基于小波分析的基桩内部应力监测数据异常辨识  

Identification of Abnormal Attribute for Observed Stress in Foundation Pile Based on Wavelets

作　　者：李筱艳[1] 阮怀宁[2] 陈志坚[1] 

机构地区：[1]河海大学地球科学与工程学院,江苏南京210098 [2]河海大学岩土工程研究所,江苏南京210098

出　　处：《岩土工程技术》2009年第6期292-295,共4页Geotechnical Engineering Technique

基　　金：国家"十一五"科技支撑资助项目(2006BAG04B01)

摘　　要：小波分析在时频域具有良好的局部化特征,采用小波分解方法可简单、快捷地计算出数据序列的奇异性指数,以检出数据序列中的随机突变信号。通过试验数据验证,奇异性指数对随机突变信号的检出是正确有效的。根据多传感器监测系统中突变信号的分布规律,进行异常属性自动辨识。研究说明基于小波分析的异常属性识别是一种新颖有效的方法。Wavelet has good localization characteristic in time and frequency domain. Strangeness index of data serial can be calculated simply and conveniently with wavelet to detect stochastic jump value. The detected method is correct and effectual by validating test data. Abnormity attribute can be identified automatically, according to distribution rules of jump value of multi-sensors. It states that identification of abnormality attribute based on wavelet analysis is a novel and valid method.

关 键 词：异常属性 应力监测 奇异性指数 

分 类 号：TU473.1[建筑科学—结构工程]