基于小波包频带能量检测技术的结构损伤诊断  

Damage diagnosis of structure based on wavelet packet frequency band energy detecting technology

作　　者：何振宇[1] 张森文[2] 

机构地区：[1]华南理工大学电子与信息学院,广东广州510641 [2]暨南大学应用力学研究所,广东广州510632

出　　处：《暨南大学学报（自然科学与医学版）》2007年第5期431-434,共4页Journal of Jinan University(Natural Science & Medicine Edition)

基　　金：国家自然科学基金项目(10372036)

摘　　要：小波包变换能将振动信号按任意时频分辨率分解到不同频段,而各频段信号的能量变化包含着丰富的信息.以悬臂梁为实验对象,对不同损伤程度下的振动信号进行小波去噪后,通过4尺度小波包分解和重构,得到了其能量谱图.实验结果表明:损伤程度不同,小波包能量谱图明显不同.因此可将敏感频带的能量值作为损伤程度的特征值进行损伤诊断.The transform of wavelet packet can decompose signals into different frequency bands at random time-frequency resolutions, while the energy changes of signals at various frequency bands contain rich information that may be used in damage frequency band energy analysis combined with ver beam, and the vibration signal of it under wavelet transform. A four measure wavelet pac diagnosis. A detecting technique based on wavelet packet experiments is given. The experimental object is a cantile- different damage conditions was collected and denoised by ket is applied into the decompositions and reconstruction of the signal, and the energy spectrum is obtained. As experiments demonstrate： under different conditions of structure, there are evident differences among the energy spectrum. Therefore, the energy components can be chosen as damage indices. damage sensitive

关 键 词：小波包 振动信号 分解与重构 损伤诊断 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置] O329[自动化与计算机技术—控制科学与工程]