Identification of modal parameters from measured input and output data using a vector backward auto-regressive with exogeneous model

This paper proposes a modal identification system based on vector backward auto-regressive with exogeneous (VBARX) model. The model is an extension of vector backward auto-regressive (VBAR). Both the backward models offer the same benefits in selecting physical modes, since both can provide a determinate boundary that separates system modes from spurious modes. The VBAR model can identify the structural parameters from only output data. In some circumstances, if the input data are available, the extended model, VBARX model, provides an additional advantage over the VBAR model. In this study, an equivalent state-space model derived from measured input and output data is transformed from the VBARX model. Consequently, the structural modal parameters can be estimated accurately using the equivalent state-space model. Two examples of modal identification are presented to demonstrate the availability and effectiveness of the proposed VBARX method. (1) Numerical data simulated in a 3-d.o.f. dynamic system with various types of input data and various noise levels. (2) Experimental data obtained from the National Center for Research on Earthquake Engineering (NCREE) in Taiwan, concerning five-story 12-scale steel structure under a shaking table test.