Development of Structural Neural Network Design Tool for Buckling Behaviour of Skin-Stringer Structures Under Combined Compression and Shear Loading

Abstract

Stiffened panels are commonly used in aircraft structures in order to resist high compression and shear forces with minimum total weight. Minimization of the weight is obtained by combining the optimum design parameters. The panel length, the stringer spacing, the skin thickness, the stringer section type and the stringer dimensions are some of the critical parameters which affect the global buckling allowable of the stiffened panel. The aim of this study is to develop a design tool and carry out a geometric optimization for panels having a large number of stringers. The panel length and the applied compression-shear loads are assumed to be given. In the preliminary part, a simplified panel with minimized number of stringers is found. This panel gives the same equivalent critical buckling load of panels having larger number of stringers. Additionally, the boundary conditions to be substituted for the outer stringer lines are studied. Then the effect of some critical design parameters on the buckling behavior is investigated. In the second phase, approximately six thousand finite element (FE) models are created and analyzed in ABAQUS FE program with the help of a script written in Phyton language. The script changes the parametric design variables and analyzes each skin-stringer model, and collect the buckling analysis results. These design variables and analysis results are grouped together in order to create an artificial neural network (ANN) in MATLAB NNTOOL toolbox. This process allows faster determination of buckling analysis results than the traditional FE analyses.