Modelling and diagnosis of solid oxide fuel cell (SOFC)

Author: 
Date created: 
2021-04-20
Identifier: 
etd21366
Keywords: 
SOFC
Pseudo-2D
3D
Parameter estimation
Diagnosis
Machine learning
Abstract: 

The development of mathematical models and numerical simulations is crucial to design improvement, optimization, and control of solid oxide fuel cells (SOFCs). The current study introduces a novel and computationally efficient pseudo-two-dimensional (pseudo-2D) model for simulating a single cell of a high-temperature hydrogen-fueled SOFC. The simplified pseudo-2D model can evaluate the cell polarization curve, species concentrations along the channel, cell temperature, and the current density distribution. The model takes the cell voltage as an input and computes the total current as an output. A full-physics three-dimensional model is then developed in ANSYS Fluent, with a complete step-by-step modeling approach being explained, to study the same cell with the identical operating conditions. The 3D model is validated against the other numerical and experimental studies available in the literature. It is shown that although the pseudo-2D solution converges significantly faster in comparison with the 3D case, the results of both models thoroughly match especially for the case of species distributions. The simplified model was then used to conduct sensitivity analysis of the effects of multi-physiochemical properties of porous electrodes on the polarization curve of the cell. A systematic inverse approach was then used to estimate the mentioned properties by applying the pattern search optimization algorithm to the polarization curve found by the pseudo-2D model. Finally, nine different input parameters of the model were changed to find the hydrogen distribution for each case, and a huge dataset of nearly half a million operating points was generated. The data was successfully employed to design a novel classifier-regressor pair as a virtual hydrogen sensor for online tracking of hydrogen concentration along the cell to avoid fuel starvation.

Document type: 
Thesis
Rights: 
This thesis may be printed or downloaded for non-commercial research and scholarly purposes. Copyright remains with the author.
File(s): 
Supervisor(s): 
Krishna Vijayaraghavan
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.
Statistics: