This thesis project explores the fatigue behavior and failure prediction of mechanical components produced through Additive Manufacturing (AM) processes, with a particular focus on the application of the Average Strain Energy Density (ASED) method and the Theory of Critical Distances (TCD). The work has been divided into several phases: initially, the ASED and TCD methods were introduced, highlighting their advantages over traditional approaches for fatigue prediction. Subsequently, an overview of additive technologies was provided, also illustrating the production chain of a component. At this point, the ASED method was calibrated using experimental data from specimens made of Ti6Al4V through Laser Powder Bed Fusion, demonstrating a good capability of the method in predicting failures. Following this, the ASED method was applied to a significant number of experimental data obtained from various studies, presenting a synthesis of the results by categorizing them first by material and then by process. The results show that the ASED method can be successfully applied to predict the fatigue behavior of AM components. An attempt was then made to correlate some typical parameters of the ASED method with certain specimen-related parameters, such as surface roughness, hardness, material microstructure, and potential defects, including porosity, lack of fusion, or the presence of residual stresses. Finally, a comparison was proposed with the other prediction method introduced, the TCD method, to verify which of the two provides the best results. This research contributes to the development of efficient numerical tools for the design of AM components, opening new perspectives for the optimization of complex mechanical parts subjected to cyclic loads.
Questa progetto di tesi esplora il comportamento a fatica e la previsione del cedimento di componenti meccanici prodotti mediante processi di Additive Manufacturing (AM), con particolare attenzione all'applicazione del metodo dell'energia di deformazione media (ASED) e del metodo delle distanze critiche (TCD). Il lavoro è stato suddiviso in diverse fasi: inizialmente, è stato presentato i metodi ASED e TCD, evidenziandone i vantaggi rispetto ad approcci tradizionali per la previsione della fatica. Successivamente, è stata fatta una panorimica sulle tecnologie additive, illustrando anche la catena di produzione di un componente. A questo punto è effettuata una calibrazione del metodo ASED su dati sperimentali relativi a provini prodotti in Ti6Al4V tramite Laser Powder Bed Fusion riscontrando una buona capacità del metodo nel prevedere i cedimenti. Successivamente, il metodo ASED è stato applicato a un buon numero di dati sperimentali ottenuti da diversi articoli, proponedo una sinstesi dei risultati categorizzando prima per materiale e poi per processo. I risultati mostrano che il metodo ASED può essere applicato con discreto successo per prevedere il comportamento a fatica di componenti AM. Si è, poi, cercato di correlare alcuni parametri tipici del metodo ASED con alcuni parametri relativi ai provini, come la rugosità superficiale,la durezza, la microstruttura del materiale e gli eventuali difetti, come porosità e mancanza di fusione o la presenza di stress residui. Infine, si è proposto un confronto con l'altro metodo di predizione introdotto, il TCD method, , verificando quale dei due fornisca i migliori risultati. Questa ricerca contribuisce allo sviluppo di strumenti numerici efficienti per la progettazione di componenti AM, aprendo nuove prospettive per la l'ottimizzazione di componenti meccanici complessi soggetti a carichi ciclici.
Prediction of failure of mechanical parts produced through Additive Manufacturing process
AGOSTINI, LEONARDO
2024/2025
Abstract
This thesis project explores the fatigue behavior and failure prediction of mechanical components produced through Additive Manufacturing (AM) processes, with a particular focus on the application of the Average Strain Energy Density (ASED) method and the Theory of Critical Distances (TCD). The work has been divided into several phases: initially, the ASED and TCD methods were introduced, highlighting their advantages over traditional approaches for fatigue prediction. Subsequently, an overview of additive technologies was provided, also illustrating the production chain of a component. At this point, the ASED method was calibrated using experimental data from specimens made of Ti6Al4V through Laser Powder Bed Fusion, demonstrating a good capability of the method in predicting failures. Following this, the ASED method was applied to a significant number of experimental data obtained from various studies, presenting a synthesis of the results by categorizing them first by material and then by process. The results show that the ASED method can be successfully applied to predict the fatigue behavior of AM components. An attempt was then made to correlate some typical parameters of the ASED method with certain specimen-related parameters, such as surface roughness, hardness, material microstructure, and potential defects, including porosity, lack of fusion, or the presence of residual stresses. Finally, a comparison was proposed with the other prediction method introduced, the TCD method, to verify which of the two provides the best results. This research contributes to the development of efficient numerical tools for the design of AM components, opening new perspectives for the optimization of complex mechanical parts subjected to cyclic loads.File | Dimensione | Formato | |
---|---|---|---|
Agostini_Leonardo.pdf
embargo fino al 28/02/2026
Dimensione
15.77 MB
Formato
Adobe PDF
|
15.77 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/20.500.12608/82262