基于WPD-LMD和排列熵的结构损伤识别方法

为了直接从结构响应提取损伤敏感参数,对激励未知情况下的结构损伤模式进行识别,提出了基于小波包分解-局部均值分解方法(wavelet packet decomposition-local mean decomposition,WPD-LMD)和排列熵的结构损伤检测方法.该方法首先对结构振动响应进行小波包分解,将振动信号分解为一系列窄带信号,然后对窄带信号进行局部均值分解,能有效提取低能量分量.通过计算损伤前后分量信号的排列熵,对结构损伤进行了检测,最后通过计算测试数据和样本数据之间的相对排列熵,对损伤模式进行了识别.实验分析结果表明,所提出的方法能有效地对结构损伤进行识别.

Damage Identification Based on Wavelet Packet Decomposition-local Mean Decomposition and Permutation Entropy

To extract the damage sensitive features directly from the vibration response, a damage identification method based on the wavelet packet decomposition-local mean decomposition and the permutation entropy was proposed, without requiring the exciting force. The wavelet packet decomposition was applied to decompose the original vibration signal into a set of narrow band signals, and then, the local mean decomposition method was used to decompose the narrow band signal, which can extract the low energy components effectively. After that, the permutation entropy was obtained to detect the structural damage. The relative permutation entropy between the test data and the sample data was obtained to identify the damage pattern. The experiment results show that the proposed method can be used to identify the structural damage effectively.