System identification techniques and Hinfinity control strategies applied to automotive PEM fuel cell propulsion systems.

摘要

The problem of controlling the net output power of an automotive Proton Exchange Membrane (PEM) fuel cell propulsion system has been investigated in this work. Observed input-output data System Identification techniques were utilized to obtain a set of Single Input Multi Output (SIMO) models as well as a set of Multi Input Multi Output (MIMO) models for the system. One model out of each set was selected to represent the PEM fuel cell propulsions system via the process of model validation. The order of each selected model was reduced using model reduction techniques. A robust Hinfinity loop shaping controller was designed to set the level of the net output power of the PEM fuel cell system to follow a reference input corresponding to load demand. The SIMO model was employed in the design of such control system. The performance of the compensated system (linear and nonlinear) was deemed very satisfactory. Another Hoo suboptimal controller was designed using the MIMO model to control the output power of the PEM fuel cell system following a change in current demand. The performance of the compensated system (linear and nonlinear) was also deemed very satisfactory.