随机激励下结构突发性损伤检测方法的比较

Comparative on Abrupt Damage Detection Methods for Structure under Random Excitation

受到随机激励的结构在服役过程中突发损伤时,会导致结构局部刚度的突然下降,从而引起结构振动响应的突然变化,从振动响应中提取信息来检测结构的突发损伤,可实现结构损伤的在线监测。通过一个随机激励下框架结构模型的突发损伤在线检测实验,对比研究了经验模分解方法以及高通滤波方法在结构突发损伤检测中的可行性和有效性。结果表明:结构的突发性损伤会使损伤部位相邻结构的振动响应幅值发生突变,并使振动响应信号在损伤时刻附近出现高频成分。采用EMD分解方法的间歇检验滤波和常规高通滤波方法都可以将响应信号中的高频成分分离出来,并以此为指标进行突发损伤的在线检测,而且可以准确检测出突发损伤发生的时刻和位置。

For an in-service structure under of local stiffness, which vibration response. It is can be reflected by random excitation, abrupt damage usually induces a sudden reduction structural vibration response, and hence it results in obvious change of possible to extract damage indicator from vibration response to realize online monitoring for abrupt damage by means of signal processing. In this paper, a comparative study for empirical mode decomposition （EMD） method and high pass filtering method is conducted on their feasibility and effectiveness in abrupt damage detection using damage detecting tests of a frame structure model under random excitation. The results show that when an abrupt damage occurs to the structure, abrupt change also occurs to the vibration response of the sub-struc- ture near the damage position, and high frequency response component will appear in the response when damage is occurred. The intermittency check of EMD is used as a filtering measure to extract the high frequency component and also used as index to detect the time and location of abrupt damage.