This Poster presents an online estimation method for Open Circuit Voltage (OCV) using an Extended Kalman Filter (EKF) and a self-evaluation criterion to improve State of Charge (SoC) estimation accuracy in LiFePO₄ (LFP) batteries. Traditional OCV estimation relies on offline measurements, making real-time applications challenging. To address this, the proposed approach continuously adjusts OCV values using capacity differences in small SoC intervals.

The method follows a four-step process:

1.EKF-Based SoC Estimation – The EKF estimates SoC, with updates triggered for SoC changes exceeding 1%.
2.OCV Adjustment via Self-Evaluation – Capacity differences between expected and actual values are used to modify OCV values iteratively, ensuring monotonicity.
3.Scaling the OCV Curve – Adjusted OCV values are proportionally scaled while keeping 0% and 100% SoC voltages unchanged.
4.OCV Curve Update – The new OCV curve is used in the next iteration for SoC estimation.
Tested in an energy storage system (ESS) with a 150 Ah LFP battery, the method underwent 200 discharge iterations, showing significant accuracy improvements. The SoC estimation error remained below 0.3%, and OCV values gradually aligned with ground truth. The approach is adaptive, robust, and self-correcting, making it ideal for real-world applications.

Future work aims to extend the method to NCM and NCA-based batteries and integrate capacity estimation for enhanced battery health monitoring.