Fractional-order dynamic model-based terminal voltage estimation for vanadium redox flow battery

Abstract

The vanadium redox flow battery (VRFB) is a prominent large-scale energy storage technology, distinguished by its exceptional cycle life and high safety performance. The development of an accurate model capable of replicating VRFB characteristics under real operating conditions is imperative for the advancement of control strategies and condition monitoring techniques. While traditional integer-order models have been extensively utilized, they are inadequate in describing the complex electrochemical behaviors of VRFBs, particularly in capturing the non-integer-order dynamic characteristics inherent in the polarization process.

To address this limitation, the present study proposes a novel dynamic model for VRFBs based on fractional-order theory. The experimental methodology involves constant-current intermittent discharge testing. Subsequently, a parameter identification method incorporating nonlinear least squares is employed to determine key model parameters, including internal ohmic resistance, polarization resistance, fractional order capacitance, and fractional order. The identified parameters are then applied to simulate the performance of a single VRFB cell. The simulation results demonstrate that the battery's constant-current intermittent discharge output voltage follows a characteristic three-stage pattern across varying discharge rates. In conclusion, to validate the proposed model, the battery's voltage response is subjected to experimental testing under a variety of constant-current intermittent discharge conditions. The findings of the study demonstrate a high degree of concordance between the model and the experimental data. The absolute error recorded was no greater than 3.58 %, and the root-mean-square error was no greater than 5.67 mV. This level of accuracy serves to confirm the model's reliability for practical applications. In addition, the established theoretical framework provides a solid foundation for advanced VRFB control strategies and state monitoring, which contributes to the optimization of large-scale energy storage systems.