Hits:

Indexed by:期刊论文

Date of Publication:2010-09-01

Journal:STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL

Included Journals:SCIE、EI、Scopus

Volume:9

Issue:5

Page Number:433-445

ISSN No.:1475-9217

Key Words:composite fuel tank; experiment; vibration based; damage detection

Abstract:Demand of composite fuel tank for space vehicles keeps increasing globally in the aerospace industry, and requires of a mean to detect the structural damage or predict its service life. As important and valid tools, in the past years, lots of nondestructive evaluation techniques have been developed, and they are carried out according to a regular time schedule. The structure has to be taken out of service during each inspection, causing serious financial implications for the user. That is why much research effort is now focused on real-time monitoring techniques. Optical grating fiber attracts more and more attention due to its excellent sensing and mechanical performances and ability of monitoring online. But it still can only measure the local deformation, but not the damage that is far away from the location of optical fiber sensors. Statistical model-based vibration-based monitoring methods have proved to be useful for the detection and localization of structural damages which might occur far from the location of the sensors. An experimental investigation on the modal properties and dynamic response of composite fuel tank filled with fluid under external excitation in the transverse direction are reported, and a vibration signal, which was obtained when the composite fuel tank filled with water was under impulsion until it failed. Proposed in this study is a method to detect the damage of the composite fuel tank with the scarce gradual damaged vibration signal. The experiment setup is also described in detail with inclusion of modeling a damaged situation of a structure. The damage detection algorithm is developed by applying a residually generated form of a subspace-based covariance-driven identification method and the vibration signal analysis. Results of the modal identification and the damage detection are reported in this article.