自然激励技术在时延相关白噪声激励系统中的应用

工况模态分析(operational modal analysis,OMA)是一种只基于输出数据进行系统模态参数辨识的技术。由于系统所受激励未知,通常假设其为互不相关的白噪声。自然激励技术(natural excitation technique,NExT)正是一种基于该假设的从响应数据中提取自由衰减信号的高效方法。然而对于真实工况,系统所受激励之间互不相关的假设过于严格,例如行驶的车辆等所受激励为时间延迟相关的白噪声。针对这类情况,基于复模态系统讨论NExT理论的适用性。结果表明,NExT是否适用取决于从相关函数中截取的用来进行OMA分析的数据段的位置,当数据段位于Tε区间内,NExT是适用的。在ε已知的情况下,可以基于NExT对系统进行工况模态分析。最后通过仿真进一步验证该结论。

Application of Natural Excitation Technique in the Structures under Time-delay-correlative White Noise Excitations

Operational modal analysis (OMA) is a kind of modal parameter identification technology based on output-only data. Because the excitation of the system is unknown, it is usually assumed to be uncorrelated Gaussian white noise. Natural excitation technique (NExT) enormously promotes the development of OMA, which is an efficient method for extracting free attenuation response signals from output-only data. However, for real systems, the assumption that the excitations are all uncorrelated is too strict. For the case of a moving vehicle, the excitations are time-delay-correlative white noises. Based on a complex modal system, the applicability of NExT is discussed. The result shows that the applicability of NExT depends on the time delay ε. When the data fragment cut from the correlation function of the system responses locates in the interval T>ε, NExT is valid, and when it locates in 0