The objective of this thesis is the design of several control strategies for a hybrid power generator composed by a fuel cell and an auxiliary energy storage source. The Linear Matrix Inequalities (LMI) tools are extensively used in this dissertation as a solution to the mutivariable robust control problem. As a first approach, the control methodology is consecrated to the electrical power management sub-system of the fuel cell. Different strategies are proposed to control the hybrid boost power converter configuration for DC voltage applications. The methodology is extended to AC islanded applications considering the additional control of a voltage inverter. The validation on a dedicated test-bench, of a part of the proposed control strategies, is presented. In a second approach, the control of the air supply system is addressed. The management of the air dynamic entering the fuel cell is assured by the control of the air flow of a compressor. The air supply sub-system is controlled to keep a desired oxygen excess ratio, this allow to improve the fuel cell performance. An introduction to the control of Linear Varying Parameter (LPV) systems is also presented. Robustness analysis studies are performed, these robust properties are contrasted with several classic control strategies, demonstrating the advantage and the importance of multivariable robust methodologies.
