The purpose of this Master's thesis is to assess the industrial feasibility of high-quality Large Eddy Simulations for accurate peak pressure load estimation on buildings. The discussion will start with a brief introduction to the topic, followed by the state-of-the-art assessment presenting the current academic work on the subject and the main methods employed during the research. The presentation of the results obtained will first focus on the synthetic generation of turbulence utilising the PRFG method applied as inlet condition in an empty domain, which shows satisfying results and is able to reproduce the prescribed Eurocode categories. The evolution of the turbulence along the domain will be discussed, as well as the dependency of the statistical quantities on the mesh refinement. The discussion will continue by analyzing the results obtained by applying the method as inflow condition on a slender tower to observe the pressure coefficients, which in turn are not satisfactory, partially due to numerical difficulties associated with sharp edges. The main challenges typical of this category of CFD simulations will be discussed. At the end a pilot case will be proposed to assess computational charges in terms of lead time and monetary costs in comparison with wind tunnel testing, to study the competitiveness of the method from an industrial point of view. Some smaller assessments in terms of reduced number of wind directions studied and acquisition time show promise for applications in an industrial setting.
Questa tesi magistrale si pone come obbiettivo di valutare la fattibilità dal un punto di vista industriale dell'utilizzo della fluidodinamica computazionale (Large Eddy Simulation) allo scopo di stimare accuratamente i picchi di pressione sugli edifici. La discussione inizierà con una rapida introduzione al topic, seguita dalla presentazione dello stato dell'arte, i recenti sviluppi in ambito accademico e i principali metodi utilizzati nel corso della ricerca. La presentazione dei risultati ottenuti inizialmente verterà sulla generazione sintetica di turbolenza utilizzando il metodo PRFG$^3$ applicato come condizione di ingresso in un dominio vuoto, la quale mostra risultati soddisfacenti, in grado di riprodurre le prescrizioni delle categorie Eurocode. Verrà discussa l'evoluzione della turbolenza all'interno del dominio, così come la dipendenza delle quantità statistiche del raffinamento della maglia di calcolo. Successivamente, verranno analizzati i risultati ottenuti applicando questo metodo come condizione di ingresso in una simulazione contenente una torre a base quadrata, al fine di osservare i coefficienti di pressione, che risultano non soddisfacenti, parzialmente a causa delle difficoltà numeriche associate agli angoli vivi. Verranno discusse le principali problematiche associate a questa categoria di simulazioni fluidodinamiche. Infine, verrà proposto un caso pilota al fine di valutare i costi industriali in termini monetari e di tempi di consegna. Degli studi ridotti in termini di numero di direzioni del vento studiate e tempi di acquisizione statistica mostrano buone potenzialità industriali.
Synthetic turbulence generation methods in openFOAM: industrial feasibility of peak pressure estimation for wind loads assessment in CFD
Leone, Tiziano
2022/2023
Abstract
The purpose of this Master's thesis is to assess the industrial feasibility of high-quality Large Eddy Simulations for accurate peak pressure load estimation on buildings. The discussion will start with a brief introduction to the topic, followed by the state-of-the-art assessment presenting the current academic work on the subject and the main methods employed during the research. The presentation of the results obtained will first focus on the synthetic generation of turbulence utilising the PRFG method applied as inlet condition in an empty domain, which shows satisfying results and is able to reproduce the prescribed Eurocode categories. The evolution of the turbulence along the domain will be discussed, as well as the dependency of the statistical quantities on the mesh refinement. The discussion will continue by analyzing the results obtained by applying the method as inflow condition on a slender tower to observe the pressure coefficients, which in turn are not satisfactory, partially due to numerical difficulties associated with sharp edges. The main challenges typical of this category of CFD simulations will be discussed. At the end a pilot case will be proposed to assess computational charges in terms of lead time and monetary costs in comparison with wind tunnel testing, to study the competitiveness of the method from an industrial point of view. Some smaller assessments in terms of reduced number of wind directions studied and acquisition time show promise for applications in an industrial setting.File | Dimensione | Formato | |
---|---|---|---|
LEONE_Thesis_Final_Modified.pdf
non accessibile
Descrizione: finale
Dimensione
72.48 MB
Formato
Adobe PDF
|
72.48 MB | Adobe PDF | Visualizza/Apri |
LEONE__Executive_Summary.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: executive summary
Dimensione
875.68 kB
Formato
Adobe PDF
|
875.68 kB | 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/10589/215532