Treffer: Natural frequencies of composite cylindrical shells under uncertainty: a nonparametric approach.

Uncertainty in the natural frequencies of composite cylindrical shells is typically assessed with parametric models that assume specific probability laws for material and geometric inputs. Here, we propose a fully data-driven, nonparametric alternative based on the orthogonal bootstrap, integrated into a semi-analytical Rayleigh–Ritz framework for shells subjected to axial load and internal pressure. This study makes four key contributions: (1) It introduces the orthogonal bootstrap to the structural dynamics of composite shells, enabling uncertainty propagation directly from empirical samples without prescribing distributions; (2) it provides a head-to-head comparison between the nonparametric approach and a conventional parametric Monte Carlo scheme (Beta-distributed), clarifying the conditions under which each method is most advantageous; (3) it incorporates stress-stiffening effects (geometric stiffness) from axial compression and internal pressure into the Rayleigh–Ritz formulation, enabling probabilistic frequency estimates under combined loading; and (4) it validates the semi-analytical results against a refined finite-element model, demonstrating an order-of-magnitude reduction in computational time with orthogonal resampling. The results reveal that frequency variability is primarily dominated by the radius-to-thickness ratio and fiber elastic modulus, while density predominantly influences mass-driven shifts. The proposed framework offers a transparent, efficient means of quantifying uncertainty in composite shell dynamics without the risk of distributional misspecification, and it can be readily extended to other laminated configurations and loading scenarios. [ABSTRACT FROM AUTHOR]

Copyright of Journal of the Brazilian Society of Mechanical Sciences & Engineering is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)