Date of Award

2011

Document Type

Campus Access Dissertation

Department

Chemical Engineering

First Advisor

Ralph E White

Abstract

In this research work, a variety of thermal models for lithium-ion cells and battery systems with different materials and configurations were developed by incorporating the cell energy balance equations and the temperature-dependency of cell parameters into the different electrochemical models.

Among these models, the single particle (SP) thermal model has high time efficiency in computation due to the relatively small number of equations. The single particle thermal models were firstly developed for single cells with different chemistry under isothermal conditions. The simulated cell voltage and temperature were fit with experimental data, and values for certain parameters were obtained from curve fitting. To solve the solid phase diffusion equation in single particle models, a new approximated eigenfunction solution was developed; and with this method, the solution of the partial differential equation in spherical particles can be accurately approximated by only a few ODE s for eigenfunctions.

By coupling the single particle thermal models for several individual isothermal cells with the series/parallel circuit correlations and the inter-cell heat transfer conditions, these thermal models where extended and applied to battery packs with thermal management designs. The simulation results from the battery thermal model were also validated by experimental data.

The multi-scale multi-dimensional (MSMD) thermal models were developed to study the temperature distribution in the bodies of large cells. In the multi-scale model, heat transfer occurs in the cell domain and the distributed heat generation rate is obtained from the electrochemical model in the electrode domain. The distribution of current in the cell domain is determined by the limiting electrical condition. The single-particle model with polynomial approximation was coupled with heat conduction in 3D cell domain and the model predictions of temperature at different locations were validated by experimental data. The pseudo two-dimensional (P2D) model which is much more complicated than the single particle thermal model was used to replace the single particle model in the electrode domain. To avoid the significant computational difficulties, the P2D model is not fully distributed in the 3D domain; instead, the P2D model is only solved at the average temperature and a linear approximation method is applied to calculate for the distributed heat generation rate.

Share

COinS