Display Full Text | Download Full Text
3.37 MB PDF file

Degree

Doctor of Philosophy, University of Akron, Electrical Engineering, 2008.

Abstract

This dissertation presents a methodology for using cycling data collected from several similar electrochemical cells to generate an aging model that predicts how the parameters in a first-principles dynamic model of a cell will change as the cell ages. Nine standard 18650 lithium-ion cells were cycled in three sets. Aging models were applied to the identified parameters of the dynamic models. These aging models were then validated by comparing their predictions with the original cycle data resulting in RMS voltage errors of less than 5% over the entire life of the cells. These aging models provide an accurate means of predicting the parameters for the dynamic cell model based on the life fraction of the cell and the maximum charging voltage. Unlike other aging models presented in the literature, the aging models presented here address the external performance of the cells. The aging model containing first-order temperature correction terms for the charge diffusion and current polarization term produced the smallest errors when compared with the original data. Incorporation of the aging model into a battery management system (BMS) will allow the BMS to better track capacity and remaining life of a cell. The methodology presented here could be applied to other cell chemistries.

Subject Headings

Chemical engineering; Electrical engineering

Keywords

electrochemical cell; lithium-ion; aging; cycling

Advisor

Tom Hartley, PhD (Advisor)

Pages

238p.

Document number: akron1226887071
Permalink: http://rave.ohiolink.edu/etdc/view?acc_num=akron1226887071

This ETD has been downloaded 3675 times (through March 2013)