A novelty detection method to diagnose damage in structures: An application to an offshore platform

被引:0
作者
Surace, C [1 ]
Worden, K [1 ]
机构
[1] Politecn Torino, Torino, Italy
来源
PROCEEDINGS OF THE EIGHTH INTERNATIONAL OFFSHORE AND POLAR ENGINEERING CONFERENCE, VOL 4 | 1998年
关键词
damage detection; neural networks; offshore platforms;
D O I
暂无
中图分类号
U6 [水路运输]; P75 [海洋工程];
学科分类号
0814 ; 081505 ; 0824 ; 082401 ;
摘要
The problem of diagnosing damage to offshore structures is of great importance since damage is accumulated continually during service life due to environmental forces such as waves, wind, current and seismic actions. Damage tends to alter the modal properties of the structural system such as the natural frequencies, and one approach to its detection currently gaining considerable research attention uses the vibrational response to diagnose the state of the structure. However application to offshore structures is made particularly difficult since their natural frequencies undergo significant alteration over time as a result of various factors such as changes of the deck mass due to the quantity of oil stored. As a consequence, in order to be detected, damage must produce greater variations in the natural frequencies. To overcome this problem the approach of 'novelty detection' is proposed. The article describes this method and its application to a numerical simulation of a steel offshore platform. Using the results obtained it has been possible to analyse the effects of measurement noise on the performance of the method and assess the sensitivity of the method to structural alterations.
引用
收藏
页码:64 / 70
页数:3
相关论文
empty
未找到相关数据