![Sample our Mathematics & Statistics journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5943/TOC-Subject-Sample-2021-Mathematics-Statistics.jpg"})
Abstract
In this article, a novel method for estimating inertial and stiffness parameters for aircraft structures is presented. The method is based on a combination of the finite element method (FEM) and artificial neural networks (ANNs). ANNs are known for their non-linearity and input/output mapping features and the proposed procedure aims to develop network architecture and training data capable of overcoming many of the shortfalls associated with previous parameter estimation techniques, such as uniqueness of solution and inadequate performance in the presence of uncertainties. The proposed parameter estimation technique is used to determine inertial and stiffness properties of a linear FEM comprised of planar Hermitian beam elements. It achieves this with surprising accuracy. The stiffness distribution is estimated from static load/deformation considerations, while the inertial distribution is estimated from the modal characteristics of the model. Finite Element Analysis in MATLAB® is used to generate the training data for the networks, which are simulated using its Neural Network Toolbox.
Acknowledgements
The authors would like to acknowledge the support of the DSTO/RMIT Center of Expertise in Aerodynamic Loading (CoE-AL), School of Aerospace, Mechanical & Manufacturing Engineering, RMIT University and the Department of Mechanical and Aerospace Engineering, University of Texas at Arlington.