期刊
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
卷 9, 期 1, 页码 59-74出版社
SAGE PUBLICATIONS LTD
DOI: 10.1177/1475921709341017
关键词
damage detection; data reduction; feature subset selection; energy-efficient sensor network; wavelet shrinkage
资金
- NSF [CMMI 0427878, CMMI 0428210]
While wireless sensors are increasingly adopted in various applications, the need of developing data reduction methods to alleviate data transmission rate issue between the sensors and the data interpretation unit becomes more urgent. This article presents a new data reduction method for sensors used in structural health monitoring application. Our goal is to achieve an effective data reduction capability while maintaining adequate power for damage detection. We propose to establish an explicit measure of damage detection capability for the features in the response signals and use this measure to select the subset of the features that balance between the degree of data reduction and the damage detection capability. We also explore a computationally efficient procedure searching for the best subset of the features. This new method is tested on experimentally obtained Lamb wave signals for beam damage detection. Performance comparisons with respect to the existing methods demonstrate the strength of the proposed method.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据