Fatigue life from kurtosis controlled excitations

Abstract Components of ground and flight vehicles are subjected to random vibration excitations. A common approach to qualify such components is to expose them to a random Gaussian excitation, defined by the power spectral density (PSD) of the vibration under consideration. In real life, however, it is common to experience non-Gaussian acceleration inputs such as road irregularities in the automotive world or turbulent pressure fluctuations for the aerospace sector. Traditional Gaussian random test signals do not accurately represent the bursts and peaks seen in service use. The consequence of not using the right type of test signal during vibration testing of the product leads to higher field failure rates and added warranty costs. Modern controllers can generate non-Gaussian excitation signals with a given PSD and kurtosis. A simulation with kurtosis control makes the vibration test more realistic and therefore closer to real-world excitations. This paper addresses the question of linking fatigue damage with the prescribed input kurtosis. Direct applications of these results include improved fatigue life estimations and a method to accelerate shaker tests by generating high kurtosis, non-Gaussian drive signals.