A Reference Model LMI Approach to Fault Detection and Isolation

摘要

An H_infinity approach for the robust fault detection and isolation filter design is presented in this paper, using a linear matrix inequality formulation. To detect and isolate the faults, an estimation of the input fault signal vector is implemented, and a reference model is used in order to shape the desired performance of the filter. The filter problem is formulated using a general linear fractional transformation framework, and the solution is based on the Bounded Real Lemma applied to the closed-loop system. The conditions for the existence of the solution are presented, forming a convex problem for the full-order filter. These conditions are expressed in terms of linear matrix inequalities with matrix rank constraints, and a parameterization of all admissible filters, which correspond to a feasible solution, is provided. Finally, the methods are demonstrated using a structural system simulation example, which include a faulty actuator and external disturbances.