Identification of elastic parameters of composite using experimental data on modal characteristics of samples

In order to determine the elastic parameters of polymer composites, a mixed numericalexperimental method of identification was developed in the 1990s - 2000s, based on the use of experimental data on natural frequencies and eigenmodes of oscillation of samples. Its practical application involves the choice of a shape and a size of samples, a set used to identify the natural frequencies and eigenmodes of vibration, a methods of their experimental determination, a finiteelement model for modal analysis, and an algorithm for solving the identification problem. The object of the research is laminated polymer composite materials reinforced with carbon or glass fibers. The aim of the work is to develop practical aspects of implementation and assessment of the accuracy of the mixed experimental-calculation method for identifying of the elastic parameters of such materials based on experimental data on natural frequencies and eigenmodes of vibration of standard samples. Parameter identification for the material is considered as an optimization problem with an optimization function, which is a weighted sum of squares of differences between the experimental and calculated values of natural frequencies. A procedure was developed to implementing the main steps of the suggested technique: experiments, calculations and control of results. An error analysis of main steps was carried out, and the accuracy of the determined parameters of the ply was estimated. To determine the natural frequencies and eigenmodes of oscillation of the samples, the method of three-component scanning laser vibrometry was used. The experimental technique was established and parameters of the technique were chosen to ensure the necessary accuracy in determining of the natural frequencies. The parameters of the identification procedure and the finite-element model of the sample were selected. To control the obtained values of the elastic parameters, the natural frequencies of the samples were calculated, including those not used in the identification procedure. The error assessment of the in determined elastic parameters was performed on three different series of samples of carbon-fiber-reinforced laminates with the same material of the plies and different ply numbers and stacking orders. The developed technique can be recommended to determine the parameters of material models required for strength and vibrations assessment of products manufactured with layered composites.