Monitoring of State of Charge and State of Health of Vanadium Redox Flow Battery via Extended Kalman Filter

Abstract

Vanadium Redox Flow Batteries (VRFBs) offer a compelling pathway for large-scale, long-duration energy storage, where reliable operation depends on accurate tracking of both the state of charge (SOC) and the state of health (SOH). This work introduces an Extended Kalman Filter (EKF) framework coupled with a first-order equivalent-circuit model to jointly estimate SOC, internal resistance, and effective capacity. Unlike conventional VRFB management strategies, the proposed method eliminates the need for dedicated open-circuit voltage (OCV) cells, enabling SOC estimation using only stack measurements. Capacity degradation is captured through event-triggered pseudo-measurements at end-of-discharge, allowing the algorithm to distinguish between resistance growth and true energy-capacity loss, which is particularly relevant for VRFBs where power and energy capabilities are inherently decoupled. The framework is validated using experimental data from a 5 kW VRFB test bench and further evaluated under synthetic aging scenarios, including increasing ohmic resistance, progressive capacity fade from 60 Ah to 40 Ah, and measurement noise. Across these tests, the EKF achieves an SOC estimation error of 0.66%, internal resistance estimation within 3%, and capacity estimation below 1.6% under nominal conditions, while maintaining strong robustness under sensitivity analysis. Overall, the proposed method advances the safe, efficient, and durable operation of VRFBs.