Accelerated fading recognition for lithium-ion batteries with Nickel-Cobalt-Manganese cathode using quantile regression method. (15th December 2019)

• Record Type: Journal Article
• Authors: Zhang, Caiping; Wang, Yubin; Gao, Yang; Wang, Fang; Mu, Biqiang; Zhang, Weige
• Is Part Of: Applied energy. Volume 256(2019)
• Publication Date: 2019-12-15
• Subjects: Nickel-Cobalt-Manganese lithium-ion battery -- Accelerated aging -- Sudden degradation -- Recognition -- Quantile regression; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2019.113841
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Battery knee point recognition new method using quantile regression is proposed. The dynamic boundary determination method for the whole lifetime is developed. Recognition result is effective even if the input is disturbed. The proposed method has strong reliability and stability at different conditions. Abstract: The requirement for energy density of lithium-ion batteries becomes more urgent due to the rising demand for driving range of electric vehicles in recent years. Meanwhile, the performance stability of batteries with high energy densities tends to deteriorate, leading to accelerating degradation and safety issues. As a result, it is critical to explore the reasons that yield the sudden degradation and to recognize the degradation knee point of Nickel-Cobalt-Manganese batteries commonly used for electric vehicles. Existing results have disclosed that the lithium deposition of negative electrode dominates the sudden degradation of battery capacity. This paper extracts key parameters that characterize the aging status to facilitate knee point recognition in engineering practice. Furthermore, a novel method that integrates quantile regression and Monte Carlo simulation method to identify the accelerated fading knee point is introduced. The dynamic safety boundary determination method for the whole battery lifetime is proposed to update and monitor the safety zone. It is verified by experiments that the recognition results of capacity degradation knee point appear within 90–95% capacity range at 25 °C, 35 °C and 45 °C conditions, which can provide an early warning before the battery fails. Using the proposed method for recognizing the sudden degradation of capacity, recognition result is effective even if the input is disturbed and has strong reliability and stability under different conditions. It is helpful to promote the sustainable and stable development of the electric vehicles and improve advanced applied energy technologies.