Gain-scheduled control for an automotive traction pem fuel cell system

To be practical in automotive traction applications, fuel cell systems must provide power output levels of performance that rival that of typical internal combustion engines. In so doing, transient behavior is one of the keys for success of fuel cell systems in vehicles. From a model-based control perspective, regulation of the fuel cell system through transients is critical, where the response of a fuel cell system depends on the air and hydrogen (flow and pressure regulation) and heat and water management. The focus of this paper is on the air/fuel supply subsystem in tracking an optimum variable pressurization and air flow for maximum system efficiency during load transients. The control-oriented model developed for this study considers electrochemistry, thermodynamics, and fluid flow principles for a 13-state, nonlinear model of a pressurized fuel cell system. For control purposes, a model reduction is performed by converting some of the model dynamics to simple algebraic relationships. A single reference input, the power demanded by the user, is utilized to produce a corresponding reference air flow and back-pressure valve opening, after passing through a static calculation and a tabulated map. Because of the complexity of the full nonlinear model (used in simulation as the truth model), where several maps are used rather than functional forms, two different control techniques are examined separately, each using a feedforward component. The first technique uses an observer-based linear optimum control which combines a feed-forward approach based on the steady state plant inverse response, coupled to a multi-variable LQR feedback control. An extension of that approach, for control in the full nonlinear range of operation, leads to the second technique, nonlinear gain-scheduled control.