H∞ filtering for fuzzy singularly perturbed systems

This paper considers the problem of designing H. filters for fuzzy singularly perturbed systems with the consideration of improving the bound of singular-perturbation parameter epsilon. First, a linear-matrix-inequality (LMI)-based approach is presented for simultaneously designing the bound of the singularly perturbed parameter E, and H. filters for a fuzzy singularly perturbed system. When the bound of singularly perturbed parameter E is not under consideration, the result reduces to an LMI-based design method for H. filtering of fuzzy singularly perturbed systems. Furthermore, a method is given for evaluating the upper bound of singularly perturbed parameter subject to the constraint that the considered system is to be with a prescribed H. performance bound, and the upper bound can be obtained by solving a generalized eigenvalue problem. Finally, numerical examples are given to illustrate the effectiveness of the proposed methods.