Holes Identification in Composite Structures by Artificial Neural Network

Y.C. Liang and C. Hwu (Taiwan)


composite structure, artificial neural network, inverseproblem.


In this paper, the on-line identification of holes in the composite structures is performed successfully by using the artificial neural network (ANN). The detectors of the inverse problem are the static strains simply measured by the strain gauges, and the system of on-line identification is accomplished through the back propagation neural network (BPN). The network of BPN with two hidden layers is designed to express the highly nonlinear relationship between the strains and the size, location and orientation of holes. The optimal learning efficiency and accuracy are also showed by the numerical results.

Important Links:

Go Back