Measurement of natural frequency of track based on Bayesian estimation

Abstract

Track modal parameters obtained by the vibration test contain uncertain factors due to the influence of the test environment, the test equipment, the operator, and the simplification of the modal analysis process. In order to estimate the statistical characteristics of the physical parameters of the track, the conjugate prior distribution is selected in the Bayesian estimation method, and the posterior distribution is derived from the experimental data, and the estimated value of the mean and variance of the natural frequency of the track are obtained. The reliability of the method is verified by simulation and experiment. The results show that when only a few samples can be obtained, the full use of prior information and test data can estimate the frequency parameters and reduce the length of the confidence intervals of the mean and the variance. The proposed method has practical reference value in engineering.