Identification of non-proportional structural damping using experimental modal analysis data

Abstract

A verified computational model of a complex structure is crucial for reliable vibro-acoustic simulations. Mass and stiffness matrices of such a computational model may be constructed correctly, provided all the design information is available. Since it is an unknown, the damping matrix is usually populated through mathematical models based on some assumptions. In the current study, it is proposed to use the identified non-proportional structural damping matrix in the computational model. Structural damping matrix can be identified using the complex frequency response functions obtained from experimental modal analysis data. No matter what type of a damping mechanism a structure has; proportional or non-proportional, the frequency response functions of the system can be measured. First, the calculation procedure for the non-proportional structural damping matrix is explained. The damping matrix of an analytical model is identified successfully using the proposed procedure. The same procedure is then applied through a case study. Computational model of a test vehicle is constructed. Next, the test vehicle is subjected to a modal test to measure the frequency response functions of the structure. Incompleteness of the measured data and the requirements of the procedure are discussed, as well. The described procedure can be used in any model updating framework.

Enhanced method for optimum driving point identification for modal testing

“…However, a full modal description of the structure usually cannot be obtained by this method. If an EMA is carried out with the aim of identifying all modal parameters in a certain frequency range, as it is often necessary for validating numerical models, the planning and selection of driving points is either based on testing multiple possible excitation points on the system prior to the actual EMA (Donaldson and Mechefske, 2020;Oktav, 2020) or on numerical pre-calculations. General requirements for the selection of driving points are the avoidance of areas of nodal lines and areas sensitive to double hits.…”

Cumulative power spectral density‐based damping estimation

“…The random decrement technique based on the transformation of modal response into free vibration response has offered a promising procedure to estimate the damping ratio in the time domain, however, it has some variations in the estimated damping depending on the initial condition, segment set, and the non-Gaussian nature of the external load. 6,[30][31][32] More advanced techniques such as the eigenvalue realization algorithm extended by the natural excitation technique (NExT-ERA), 33,34 autoregressive model (AR) [35][36][37][38] and stochastic subspace iteration 39,40 require familiarity with its mathematical details, have been widely utilized for SI. However, the uncertainties associated with model parameters such as the selection of appropriate model order, the time lag of correlation, and so forth.…”

A time-domain method for free vibration responses of an equivalent viscous damped system based on a complex damping model

“…A complex damping model 1 is commonly an internal damping model, which is also called the structural damping model 2,3 and frequency-independent damping model. 4,5 The damping parameter of the complex damping model is the loss factor.…”