The high strength-to-weight and stiffness-to-weight ratio materials, such as laminated composites, are advantageous for modern aircraft. Laminated composites with initial flaws are susceptible to delamination under buckling loads. PDA tools help enhance the industry’s understanding of the mechanisms for damage initiation and growth in composite structures while assisting in the design, analysis, and sustainment methods of these composite structures. The global-local modeling approach for the single-stringer post-buckled panel was evaluated through this effort, using Teflon inserts to simulate the defect of damage during manufacturing. This understanding is essential for designing the post-buckled structure, reducing weight while predicting damage initiation location, and addressing a potential design review for future aircraft repairs. In this work, the initial damage was captured with Teflon inserts as the starting configuration; and any reference to the damage initiation refers to any damage beyond the “initial unbonded region.” The effort aims to develop, evaluate, and enhance methods to predict damage initiation and progression and the failure of post-buckled hat-stiffened panels using multiple Abaqus FEA Virtual Crack Closure Technique (VCCT) definitions. Validation of the PDA using the VCCT material model was performed on a large single-stringer panel subjected to compressive loading. The compressive loading of the panel caused the skin to buckle before any damage began to occur locally. In addition, comparisons are made for critical aspects of the damage morphology, such as a growth pattern that included delamination from the skin-stiffener interface to the skin and ply interfaces. When compared against the experimental data produced through the NASA Advanced Composites Project (ACP), the present model captured damage migration from one surface to another, and model validations were ~5% of the experimental data.
Determination of the 1st Buckling and Collapse Loads for Integrally Stiffened Panels by Artificial Neural Network and Design of Experiment Methodology
