Non-destructive identification of effective elastic properties of carbon fiber reinforced polymer laminates using modal parameters

This work presents a non-destructive approach for determining the effective elastic properties of carbon fiber reinforced polymer laminated composites. Two types of laminates were considered: unidirectional and twill-weave carbon-fiber composites. The purpose of the study is to estimate the quality of a numerical-experimental method for identifying the elastic moduli and Poisson’s ratios of orthotropic materials. The method combines vibration testing data with finite element modeling and global optimization. To validate the performance of the identification methodology, data from a virtual experiment on the numerical determination of the natural frequencies and vibration modes of a free rectangular plate were used. In the first stage of the identification procedure, a refined finite element model of the plate with initial guess values of the elastic constants was employed to determine the natural frequencies. The initial guess values of elastic constants were obtained using the design of experiments method. In the second stage, response surfaces were constructed to approximate the structural behavior of the composite plate. The elastic properties were analyzed to determine the influence of parameters that do not affect the natural frequencies. Finally, identification of material properties was performed by minimizing the error functional between the virtual experimental and numerical frequency values. The minimization was carried out using the differential evolution algorithm for global optimization. The results indicate that, although the methodology accurately identifies the dominant in-plane elastic constants of orthotropic composite laminates, it does not permit reliable estimation of out-of-plane properties due to their negligible influence on the measured natural frequencies. Consequently, the approach should be considered as a partial identification tool focused on the most dynamically influential parameters.