Analysis of non-uniform hexagonal cross-sections for thin-walled functionally graded beams using artificial neural networks

We study static mechanical behavior of non-uniform hexagonal cross-sections for thin-walled functionally graded beams using a non-traditional computational approach based on artificial neural network. One the main objectives our is to save cost optimization process, which usually time-consuming by traditional methods such as finite element method (FEM). In this study, 1000 data sets randomly generated FEM through iterations are used training process get optimal weights. Based these obtained weights, beam behaviors under changes in material distribution thickness could then be predicted. model, ANN's inputs gradation index power-law and thickness, while outputs compliance displacements. The computed results verified against those derived from FEM.