Practical Online Estimation of Lithium-Ion Battery Apparent Series Resistance for Mild Hybrid Vehicles

• Published in: IEEE transactions on vehicular technology Vol. 65; no. 6; pp. 4505 - 4511
• Main Authors: Lievre, Aurelien, Sari, Ali, Venet, Pascal, Hijazi, Alaa, Ouattara-Brigaudet, Mathilde, Pelissier, Serge
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-06-2016; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); Institute of Electrical and Electronics Engineers
• Subjects: Batteries; Current measurement; Electric batteries; Electric power; Electrical resistance measurement; Engineering Sciences; Estimation; Evolution; Hybrid vehicles; Impedance; Lithium; Lithium-ion batteries; Monitors; Noise measurement; Online; Other; Rechargeable batteries; Resistance; Series resistance; State of Health; Vehicles; Exponential moving average; lithium-ion batteries; mild hybrid vehicle; series resistance; state of function; online parameter identification; state of health (SoH); BATTERIE; MILD HYBRID VEHICLE; EXPONENTIAL MOVING AVERAGE; VEHICULE HYBRIDE; SERIES RESISTANCE; STATE OF HEALTH; LITHIUM-ION BATTERIES; LITHIUM; ONLINE PARAMETER IDENTIFICATION; STATE OF FUNCTION
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2015.2446333

In hybrid vehicles, lithium-ion cells constituting a battery pack are frequently used to provide and recover high power to assist the vehicle's internal combustion engine (ICE) powertrain. This usage is more present in mild hybrid applications where the battery does not have long discharge time. Under such conditions, the pack's series resistance R S proved to be an important parameter to monitor since its evolution depends on the cell's characteristics (manufacturing tolerance, temperature, etc.). This resistance, which is monitored by the battery management system (BMS), reflecting the available power level in the cell can be used as an indicator to enhance the security of the battery pack. Its evolution can be used to quantify its aging (state of health: SoH). This paper presents an online approach to identify the cell's series resistance based on a direct estimation of R S . This parameter can be usually identified through the voltage drop occurring across the cell caused by a high current variation profile (mild hybrid conditions). These estimated values are then filtered with an "exponential moving average" method to limit the measurement noise effect. This approach provides good results for mild hybrid conditions, while minimizing the computing power required.