(592h) Modeling and Optimization of Polymer Electrolyte Fuel Cells

Authors: 
Biegler, L. T., Carnegie Mellon University
Jhon, M. S., Carnegie Mellon University


We develop a generalized modeling and optimization framework that includes rigorous computational fluid dynamics and catalytic reaction phenomena, suitable for analysis of polymer electrolyte fuel cell (PEFC) systems.

Very recently [1], we constructed a multi-dimensional, multi-physics PEFC model that accounted for the transport processes within the gas channels (GC), gas diffusion layer (GDL), and the polymer electrolyte membrane (PEM). These set 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 possessing multiple experimental operating conditions.

In this paper, we improved our previous integrated PEFC model substantially by incorporating a cathode catalyst layer (CL) model in depth. We adopted an agglomerate CL model, and recasted it to a compact form that is easily convertible to a pseudo-homogenous CL model, for optimization. As a consequence, our new PEFC model captures the essential transport processes and accounts for species, proton, and electron transport within the GC/GDL/PEM as well as reaction/transport processes within the CL. The resulting nonlinear partial differential algebraic equations are fully discretized using a finite difference method and the nonlinear program is linked to IPOPT.

Before solving integrated PEFC model along with optimization, we analyzed our CL model. In our preliminary study, we performed Platinum (Pt) minimization for a specified cell performance as well as, obtained optimal Pt distribution along CL width, which is obtained by solving multi-zone partial differential equation-constrained optimization problem. In this preliminary work, CL was coupled with nearby regions, e.g., GDL and PEM. We intend to optimize the performance of fully integrated PEFC model within both CL and PEM regions. Specifically we will examine Pt and Nafion distributions within CL, and also obtain optimal chemical structures for PEM.

Our proposed framework provides a robust and fast solution methodology, and is general and straightforward for modeling extensions, for addressing other critical issues in the fuel cell technology that require large-scale optimization methodologies.

References:

1. Jain P, Biegler LT, Jhon MS. Parametric Study and Estimation in CFD-based PEM Fuel Cell Models. AICHE J, to appear.

2. Wächter A, Biegler LT. On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming. Math Program. 2006;106:25-57.