Structural Health Monitoring (SHM) system can monitor damage of components in real time, allowing for the maintenance operations, giving the possibility to replace the classical scheduled maintenance approach with a continuous monitoring of the structural integrity of components. This approach is particularly interesting when it’s applied to innovative materials such as sandwich panels. These structures are largely adopted in the aeronautical field and in particular for the construction of helicopter frames. For this scope the most common configuration of sandwich panel is the one with Nomex honeycomb core and Al2024 aluminium skins. Sandwich panels are exposed to low velocity impacts which can reduce drastically their strenght, compromising passenger's safety. Aim of the present work is to demonstrate the possibility of development of a SHM system: it should be able to monitor low-velocity impacts on sandwich panels with Nomex honeycomb core and aluminium Al2024-T3 skins. Firstly a finite element model of the sandwich panel was created using ABAQUS 6.10 able to simulate the impact phenomena. One of the basic requirements is that the model has to be "light", allowing reduced simulation times and maintaining a good approximation. The results obtained were compared with the experimental results, with a good correlation of results. As a result, an artificial neural network was carried out using the software Matlab. The purpose of the network is to quantify the impact energy and locate the point of impact in the sandwich panel. The data of elastic deformation obtained by the FEM simulations were used to train and test the network created, evaluating its performance. At this point the skins of the sandwich panel have been analyzed and modeled in detail to study the wave propagation of elastic strain. Two different strategies were followed for modeling FEM using ABAQUS 6.10: a model based on solid elements and another one with shell elements. These were compared both in terms of time of calculation, and in terms of approximation of the model. Based on the results obtained in the study of elastic wave propagation, the instrumentation for the experimental tests has been chosen but, for reasons of time, were not carried out and it was not therefore possible to make a comparison with the data obtained in simulations. Finally, these results are reported and summarized; moreover new ideas for future developments are proposed.
Il monitoraggio strutturale (SHM - Structural Health Monitoring) permette di monitorare in tempo reale il danneggiamento dei componenti consentendo così l'ottimizzazione delle operazioni di manutenzione, dando la possibilità di passare da un approccio di manutenzione programmata ad un approccio di tipo correttivo. Questo approccio risulta essere particolarmente interessante quando viene applicato a materiali innovativi come i pannelli sandwich. Questi sono ampiamente utilizzati in campo aeronautico, in particolare per le fusoliere degli elicotteri. La configurazione più comune utilizzata per la realizzazione di telai di elicotteri è quella costituita da una struttura a nido d'ape in Nomex e le pelli in alluminio Al2024. I pannelli sandwich sono esposti ad impatti a bassa velocità che possono ridurne drasticamente la resistenza, compromettendo la sicurezza dei passeggeri. Lo scopo di questo lavoro di tesi è quello di sviluppare un sistema SHM in grado di monitorare gli impatti a basse velocità sui pannelli sandwich composti da un cuore a nido d'ape in Nomex e pelli in alluminio Al2024-T3. Inizialmente è stato utilizzato un modello a elementi finiti (FEM) del pannello sandwich mediante ABAQUS 6.10 con il quale è stato possibile simulare i fenomeni d'impatto ad energie tali per cui si ha plasticizzazione del pannello. Uno dei requisiti fondamentali che il modello deve rispettare è quello di permettere tempi di simulazione ridotti mantenendo comunque una buona approssimazione. I risultati ottenuti sono stati quindi confrontati con i risultati sperimentali, riscontrandone una buona correlazione. Tale modello è stato utilizzato per creare un database di informazioni utilizzate per allenare e testare una rete neurale artificiale creata mediante Matlab. Lo scopo della rete è quello di quantificare l'energia d'impatto e localizzare il punto d'impatto nel pannello sandwich. Tale rete neurale artificiale è stata poi ottimizzata e ne sono state valutate le prestazioni. Al fine di studiare nel dettaglio la propagazione dell'onda elastica, sono state analizzate le pelli in alluminio per basse energie d'impatto, ossia per energie tali per cui la lastra non subisce deformazioni plastiche permanenti. Sono state seguite due strategie differenti per la modellazione FEM, sempre con l'ausilio di ABAQUS 6.10: la modellazione mediante solid elements e la modellazione mediante shell elements. Queste sono state confrontate tra loro sia in termini di tempistiche di calcolo, sia in termini di approssimazione del modello. I risultati ottenuti dallo studio della propagazione dell'onda elastica sono stati utili alla progettazione dei test sperimentali, oggetto di ricerche future. Infine vengono riportati e riassunti i risultati ottenuti e vengono proposti dei suggerimenti per eventuali sviluppi futuri.
Approccio numerico per la diagnosi d'impatto a basse velocità su pannelli sandwich
ARIOTTI, MONICA
2011/2012
Abstract
Structural Health Monitoring (SHM) system can monitor damage of components in real time, allowing for the maintenance operations, giving the possibility to replace the classical scheduled maintenance approach with a continuous monitoring of the structural integrity of components. This approach is particularly interesting when it’s applied to innovative materials such as sandwich panels. These structures are largely adopted in the aeronautical field and in particular for the construction of helicopter frames. For this scope the most common configuration of sandwich panel is the one with Nomex honeycomb core and Al2024 aluminium skins. Sandwich panels are exposed to low velocity impacts which can reduce drastically their strenght, compromising passenger's safety. Aim of the present work is to demonstrate the possibility of development of a SHM system: it should be able to monitor low-velocity impacts on sandwich panels with Nomex honeycomb core and aluminium Al2024-T3 skins. Firstly a finite element model of the sandwich panel was created using ABAQUS 6.10 able to simulate the impact phenomena. One of the basic requirements is that the model has to be "light", allowing reduced simulation times and maintaining a good approximation. The results obtained were compared with the experimental results, with a good correlation of results. As a result, an artificial neural network was carried out using the software Matlab. The purpose of the network is to quantify the impact energy and locate the point of impact in the sandwich panel. The data of elastic deformation obtained by the FEM simulations were used to train and test the network created, evaluating its performance. At this point the skins of the sandwich panel have been analyzed and modeled in detail to study the wave propagation of elastic strain. Two different strategies were followed for modeling FEM using ABAQUS 6.10: a model based on solid elements and another one with shell elements. These were compared both in terms of time of calculation, and in terms of approximation of the model. Based on the results obtained in the study of elastic wave propagation, the instrumentation for the experimental tests has been chosen but, for reasons of time, were not carried out and it was not therefore possible to make a comparison with the data obtained in simulations. Finally, these results are reported and summarized; moreover new ideas for future developments are proposed.File | Dimensione | Formato | |
---|---|---|---|
Tesi_Monica_Ariotti.pdf
accessibile in internet per tutti
Descrizione: Testo della Tesi
Dimensione
7.59 MB
Formato
Adobe PDF
|
7.59 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/79821