Application of EEMD Method and ARX Model to Identify the Characteristic Parameters of Structural Systems

碩士 === 國立臺灣大學 === 工程科學及海洋工程學研究所 === 101 === Structural system identification has been well-developed recently, mature enough to yield an effective result in simulation through proper mathematics modeling. However, the identification in practice encounters signal interference that dampens the result....

Full description

Bibliographic Details
Main Authors: Wei-Ting Hsu, 徐偉庭
Other Authors: 柯文俊
Format: Others
Language:zh-TW
Published: 2012
Online Access:http://ndltd.ncl.edu.tw/handle/17630885991771121982
Description
Summary:碩士 === 國立臺灣大學 === 工程科學及海洋工程學研究所 === 101 === Structural system identification has been well-developed recently, mature enough to yield an effective result in simulation through proper mathematics modeling. However, the identification in practice encounters signal interference that dampens the result. For better precision, mathematics modeling order would be lifted, yet the process of selecting eigenvalue becomes longer. This thesis applies one of the most important step in HHT – EEMD. EEMD as a preset uses its properties to decompose signal into IMF and filters out high-frequency signal. In addition, along with ARX model and state-space system identification theorem, the structural characteristic parameters can be identified. The EEMD method will be tested through computer simulation that includes interfered free vibration and forced vibration signal under either single or three-degrees-of-freedom. The method will later be applied onto two virtual systems: a simple and efficient virtual perpendicular beam structure crash test and seismic waves to test five-story steel frame structure of NCREE. As the result reveals, the method in the thesis provides an effective result against signal interference.