A Thick Composite-Beam Model for Delamination Prediction by the Use of Neural Networks

A composite beam with a delamination has been modeled, accounting for the Poisson effect and transverse shear deformation. Delaminations may be caused by imperfections introduced during the fabrication process or by external loads during the operational life, such as impact by foreign objects. A consequence of delamination is the change in stiffness of the structure, and this affects the modal frequencies of the structure. The applicability of neural networks in determining delaminations in laminated composite beams is examined in the present study. The modal frequencies are obtained by using the beam model for different delamination sizes and locations. The beam model developed is applicable to various boundary conditions. A back-propagation neural network is trained to predict the delamination size and location from the natural frequencies of the beam. The neural-network model is found to be quite successful in determining the delamination size and location