ABSTRACT
The purpose of this study is to figure out the relationship between sampling frequency and fatigue damage calculated in time-domain. Three random loadings with representative Power Spectral Densities (PSDs) are studied to ascertain the relationship. Based on the simulation analysis, the sampled signal with sampling frequency lower than eight times of the upper limit of the frequency range cannot be used to calculate fatigue damage directly for time-domain method. The reasons why the sampling frequency should be so high are investigated qualitatively. Based on the analysis results, the sampling frequency empirically should be at least nine times of the upper limit of the frequency range of the power spectral density. When the sampled signals cannot reach this requirement, Shannon’s sampling theorem should be used to reconstruct the loading time history. And then the resampled signal with sampling frequency high enough is used to estimate the fatigue damage. Lastly, the numerical example verified the correction of the proposed approach.
