• Comparison of FDS, FDmS, and MI-FDS using numerical and experimental studies • FDS and FDmS tests show poor correlation with actual inflicted damage • MI-FDS accurately reproduces fatigue damage and crack mode from real missions The synthesis of accelerated random vibration tests is crucial for assessing the durability of components subjected to vibrations during their operational lifetime. The most accredited mission synthesis approach is based on the concept of damage potential, obtained by evaluating the Fatigue Damage Spectrum (FDS) of the vibration environments. However, the formulation of the FDS requires to simplify the Device Under Test (DUT) as a Single-Input-Single-Output system, while real components are, in general, Multi-Input Multi-Output (MIMO) systems. In the case of random vibration testing, this strict assumption neglects the phase and correlation between multiple axes of excitation, defined by means of the Cross Spectral Densities (CSDs). To address the limitations of the FDS , two new methodologies have been recently formalized: the Fatigue Damage multi-Spectrum (FDmS) and the Multi-Input Fatigue Damage Spectrum (MI-FDS) . These two methodologies propose a modification of the traditional FDS that allows to include the CSD terms of a multi-axis random vibration environment in the computation of the fatigue damage potential. The FDmS and the MI-FDS are currently the only techniques that tackle the problem of estimating the fatigue damage potential of multi-axis random vibration environments. The purpose of this paper is to provide an in depth study of the currently available mission synthesis techniques based on the FDS for multi-axis random testing, which is not present in the literature. In particular, a numerical and experimental study are here presented. Both studies have the objective of verifying the potential of different test tailoring approaches in a complex mission synthesis task. Ultimately, this paper provides a detailed analysis of the currently available FDS based mission synthesis approaches for multi-axis random vibration environments.
Proner et al. (Wed,) studied this question.