Approximate Dynamic Programming based control for Water Gas Shift reactor

Abstract The Water Gas Shift (WGS) reactor is an important component in a fuel processing system. It is necessary to maintain low CO levels during steady state and transient operation of fuel processing system when used onboard in a vehicle. The WGS reactor plays an important role in this regard. The focus of this paper is a model based controller for the regulation of CO level in the presence of the possible disturbances from upstream temperature and flow rate. The use of model based controllers such as nonlinear Model Predictive Control (nMPC) suffers from large computational load. On the other hand Approximate Dynamic Programming (ADP) based controller will result in better performance with lower computational load. In this paper, a non-adiabatic WGS reactor modeled by hyperbolic partial differential equations is considered. The performance of the ADP is illustrated through numerical simulations.