2RC model based state of charge estimation of lithium-ion battery

Abstract

Safe reliable and efficient utilization management of rechargeable lithium-ion batteries relies on accurate real-time state-of-charge (SOC) estimation. Real-time estimation of the state of charge (SOC) of the battery is a crucial need in the growing fields of plug-in hybrid electric vehicles and smart grid applications. Model-based SOC observers have been widely used to estimate the SOC due to their high accuracy and robustness. The accuracy of the model used to characterize the battery's properties directly affects the estimation algorithm's accuracy. We employ a piecewise linear approximation with variable coefficients to represent the intrinsically nonlinear relationship between the open-circuit voltage (VOC) and the SOC of the battery, taking into account a resistance-capacitance (RC)–equivalent circuit to simulate the dynamics of the battery. Numerous experimental test results on lithium (Li)-polymer batteries demonstrate that the RC parameters also change in response to changes in the SOC and charging/discharging rates, in addition to the VOC–SOC relationship coefficients. The typical battery state-space model's observability matrix is ill-conditioned when the battery's slope from state-of-charge to open-circuit voltage (VOC) profile is near zero. Normally overlook this issue when we estimate the state of Charge and get unreliable SOC estimation. So choose the state carefully and resolve the observability problem. And model the battery parameter and thermal effect with the proper dependency of SOC and temperature (T). This article describes and analyzes the SOC estimation of both the conventional state space and new state space based on simulation.