A Bayesian estimator of operational modal parameters for linear time-varying mechanical systems based on functional series vector TAR model

Operational modal analysis of time-varying mechanical dynamic systems is a useful but challenging task. This paper presents a Bayesian estimator for modal parameters of linear time-varying mechanical dynamic systems in the framework of the functional-series vector time-dependent autoregressive (FS-VTAR) model with output-only measurements. The proposed Bayesian estimator cannot only give the modal parameters with the estimation of mean value, but also supplies the uncertainty with the creditable interval estimation. A series of numerical examples have illustrated the advantages of the proposed Bayesian estimator against the overestimate, the better performance for the short data and the capability of supplying uncertainty of estimates. An experimental example validates the proposed estimator further.