Predicting buckling loads of perforated rectangular isotropic panels using Gene Expression Programming and Artificial Neural Network

Abstract

This study aims to develop semi-empirical formulas for foretelling the buckling loads of clamped and simply supported rectangular isotropic panels with central circular perforation exposed to different ratios of biaxial load conditions. The empirical formulas were developed and evaluated using Gene Expression Programming (GEP), Artificial Neural Network (ANN), and Finite Element Method (FEM). A total of 714 data set, generated using the FEM, is used to establish and validate the empirical formulas. This study investigates the effect of perforation sizes, plate aspect ratios, and biaxial load ratios on the buckling strength of perforated panels. The proposed formulas will allow a quick and easy estimation of buckling loads for perforated rectangular panels with acceptable accuracy without the need for sophisticated calculations. The results of the empirical formulas were comparable reasonably well with the results of the finite element analysis (FE) and available literature findings.