Skip to main navigation Skip to search Skip to main content

Structural system identification based on frequency-time domain information

  • Yong Ding*
  • , Guoshan Xu
  • , Qi Lin
  • , Bin Wu
  • *Corresponding author for this work
  • Harbin Institute of Technology

Research output: Contribution to journalArticlepeer-review

Abstract

The information in time domain and frequency domain can be used for structural model updating. A frequency-time domain method is proposed for structural model updating and external excitation identification in this paper. First, the parameters in frequency, such as frequency and mode shape, are calculated from the acceleration and structure model updating is conducted with these parameters. Second, a state space force identification method is proposed. The model updating is also updated in the second stage. A regularization method is utilized in the force identification and constraint optimization method is applied in the structural model updating. IIRS condensation method is used for computational effort. In the numerical simulations, a frame structure is studied. The structural model is updated with white noise excitation. The structure damage is detected with the incomplete measured information during the earthquake excitation. A 14-storey shear wall building is investigated to validate the proposed method for force identification and damage detection. The structure response is reconstructed with small error. Promising identification results can be obtained with the proposed method even when the measured response is polluted with 10% noise.

Original languageEnglish
Pages (from-to)15-19
Number of pages5
JournalTumu Gongcheng Xuebao/China Civil Engineering Journal
Volume45
Issue numberSUPPL.1
StatePublished - Jul 2012

Keywords

  • Force identification
  • Mid-story isolation
  • Parameter identification
  • Structural model updating
  • Substructure

Fingerprint

Dive into the research topics of 'Structural system identification based on frequency-time domain information'. Together they form a unique fingerprint.

Cite this