Application Of Structural Modal Parameter Identification Based On Natural Excitation Technique

Large-scale structural health monitoring technology is a research focus in recent years. Damage diagnosis is the difficulty and core issues of structural health monitoring. Most of the existing structural damage detection methods are based on the vibration characteristics of structural; therefore, modal parameter identification is a key problem of structural health monitoring technology. Traditional modal parameter identification methods exert artificial excitation to measure input and output signals at the same time. According to frequency response function or impulse response function to identify the structural modal parameters. That only applies to small structures. It is difficult to exert artificial excitation on large-scale structural. Structural modal parameter identification under ambient excitation caused widespread concern, for it has many advantages, such as low cost and no interruption of structural service conditions.Structure modal parameter identification under ambient excitation was used to monitor the structural health online. Firstly, fundamental concepts and main research methods were summarized, and then the theoretical basis for the following research is established. Secondly, Natural Excitation Technique and two kinds of time-domain modal parameter identification methods were introduced. The numerical simulation of three DOF frame was indentified to describe the operation process of NExT-ERA and NExT-ITD. Finally, Issues of determining system order number and extracting accurate modal about ERA and ITD were discussed.NExT assume that external excitation was Gaussian white noise. In fact, ambient excitation is a kind of non-stationary random process. In order to validate the availability of the NExT-ERA and NExT-ITD under ambient excitation, the modal parameters of a model of three DOF frame and a three-span continuous bridge were calculated, respectively. Singular entropy incremental spectrum and stabilization diagram was used to determine system order number and extract accurate mode. The results were compared with which identified by peak pick method in frequency domain. The experiment and engineering application show that NExT-ERA and NExT-ITD are all effective methods of structural modal parameter identification.