Power spectrum and kurtosis separation method for multi-shaker non-Gaussian random vibration control

Abstract

An innovative power spectrum and kurtosis separation control strategy for a non-Gaussian random vibration test is proposed in this work. In the existing non-Gaussian random vibration test method, the spectrum and kurtosis control channels are identical and the responses of non-control channels usually approach Gaussian. This work aims at the practical situation where the spectrum and kurtosis control channels are not completely coincident. The key ideas of the proposed non-Gaussian separation control strategy are that the predefined reference spectral densities of the spectrum control channels are used to generate equivalent reference spectral densities of the kurtosis control channels, then the equivalent reference spectral densities combined with the reference kurtoses are used to obtain the desired drive signals. For the single-shaker separation control, the non-Gaussian separation control method can be implemented. When it comes to the multi-shaker case, the coupling problems caused by the cross spectra and inherent multi-input multi-output physical system have an influence on the setting between the reference spectra and reference kurtoses and on the simultaneous control of the response spectra and response kurtoses. Simulation examples with a mass-spring-damping system are carried out and the results are discussed. Finally, biaxial separation and non-separation non-Gaussian random vibration tests are provided to verify the feasibility of the proposed control method.