A novel method for joint estimation of SOC and SOP based on electrochemical modeling

Abstract

Lithium-ion batteries play an important role in electric vehicles, and their internal condition is crucial in battery operation safety. However, lithium-ion battery is difficult to observe the negative state owing to the relatively flat equilibrium potential of the negative electrode, resulting in an inaccurate estimation of the negative state through voltage. Therefore, a novel joint estimation method for SOC and SOP is developed in this study through electrochemical modeling, enabling precise characterization of lithium batteries internal states. First, the extended single particle model (eSPM) is established, along with a newly developed variable-scale parameter identification methodology is designed to address convergence challenges, enabling accurate extraction of the model's 21 parameters. Subsequently, a Dual Unscented Kalman Filter (DUKF)-based observer is developed to enable real-time monitoring of the battery internal state variables. Finally, the state of power prediction is accelerated by the estimated value of DUKF combined with support vector regression. The methods of this paper are validated through experiments, and the results show that the methods can realize high- accuracy battery state estimation and power prediction, which can provide technical support for detecting the safe operation of batteries.