Time frequency analysis for nonlinear identification of structures

Abstract

In this paper, techniques based on Hilbert Huang Transform (HHT) and wavelet transform (WT) are proposed to extract the desired features for identification of the presence of nonlinearity in structures. It is shown in this paper that the instantaneous frequency and Hilbert marginal spectrum obtained from HHT and holder exponent obtained from WT can be effectively used as good indicators to identify the presence of nonlinearity. A numerically simulated example and also an experimental data are analyzed to evaluate the robustness of these nonlinear indicators. In this paper, the investigations are carried out on a breathing crack problem which exhibits a bilinear stiffness effect and an eight degree of freedom spring mass system.