Abstract:

A set membership identification approach is provided to deal with the problem of fault detection of systems with models not precisely known. Compared with the former model based approaches of fault detection, the present one does not require the a priori knowledge about the true parameters of models. Simulations show the quickness and the effectiveness of this approach.

Key words: fault det ect ion, set membership ident ification, r obust ident ification, U BB err or