A Distribution-Based Damping Estimation Method for Random Vibration Response and Its Applications

Abstract

In this paper, a damping estimation method based on distribution of zero-crossing times and its applications are presented. The method is empirically derived and it is based on Rice distribution operating on probability distribution of zero-crossings. The method’s assertion is that a relationship exists between said probability distribution and the bandwidth of a mechanical system and this can be used as a means of increasing confidence in more traditional damping estimation methods. The method is applicable to responses that are due to broadband random excitation featuring a mono-harmonic response. Given that a multitude of modes will be active in aero-engines; its application has to be preceded with a suitable band-pass filter to isolate modes of interest. Such filtering is shown to affect the damping estimates. Some controlled laboratory tests are performed to verify the accuracy of the method and to study the effects of preceding filtering amongst other factors such as modal density, data length etc. The performance of the method is compared with Fourier transform based damping estimation methods. The results of application of the method to real engine measurements and carefully controlled laboratory tests are also presented.