOH）of lithium-ion batteries is a crucial prerequisite for

ensuring the safe and stable operation of energy storage systems. The key to improving the accuracy of SOH

estimation lies in the rational selection of health characteristics that can effectively reflect the state of health of

lithium-ion batteries. By analyzing the current characteristics of lithium-ion batteries during the constant voltage

charging stage，a healthy combination of features containing the slope of the first and last points of the current

curve，the standard deviation，and the mean value were extracted from the current curve data during the constant

voltage charging stage. To validate the effectiveness of the proposed feature combination，SOH estimation model

based on kernel ridge regression（KRR）and support vector regression（SVR）was designed，and model validation

was successfully completed. The experimental results demonstrate that the proposed feature combination can

achieve high-precision SOH estimation across different models，exhibiting excellent model adaptability.

      Key words：lithium-ion battery；state of health（SOH）estimation；constant voltage charging stage；kernel ridge regression（KRR）；support vector regression（SVR）