Imperfection Sensitivity and Lower-Bound Design of Axially Compressed Concave Conical Shells: Experimental Validation and GMNIA-Based Framework
摘要
This contribution focuses on the response to axial compression of concave conical shells (CCSs) affected by initial geometric imperfections, adopting an integrated experimental and computational approach. A series of controlled axial compression tests is first conducted to characterize specimen behavior and to substantiate the numerical model validation process. Subsequently, a detailed numerical investigation evaluates three frequently adopted imperfection modeling methods, including eigenmode-based formulations, single and multiple load indentations, and localized dent imperfections simulating typical manufacturing irregularities. The numerical results quantitatively demonstrate the reduction in load-carrying capacity induced by initial geometric imperfections, confirming the pronounced imperfection sensitivity of CCSs. The magnitude of this reduction is shown to depend strongly on the imperfection type, as well as on the location and orientation of these imperfections. Among the investigated scenarios, eigenmode imperfections lead to the most severe reduction, followed by single load indentation (SLI), multiple load indentation with four perturbation loads (4MLI), and dent imperfections. Based on the validated numerical framework, a closed-form empirical expression is finally proposed to estimate lower-bound knockdown factors (KDFs) for design purposes, focusing on the most critical imperfection scenario represented by eigenmode imperfections across a range of axially compressed CCS configurations. Compared with the classical NASA SP-8019 recommendations, the proposed formulation provides a more conservative prediction for thin shells (Re/t ≥ 150) while maintaining adequate safety margins, thereby supporting its use as a lower-bound design-oriented assessment tool for the investigated CCS configurations.