Dynamic Testing Modal Parameters Identification And Uncertainty Evaluation Under Unknown Excitation

Modal parameter plays an important role in structural health monitoring, model correction and response prediction. According to measurement of the excitation signal, the modal identification method can be divided into the traditional modal identification method which is easy to measure and the environment mode identification method which is difficult or not be measured. However, there are errors between experimental modal parameters and theoretical parameters during the data acquisition and modal analysis, which is probably caused by the noise in environment, simplified model, the accuracy of instruments and non-standard operation of testers. In order to assess the quality of modal parameter, we must study the uncertainty of modal parameter identification. In the field of static test, we have a very sound theory to evaluate uncertainty of measurement results. However, these methods are no longer applicable for evaluating uncertainty of dynamic testing due to time-varying dynamic test of randomness, correlation and dynamic. Therefore, it has a very important significance to develop a systematic method to determine uncertainty of modal parameters.Firstly, we need analyze the difference of geometric in uncertainty transfer under static and dynamic and analyze modal parameters of the beam using finite element theory. The distribution of five natural frequency of the beam did not appear normal when the geometry error of the beam appeared normal distribution. Sensitivity can be considered as the degree of change of structural properties caused by the designed variables or the changes of parameters. Sensitivity analysis can be an effective response the sensitivity of modal parameters to physical parameters, and a multi-degree of freedom system as an example, individually analyze the relationship between inherent frequency variation and change of mass and stiffness.In order to identify uncertainty and modal parameters of vibration system, Bayesian method and ITD method based on environmental excitation were used in this paper. The uncertainty components caused by noise, the length of data and the degrees of freedom of measurement system were analyzed by Bayesian. A more systematic evaluation method to assess the uncertainty of modal parameter identification was given combining the uncertainty components introduced by identification method and instrument calibration report. Finally, the effectiveness of Bayesian method and ITD method was verified by clamped beam. The systematic uncertainty of modal parameters obtained by different identification methods were evaluated, given the combined uncertainty.