摘要
针对桥梁结构变形信号中含有周期信号分量成分,提出了一种基于奇异值分解从桥梁结构变形信号中提取温度效应信号的新方法.首先将桥梁结构变形信号通过FIR低通滤波器,滤除活载效应等高频信号,得到一组低频信号,对其施加奇异值分解并剔除接近于零的奇异值,实现在结构变形中温度效应提取.实例分析结果显示:提取的温度变形值与源信号相关系数达到0.99,实现温度变形的精确分离,为从桥梁挠度监测信号中进行结构损伤识别提供了依据.
A new extraction method was proposed which is based on the singular value decomposition of the bridge structure deformation signal. Firstly, bridge structure deformed signal filtered by the FIR low-pass filter, the high frequency signal was filtered, and a set of low-frequency signal was obtained. Then, exert those signals to the singular value decomposition and remove the near-zero singular values to achieve a temperature effects of the structural deformation. The instance analysis showed the temperature deformation value with the source sig- nal extracted correlation coefficient of 0.99. Lastly, a precise separation of the temperature deformation was a- chieved, which was judged by structural damage identification in bridge deflection monitoring signal.
出处
《广州大学学报(自然科学版)》
CAS
2013年第3期38-44,共7页
Journal of Guangzhou University:Natural Science Edition
关键词
桥梁
变形
温度效应
奇异值分解
bridge
deformation
temperature effect
sigular value decomposition