Multivariable control design for the water gas shift reactor in a fuel processor

The water gas shift (WGS) reactor is an essential component of the fuel processor in fuel cell power plants. The WGS reactor combines CO and H/sub 2/O in the reformate stream to produce H/sub 2/ and CO/sub 2/ in a mildly exothermic, equilibrium limited reaction. Proper regulation of the steam-to-carbon ratio and the operating temperature is critical to achieving adequate CO conversion during transients. We consider the control problem for the high temperature shift reactor of a catalytic partial oxidation (CPO) based natural gas fuel processor in a PEM fuel cell power plant. The manipulated variable is the water injection rate and the measurements available are the inlet and exit temperatures of the reactor and the reformate flow from the CPO reactor. The objective is to maintain proper water content in the inlet and CO conversion in the reactor. A linear state space model is obtained by linearizing a nonlinear dynamic model of the system around an operating point of interest. A multivariable controller is designed using the LQR method. We compare various control architectures by selectively dropping the terms in the multivariable controller. It is shown that the reactor exit temperature measurement improves the observability of the system but has negligible influence on the control performance. Such multivariable control analysis provides a systematic framework for the evaluation of alternative control architectures and the impact of sensors.