Improvement of a stepwise primary diagnosis method of a beam structure using a force identification technique (Proposition of virtual measurement)

Abstract

Machine condition monitoring and diagnosis have become increasingly important. It is generally considered that larger numbers of sensors will supply useful information for diagnosis, although the number of sensors is limited in actual applications. In this study, therefore, a method of virtual measurement at many points of the structure, i.e., an approximation method of responses at non-measured points, is proposed, and the diagnostic accuracy is checked. To this end, two methods are proposed. A method is based on the modal analysis using the measured displacements, in which the displacement of the overall structure is approximated using natural vibration modes whose number is the same as that of sensors. The other method is also based on the modal analysis, though the displacements used for the approximation are modified by the displacement ratio between the actual measurement and the simulated one. The validity of each method is checked using the experimental data. From the results, it was recognized that the proposed method based on the displacement ratio is useful for virtual measurements.

Keywords

• model-based diagnosis approach
• inverse analysis
• force identification
• virtual measurement
• modal analysis