Robust fault detection using zonotope-based set-membership consistency test

This article proposes a robust fault detection method that takes advantage of a recently proposed set-membership identification procedure based on zonotopes for systems linear in the parameters. It is shown that consistency checks indicating faults can be performed in a natural manner with it zonotope description of the feasible parameter set. A zonotope-based fault detection algorithm that is able to handle systems with invariant parameters, with parameter variation bounded between samples and with unbounded variation is presented. Finally, two application examples are given, which demonstrate how the algorithm works on a simulated process (a four-tank system) and a real application example (a section of a sewer network). Copyright (C) 2008 John Wiley & Sons, Ltd.