Online estimation of model parameters and state-of-charge of Lithium-Ion battery using Unscented Kalman Filter

Abstract

For the operation of Autonomous Mobile Robot (AMR) in unknown environments, accurate estimation of internal parameters and consequently precise prediction of the battery state of charge (SoC) are critical issues for power management. Battery performance can be affected by factors such as temperature deviation, discharge/charge current, Coulombic efficiency losses, and aging. Thus, in order to increase the model accuracy, it is important to update the model parameters online. In this paper, the Unscented Kalman Filter (UKF) is employed for the online estimation of the Lithium-Ion battery model parameters and the battery SoC based on the updated model. The proposed method is evaluated experimentally, and the results are compared with that of the Extended Kalman Filter (EKF). The comparison with the EKF shows that UKF provides better accuracy both in battery parameters estimation and the battery SoC estimation.

State of charge estimation of power Li-ion batteries using a hybrid estimation algorithm based on UKF

“…The battery SOC, the voltageU d of capacitor'sC d in battery model and the voltageU e of capacitance'sC e together to form a three-dimensional state variables, battery terminal voltage y k as observation variables, combined with Ah integration method to estimate the state variables. Because of power battery model is a complex system, the established model can't completely and accurately describe the characteristics of the power battery system, in addition, measuring errors inevitably exist in measuring the terminal voltagey k and the charge or discharge currenti k .Therefore, the process noise and observation noise should be taken into account, state equation and observation equation is obtained respectively as follows [27,28]:…”

Online battery monitoring using model based approach - impedance

“…It implements a Li-ion battery. [13] describes an Unscented Kalman Filter for online estimation. [14] proposes rate data given in manufacturers' data sheets, and coulometric measurements, without the need for any battery voltage measurement under open-circuit conditions.…”

Dynamic model of lithium polymer battery – Load resistor method for electric parameters identification

“…In [27] the Gauss-Hermite quadrature filter (GHQF) is introduced to estimate battery state of charge (SOC) based on a common electrical analogue battery model useful for real time applications. The authors in [28] propose an Unscented Kalman Filter (UKF) for ‗online' estimation of the Lithium-Ion battery model parameters and the battery SOC, based on the updated model. All these techniques required not simple equations and add power consumption related to perform the necessary calculations [26].…”