Modeling State of Health In Li-Ion Cells: Aging Analysis and Cycle Performance Validation

This study presents a comprehensive analysis of the aging behavior and cycle performance of commercial lithium-ion (Li-ion) cells, with the objective of developing accurate State of Health (SoH) estimation models. Empirical data were collected through a series of cycling experiments at varied current, depth of discharge (DoD) ratings, and cycle targets, thereby creating a robust data pool. Utilizing this data, two distinct SoH estimation models were generated. The first model is a detailed analysis incorporating parameters such as cycle count, depth of discharge, and charge current. The second model is based on direct current internal resistance (DCIR) and the number of cycles. Comparative analysis demonstrates that the detailed model, developed using fine mesh tree modeling, significantly outperforms the basic linear regression model in terms of accuracy. Nevertheless, both models are viable for SoH estimation, with the simpler model offering ease of control and implementation. These models are intended to assess the SoH levels of cells collected from customers and field applications for second-life usage.