State of health assessment for lithium batteries based on voltage–time relaxation measure. (10th March 2016)

• Record Type: Journal Article
• Authors: Baghdadi, Issam; Briat, Olivier; Gyan, Philippe; Vinassa, Jean Michel
• Is Part Of: Electrochimica acta. Volume 194(2016)
• Page Start: 461
• Page End: 472
• Publication Date: 2016-03-10
• Subjects: lithium battery -- calendar aging -- cycling ageing -- state of health -- voltage relaxation; Electrochemistry -- Periodicals; Electrochemistry, Industrial -- Periodicals; 541.37
• DOI: 10.1016/j.electacta.2016.02.109
• Languages: English
• ISSNs: 0013-4686

Abstract

Highlights: Calendar aging under different storage conditions for three different battery technologies studied. Two scenarios of aging under power cycling at two different temperatures investigated for one battery technology. Relaxation profile of battery voltage just after full charge is highly correlated to aging. Linear dependence between just after charge open circuit voltage and remaining capacity demonstrated. No computational method and direct prediction of battery state of health or remaining capacity. Abstract: The performance of lithium batteries degrades over time. The degradation rate strongly depends on stress conditions during use and even at rest. Thus, accurate and rapid diagnosis of battery state of health (SOH) is necessary for electric vehicle manufacturers to manage their vehicle fleets and warranties. This paper demonstrates a simple method for assessing SOH related to battery energy capability (SOHE ). The presented method is based on the monitoring of Urelax over aging. Urelax is the open-circuit voltage of the battery measured after full charging and 30 min of rest. A linear dependence between Urelax and remaining capacity is noted. This correlation is demonstrated for three different commercial battery technologies (different chemistries) aged under different calendar and power cycling aging conditions. It was determined that the difference between two Urelax voltages measured at two different aging states is proportional to SOHE decay. The mean error of the linear model is less than 2% for certain cases. This method could also be a highly useful and rapid tool for a complete battery pack diagnosis.