A Principle for Obtaining Pragmatic Uncertainty Bounds on Modal Parameters

In the field of structural dynamics, determination of damping contributions of civil structures is a subject influenced with considerable uncertainties. Many different modal parameter estimation techniques are available to describe the dynamic behavior using only the modal response, both in frequency and time domain, but common for all is the struggle to obtain stable damping values. This paper deals with the extraction of modal parameters on simulated data through Operational Modal Analysis with focus on establishing uncertainty bounds for the damping estimates. We investigate a pragmatic principle to bound the errors, including both random and bias errors. Modal parameters are extracted by applying a low order and a high order parameter estimation method which makes it possible to establish uncertainty bounds of the total damping. These uncertainty bounds represent a lower and a higher boundary for the damping estimates corresponding to each of the identified natural modes of the test structure. Attaching a statistical confidence interval to the estimates allows for a better understanding of the uncertainties related to the damping values obtained through Operational Modal Analysis.