Global scaling of operational modal analysis modes with the OMAH method. (15th February 2019)

• Record Type: Journal Article
• Authors: Brandt, A.; Berardengo, M.; Manzoni, S.; Vanali, M.; Cigada, A.
• Is Part Of: Mechanical systems and signal processing. Volume 117(2019)
• Page Start: 52
• Page End: 64
• Publication Date: 2019-02-15
• Subjects: Operational modal analysis -- Mode scaling -- Modal mass -- Harmonic excitation -- OMAH; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2018.07.017
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: We present an enhanced method for scaling modes from operational modal analysis. The new method includes residual terms. The new method can handle closely coupled and repeated modes. We verify the method on two sets of experimental data. Abstract: Modal models from operational modal analysis (OMA) lack information about the modal scaling (or, modal mass). Many times, this is not a limitation. However, in some engineering applications, e.g. for structural health monitoring or computational model validation, a scaled modal model may be important. The authors of this paper recently presented a new approach for scaling modal models using harmonic excitation. The method was named the OMAH method, and it was shown that the approach is robust and reliable. However, the proposed technique relied on the measurement of the structural response between two arbitrary degrees of freedom (DOFs) for each mode. In the present paper, the method is extended to the case in which several excitation and response DOFs may be taken into account for the scaling. This is analogue to using global parameter estimation methods in classical experimental modal analysis. The proposed multiple-reference technique includes the capability of estimating residual terms to account for modes outside the frequency range of interest. The method is validated on two sets of data from real full scale structures, and the accuracy of the scaling using several response DOFs is shown to improve compared to using only a single excitation and response DOF. The proposed formulation, furthermore, allows one or more excitation DOFs as well as one or more response DOFs to be used for the scaling of the modal model. Thus, one single formulation can be used regardless of whether multiple-references are necessary, or not.