Session: 02-04-01: Computer Methods and Reduced Order Modeling

Paper Number: 151469

151469 - Development of Finite Element Models With Node-Wise Variations of Structural Theories for Dynamic Analyses 

This paper presents a novel modeling approach to improve the accuracy and computational efficiency of Finite Element (FE) models for structural dynamics applications. The proposed approach has three distinctive features: 1D and 2D elements are used, structural theories are higher-order, and the structural theory is a nodal property, i.e., each FE node can have a distinct set of generalized variables or degrees of freedom. The Carrera Unified Formulation (CUF) is the underlying theoretical framework of this paper; CUF allows the generation of any structural theory and related governing equations and FE arrays. 1D and 2D higher-order structural theories are used to overcome the assumptions of standard 1D and 2D FE and, therefore, handle non-classical effects, e.g., warping, cross-sectional distortions, 3D strain, and stress fields. The node-dependent kinematics (NDK) approach is exploited to let the structural theory change node-to-node by adequately manipulating the FE matrices and arrays. The node-wise distribution of the structural kinematics leads to a tight global-local scheme in which local, high-fidelity models coexist with low-fidelity regions to optimize the computational effort.

This work focuses on free-vibration analyses of compact and thin-walled structures for aerospace applications and the effect of higher-order models on the natural frequencies and modal shapes. Two results are presented: the best global structural theory for a given problem and the best local distributions of high-fidelity models, i.e., models in which the computational cost is minimized for a given accuracy threshold. The aim is to provide guidelines on how and where to refine an FE model while preserving the computational overhead. Furthermore, comparisons with structural theories from the literature and commercial FE software are provided. Perspectives on using machine learning techniques to guide the proper node-wise distributions of structural theories are drawn.

Presenting Author: Marco Petrolo Politecnico di Torino

Presenting Author Biography: Marco Petrolo is an Associate Professor and a member of the MUL2 Lab in the Department of Mechanical and Aerospace Engineering of Politecnico di Torino (www.mul2.com). His current research activities involve the multiscale analysis of composites and micromechanics and the development of higher-order structural theories. He cooperates with various institutions, including the University of Washington and NASA. He is a member of the Governing Board of the Italian Association of Aeronautics and Astronautics (AIDAA).