基于小波分析的结构损伤信号奇异性检测

结构发生损伤时,其动力特性会发生突然变化,因此会产生奇异信号。Lipschitz指数是表征信号局部奇异性特征的一种度量,可以用来识别结构损伤的发生。小波变换是一种时频域分析方法,具有多分辨率分析的特点,是奇异性信号分析的合适工具。在突变时刻小波系数出现模极大值,通过小波系数模极大值求得的Lipschitz指数可以作为衡量突变程度的指标,由此可以准确识别结构发生损伤的时刻。通过对一空间结构损伤前后加速度信号的奇异性进行分析,验证本文的相关理论。

SINGULARITY DETECTION FOR STRUCTURAL DAMAGED SIGNALS BASED ON WAVELET ANALYSIS

When damages occur on structure,the structural dynamic characteristics change abruptly and produce singular signals.Lipschitz exponent,which is the measuring of local singularity of corresponding signals,could be used to detect whether structural damage has happenned.Wavelet transform is a analysis method both for time and frequency domain,and it adapted well to singularity analysis for its multi-resolution.The modulus maximum of wavelet coefficients appear when the signal change,hence Lipschitz exponent could be computed to judge the mutation degree and the time when damages occur.The method is validated by the singularity analysis for the acceleration signals of a space structure.