Anti-Windup Co-Estimation of Open Circuit Voltage and Equivalent Circuit Model Parameters for Lithium-Ion Battery Diagnostics

• Published in: IFAC-PapersOnLine Vol. 56; no. 2; pp. 11179 - 11184
• Main Authors: Lochrie, Gabrielle, Yoon, Yongsoon
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 2023
• Subjects: Automotive Control; Fault Detection, Supervision & Safety of Technical Processes; Mechatronic Systems; Transportation Systems; Automotive Control; Mechatronic Systems; Transportation Systems; Fault Detection, Supervision & Safety of Technical Processes
• ISSN: 2405-8963; 2405-8963
• DOI: 10.1016/j.ifacol.2023.10.841

This paper presents anti-windup co-estimation of open circuit voltage and equivalent circuit model parameters for on-board diagnostics of a lithium-ion battery. The open circuit voltage and equivalent circuit model parameters of the lithium-ion battery depend on battery core temperature and state of charge. They are also subject to variation from manufacturing tolerance, system aging, and faults. For example, battery aging affects electrode stoichiometry, and an open circuit voltage curve can change as a result. And, the equivalent circuit model parameters can vary due to abnormally repeated over-charging and over-discharging. Therefore, estimating both the open circuit voltage and equivalent circuit model parameters is of significance on health monitoring of the lithium-ion battery. Toward this end, this work aims to develop an online co-estimation method of the open circuit voltage and equivalent circuit model parameters, while avoiding covariance windup under poor external excitation. Three different anti-windup recursive parameter estimation methods are investigated: 1) directional forgetting recursive least squares; 2) variable forgetting recursive least squares; 3) Stenlund-Gustafsson Kalman filter. The developed co-estimation methods are numerically illustrated with the scaled heavy-duty urban dynamometer driving schedule.