MERGING SENSOR DATA FROM MULTIPLE MEASUREMENT SET-UPS FOR NON-STATIONARY SUBSPACE-BASED MODAL ANALYSIS

Processing sensor data, from multiple non-simultaneously recorded measurement set-ups, for structural analysis is often achieved by merging identification results obtained from records corresponding to different sensor pools. Since pole matching and eigenvector gluing might not be consistent in some cases, the question arises to merge the data first and then process them globally. Subspace identification algorithms have proven efficient for performing output-only modal analysis of mechanical systems subject to uncontrolled, unmeasured, and non-stationary excitation. The purpose of this paper is to investigate as to how subspace-based output-only modal analysis algorithms can be adapted to handle the multi-patch measurements set-up. Numerical results, obtained on both laboratory and real application examples, are reported, which show the relevance and usefulness of the proposed algorithm.