Buckling analysis of functionally graded perforated plates using finite element and artificial neural network methods

In this study, the lateral and longitudinal buckling loads are presented for Functionally Graded Plates (FGPs) with one end fixed and the other end free. The properties of FGPs are calculated from the mixture rule and power law. Effects of the material order, material index, and size, location, numbers and shape of hole on the buckling loads are investigated. To verify the numerical results, theoretical solutions for some material index values are calculated for longitudinal buckling. Both buckling loads reduce with increased hole sizes, and they reach minimum values when the hole location is the closest to fixed end. Both buckling loads are minimum in plates of square hole, whose hole size is equal to hole dimensions of the other shapes. However, both buckling loads of plates with triangular holes, whose hole areas are equal, are minimum compared to other plates. Therefore, considering other damage situations in terms of buckling loads, it would be better to prefer FGPs with circular holes. Also, to reduce the processing time, this problem is trained with artificial neural networks (ANN) and the ANN is used to obtain new results for different situations. It is seen that ANN data is compatible with FE results.