State of Health Estimation of Lithium-Ion Batteries Based on Voltage Data Segment Hybrid Model

The health state estimation of lithium-ion batteries are the essential issues for the safety of energy storage stations. The important indicators often focus on the battery capacity and internal resistance. However, the measurement of capacity requires a complete charge/discharge cycle, and the measurement of internal resistance requires additional equipment. To solve the above problems, based on the voltage segment under the constant-current discharge condition of lithium-ion battery, this paper takes the sharp voltage drop of the initial discharge segment as a new healthy factor. Furthermore, facing the possibility that the new healthy factor data is polluted by noise, this factor data is reconstructed to reduce noise through multi-order Bezier curve. Subsequently, an empirical degradation hybrid model is constructed with the number of cycles. On this basis, the battery healthy state is defined by voltage segment and a new healthy state estimation model is proposed. The feasibility and effectiveness of the proposed degradation model and estimation model are verified by the aging data published by NASA and experimental platform.