Subspace-based fault detection algorithms for vibration monitoring

We address the problem of detecting faults modeled as changes in the eigens tructure of a linear dynamical system. This problem is of primary interest for structural vibration monitoring. The purpose of the paper is to describ e and analyze new fault detection algorithms, based on recent stochastic su bspace-based identification methods and the statistical local approach to t he design of detection algorithms. A conceptual comparison is made with ano ther detection algorithm based on the instrumental variables identification method, and previously proposed by two of the authors. (C) 1999 Elsevier S cience Ltd. All rights reserved.