Fault Estimation Based on Generalized Learning Observer for a Class of Nonlinear Algebro-Differential Parameter-Varying Systems

Abstract

This paper presents the design of a generalized learning observer (GLO) structure for simultaneous estimation of variable states and actuator faults in a wider class of systems known as descriptor nonlinear parameter varying (D-NLPV) systems. This generalized structure provides additional degrees of freedom in the observer design to improve robustness and reduce the convergence time for fault estimation. Its design is obtained in terms of a set of linear matrix inequalities. The performance of the proposed methodology is evaluated in the model of a heat exchanger with two countercurrent cells with actuator faults.