Aging mechanism analysis under different charging voltages and online SOH estimation of Li-ion batteries

The performance state of lithium-ion batteries directly impacts the stability of energy storage system operations. With prolonged use, lithium-ion batteries undergo complex electrochemical changes, leading to capacity degradation and reduced performance. To accurately estimate the state of health (SOH) for lithium-ion batteries in energy storage application scenarios, this study conducts aging tests on lithium-ion batteries under different charging voltages and develops an online model-based SOH estimation method. First, excitation response analysis and an extended Kalman filter algorithm are used to identify battery parameters of a simplified electrochemical model both offline and online. Then, by analyzing parameter change laws during battery aging and the correlation between the parameters and battery capacity, aging mechanisms are obtained and battery health features are further extracted. Finally, an SOH estimation model based on a support vector regression algorithm is developed with both offline and online parameter sets.