Constrained model-based control for robust operation of a solid oxide fuel cell.

Abstract

Over the years, the Solid Oxide Fuel Cell (SOFC) has grown in popularity due to its high efficiency and fuel flexibility, low maintenance requirements and environmentally friendly features. Its durability and performance can be substantially enhanced through the implementation of optimal control strategies. However, controlling the SOFC is challenging due to its nonlinear behavior and the need to manage its operational constraints simultaneously. To come up with an effective control algorithm for a SOFC, it is necessary to understand its dynamic behavior. In this work, a Wiener model is developed for control purpose. This mathematical representation enables the use of an identification algorithm to estimate unknown parameters. Furthermore, the control-oriented model uncertainty is characterized and incorporated as bounds on the model parameters. The resulting model is integrated into a Model Predictive Control (MPC) framework, designed for both regulation and optimization objectives. This control scheme ensures robust constraint satisfaction for the whole set of characterized uncertain models. The simulation results show the benefits of the proposed model-based controller both for SOFC operation and performance enhancement.