DAMAGE identification with fusion of estimates from covariance of IRF in different frequency bands. (1st December 2019)
- Record Type:
- Journal Article
- Title:
- DAMAGE identification with fusion of estimates from covariance of IRF in different frequency bands. (1st December 2019)
- Main Title:
- DAMAGE identification with fusion of estimates from covariance of IRF in different frequency bands
- Authors:
- Li, X.Y.
Ye, X.Q.
Law, S.S. - Abstract:
- Highlights: The analytical formula of CoI of every mode is derived. CoI sensitivity is obtained analytically. CoI and WPE changes and Bayesian fusion are used without need of analytical model. Numerical study and Experimental verification are performed. Abstract: Structural health monitoring is becoming more important for infrastructure worldwide in the past few decades. An accurate analytical model of the monitored engineering structure is, however, usually not available. This report presents a new non-model based damage detection method from variation of the auto (cross)-covariance of impulse response function within selected frequency bandwidths of responses obtained from the intact and damaged structural states. The covariance is computed via the wavelet packet of the responses. Bayesian fusion is applied to all the damage probability vectors obtained from the covariance under different excitations and different measurement noise. A seven-storey steel frame structure with scenarios of both single and multiple damages and complete or incomplete measurements is numerically analyzed. A similar frame was later test in the laboratory with hammer excitation. The sensors nearest to the local damages exhibit the biggest damage probability. Both numerical and experimental results show that the proposed method is effective and promising for damage detection of practical engineering structures.
- Is Part Of:
- Mechanical systems and signal processing. Volume 134(2019)
- Journal:
- Mechanical systems and signal processing
- Issue:
- Volume 134(2019)
- Issue Display:
- Volume 134, Issue 2019 (2019)
- Year:
- 2019
- Volume:
- 134
- Issue:
- 2019
- Issue Sort Value:
- 2019-0134-2019-0000
- Page Start:
- Page End:
- Publication Date:
- 2019-12-01
- Subjects:
- Damage identification -- Wavelet packet energy -- Covariance -- Bayesian fusion -- Acceleration
Structural dynamics -- Periodicals
Vibration -- Periodicals
Constructions -- Dynamique -- Périodiques
Vibration -- Périodiques
Structural dynamics
Vibration
Periodicals
621 - Journal URLs:
- http://www.sciencedirect.com/science/journal/08883270 ↗
http://firstsearch.oclc.org ↗
http://firstsearch.oclc.org/journal=0888-3270;screen=info;ECOIP ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.ymssp.2019.106327 ↗
- Languages:
- English
- ISSNs:
- 0888-3270
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 5419.760000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 12166.xml