Short-term model-based damping prediction for fluid-loaded structures

Research output: Book/ReportPh.D. thesis

39 Downloads (Pure)

Abstract

Structures vibrate when they are excited. Some structures are excited by wind and waves, some by moving traffic, and some by something else. The resulting structural vibrations are called the dynamic response. The dynamic response to a given excitation is governed by the modal parameters of the structure, which include natural frequencies, damping ratios, and mode shapes. Characterizing the dynamic response is crucial for ensuring the structural integrity of various engineering structures, such as bridges, aircraft, and wind turbines.

Various structures are influenced by varying environmental and operational conditions, which can cause the effective linear modal parameters, and especially modal damping, to vary considerably over short time periods. This thesis develops and tests methods for estimating short-term damping ratios and natural frequencies from vibration response measurements. Operational Modal Analysis (OMA) refers to modal parameter estimation based on response measurements of a system excited by an unmeasured load. Performing OMA involves addressing two fundamental challenges: distinguishing between modes in measurements containing the response of multiple modes, and accounting for the effect of the unmeasured excitation. The latter constitutes the primary challenge of short-term damping estimation. Conventional OMA methods can average out the excitation effect from the response effectively. However, they assume time-invariant modal properties and require long measurements, limiting their capabilities to track short-term damping variations.

The work is divided into two main parts. First part considers damping estimation from free response measurements, and second part considers damping estimation from forced response measurements. The former case is simpler, as there is no unmeasured excitation to be accounted for. The first part compares three methods for short-term and one for timeinvariant damping estimation. The performance of the methods is tested using simple simulated responses with prescribed modal parameters, and practical applicability is tested using responses from wind turbine impulse tests.

The second part introduces a novel OMA method for short-term damping estimation for structures influenced by environmental and operational variability. The proposed method combines physics- and data-driven system identification, leveraging prior domain knowledge to achieve short-term damping ratio and natural frequency estimation. The method is tested on two simulated and two experimental application examples. It appears promising for short-term damping estimation, given sufficient data and a suitable model structure.
Original languageEnglish
Place of PublicationKgs. Lyngby
PublisherTechnical University of Denmark
Number of pages135
ISBN (Electronic)978-87-7475-783-2
Publication statusPublished - 2024
SeriesDCAMM Special Report
NumberS351
ISSN0903-1685

Fingerprint

Dive into the research topics of 'Short-term model-based damping prediction for fluid-loaded structures'. Together they form a unique fingerprint.

Cite this