Fuzzy gain adaptive observers for robust fault detection in dynamic process
Abstract
A design method of a fuzzy gain adaptive observer, based on the conventional regulator design principle, is proposed in this paper. The observer gain is adapted using rules similar to those of fuzzy regulator through an appropriate choice of linguistic terms. The proposed method is well adapted for dynamics processes with uncertain model parameters. Numerical results are provided for performance analysis in the case of a nonlinear process and that of a nonlinear process in which some parameters have varied.