Hızlı tanı testleri ile sağlık durumları arasındaki istatistiksel olarak anlamlı korelasyonlar

With the popularity of electrical vehicles rising, research on Lithium ion batteries have the utmost importance. For an efficient and safe usage of these batteries, their performance and its decline have to be closely monitored. In this work, two techniques were used for predicting State of Health (SoH) for 8 Aspilsan INR18650 batteries: Electrochemical Impedance Spectroscopy (EIS) and Intermittent Current Interruption (ICI). In this thesis, firstly Lithium ion batteries and their working principle is discussed. Afterwards, background information for both EIS and ICI were given for understanding why they are suitable for application; moreover, how they can be utilized to understand the battery chemistry that causes the SoH decline. Following, how EIS and ICI were combined and used to correlate to battery SoH were presented. All 8 batteries were tested with a sequence made up of EIS, ICI, charging and discharging procedures in different States of Charge every 50 cycles. Then, the data acquired from these tests were checked against the capacities known from the cycling. For correlating these parameters, Distance Correlation (dCor) was applied as dCor reveals any type of association between two parameters, not limited to linearity. Afterwards, to verify the data acquired, the errors for ICI parameters were investigated. Moreover, Kramers-Kronig test was applied to all the EIS data acquired to validate the quality of the data. To add, binning using the ICI parameters were used to showcase the prediction power of using ICI parameters acquired. In the end, it is explained how and why capacity of this set of batteries can be predicted with 1.8\% error range using ICI test at specific conditions in a relatively short amount of time.