This thesis deals with solving a mechanical-structural analysis first of a clamped beam and then of a cylindrical shell structure subjected to a pressure load, represented with a Fourier series, by means of implementation and the use of a Neural Network built with TensorFlow 2.0, written in the Python programming language, and to compare the results thus obtained with the analytical results calculated using the Transfer Matrix Method for the shell structure and the elastic line theory for the clamped beam.
La tesi si occupa di risolvere una analisi meccanico-strutturale prima di una trave incastrata e poi di una struttura di tipo shell cilindrica sottoposta a un carico di pressione, rappresentato con una serie di Fourier, mediante l'impelementazione e l'utilizzo di una rete neurale costruita con TensorFlow 2.0, scritta in linguaggio di programmazione Python, e andare a confrontare i risultati così ottenuti con quelli analitici calcolati mediante l'utilizzo del Transfer Matrix Method per la struttura shell e della teoria della linea elastica per la trave incastrata.
Structural analysis using a Neural Network approach of a cylindrical shell subjected to a pressure load
CAGNATO, ENRICO
2022/2023
Abstract
This thesis deals with solving a mechanical-structural analysis first of a clamped beam and then of a cylindrical shell structure subjected to a pressure load, represented with a Fourier series, by means of implementation and the use of a Neural Network built with TensorFlow 2.0, written in the Python programming language, and to compare the results thus obtained with the analytical results calculated using the Transfer Matrix Method for the shell structure and the elastic line theory for the clamped beam.File | Dimensione | Formato | |
---|---|---|---|
Cagnato_Enrico.pdf
embargo fino al 09/01/2025
Dimensione
19.15 MB
Formato
Adobe PDF
|
19.15 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/47878