(491k) Inverse Modeling of Solid Oxide Fuel Cells

Kim, S., Carnegie Mellon University
Biegler, L. T., Carnegie Mellon University
Jhon, M. S., Carnegie Mellon University

We develop a novel inverse modeling framework for solid oxide fuel cell (SOFC) systems via a parameter estimation strategy. In our previous work [1], we constructed a multi-dimensional, multi-physics polymer electrolyte fuel cell (PEFC) model that accounted for the transport processes within the gas channel, gas diffusion layer, and the polymer electrolyte membrane sub-components of the PEFC, whereas the catalyst layer sub-component was inversely modeled through parameter estimation strategy. The resulting sets of partial differential algebraic equations were linked to our in house state-of-the-art interior point optimization algorithm, IPOPT [2] to solve challenging parameter estimation problems with multiple experimental operating conditions. We particularly explored the water transport and distribution characteristics, and overall system performance through parametric studies on the system.

In this study, we examine a simple tubular SOFC model (instead of the accurate modeling, which entails solution to highly coupled 3-D nonlinear partial differential-algebraic equations) that accounts for the essence of the SOFC assembly. Our 1-D/2-D model accounts for mass, momentum, energy and reaction terms, and through a parametric study, we found that mass transport and temperature effects dominate the fuel cell performances. A parameter estimation non-linear program is formulated using a maximum likelihood objective that minimizes the error between the model and experimental value of cell output voltage, in order to determine detailed diffusional characteristics including nano/micro scale pore size (Knudsen effects) as well as non-isothermal effects on overall system performance. Our novel framework reported here provides a robust and fast solution methodology, and is planned for modeling extensions and addressing other critical issues in SOFC technology that require large-scale simulations.


1. Jain P., Biegler L.T., and Jhon M.S., ?Parametric Study and Estimation in CFD-based PEM Fuel Cell Models,? AIChE J, 54 (8), 2089 (2008).

2. Wächter, A., Biegler, L. T., ?On the Implementation of an Interior-Point Filter Line-Search Algorithm for Large-Scale Nonlinear Programming,? Math Programming, 106, 25 (2006).