基于小波包能量变异最值指数的RC框架损伤识别

Damage Identification of a RC Frame Based on the Maximum Variation Index of Wavelet Packet Energy Spectrum

小波包变换可以将振动信号按任意时频分辨率分解到不同频带,而各频带信号的能量变化包含着丰富的损伤信息。结合八层框架模型模拟地震振动台试验结果,对不同工况下的结构加速度时程响应进行小波包分解和重构,得到小波包能量谱。对不同频带下的小波包能量谱进行统计分析,利用小波包能量变异极值指数对框架结构的损伤进行评定。结果表明:损伤程度不同,小波包能量变异极值指数明显不同。因此可将小波包能量变异最值指数作为损伤程度的指标进行损伤识别。

Wavelet packet transform can decompose a vibration signal into different frequency bands at ran- dom time-frequency resolutions. Energy changes of signals at various frequency bands contain rich information that may be used in damage identification. With the shake table test results of an eight-story frame model, the wavelet packet is applied into the decomposition and reconstruction of the acceleration responses, and the en- ergy spectrum is obtained. The maximum variation index of wavelet packet energy spectrum was obtained through statistical analysis of the energy spectrum at different frequency bands. Experiments demonstrate under different damage conditions, there are obvions differences among the maximum variation index of wavelet pack- et energy spectrum. Therefore, the maximum variation index of wavelet packet energy spectrum can be used for structural damage detecting.