Order Reduction of a Distributed Parameter PEM Fuel Cell Anode Gas Channel Model

Distributed parameter modeling is required to accurately consider space variations, which are important regarding the performance and durability of the Proton Exchange Membrane Fuel Cells (PEMFC) [1-3]. However, the number of differential and algebraic equations (DAE) obtained from the discretization of a set of partial differential equations (PDE) is very large, and this not only slows down the numerical simulations, but also complicates the design of online model-based controllers. The inclusion of complex DAE models within model-based control schemes requires a previous simplification. A method to simplify complex models consists of reducing the order while preserving the relationship between certain input and output variables, determined from the control objectives. These Model Order Reduction (MOR) techniques have been extended to DAE systems [4]. This work focuses on obtaining an order reduced model, from a PEMFC anode gas channel PDE model, which incorporates the effects of distributed parameters that are relevant for the proper functioning and performance of PEMFC. The original model is an in-house MATLAB® code, flexible enough to manipulate the underlying model equations and apply MOR techniques. The obtained order-reduced model is suitable to perform numerical simulations and design efficient controllers for the original nonlinear PDE model.