Sustainable air mobility is a critical pursuit in modern aeronautical research, aimed at reducing aircraft emissions and advancing environmental goals. The Clean Sky 2 Joint Undertaking (CSJU), in collaboration with the European Commission and the European aeronautic industry, is dedicated to addressing this challenge. Within this context, this PhD thesis focuses on developing a physics-based, low-order numerical model to predict the nonlinear aerodynamic characteristics of lifting surfaces equipped with control surfaces typical of commercial aircraft tails. By enhancing aerodynamic efficiency and reducing aircraft weight, such innovations contribute significantly to the evolution of future aircraft generations. This thesis conducts experimental and numerical investigations into the aerodynamics of tail surfaces equipped with control surfaces. The wind tunnel campaign includes various measurement techniques, such as force and moment assessments, boundary-layer transition detection via infrared thermography, Particle Image Velocimetry (PIV) surveys, and surface flow visualizations. Numerical simulations, ranging from high-fidelity Reynolds-Averaged Navier-Stokes (RANS) computational fluid dynamics (CFD) to low-to-mid fidelity models like the panel method and non-linear lifting line model, complement experimental efforts to analyze flow solutions and understand complex aerodynamic phenomena. The thesis addresses three key objectives: Firstly, it compares RANS simulations and low-to-mid fidelity models using DUST results to evaluate their accuracy and reliability in predicting aerodynamic coefficients for swept tails with control surface deflection and varying dihedral angles. Secondly, it explores Uncertainty Quantification via the Eigenvalue Perturbation Method (EPM) to identify regions in the parameter space where RANS simulations lack reliability, guiding wind tunnel experiments for selecting geometries. Lastly, it investigates the feasibility of employing a multi-fidelity database to train a reduced model, enhancing predictions across diverse operating conditions using a Bayesian calibration framework. Through these investigations, this thesis advances our understanding of aerodynamic behaviors in complex configurations and contributes to the development of efficient and reliable prediction tools for future aircraft design, aligning with the overarching goal of sustainable air mobility.
La mobilità aerea sostenibile è un obiettivo cruciale nella ricerca aeronautica moderna, mirato a ridurre le emissioni degli aeromobili e a promuovere gli obiettivi ambientali. La Clean Sky 2 Joint Undertaking (CSJU), in collaborazione con la Commissione Europea e l’industria aeronautica europea, è dedicata ad affrontare questa sfida. In questo contesto, questa tesi di dottorato si concentra sullo sviluppo di un modello numerico a bassa complessità basato sulla fisica per prevedere le caratteristiche aerodinamiche non lineari delle superfici portanti dotate di superfici di controllo tipiche degli impennaggi degli aerei commerciali. Migliorando l’efficienza aerodinamica e riducendo il peso degli aeromobili, tali innovazioni contribuiscono significativamente all’evoluzione delle future generazioni di aeromobili. Questa tesi conduce indagini sperimentali e numeriche sull’aerodinamica delle superfici di coda dotate di superfici di controllo. La campagna in galleria del vento include varie tecniche di misurazione, come le valutazioni di forze e momenti, il rilevamento della transizione dello strato limite tramite termografia a infrarossi, indagini di Particle Image Velocimetry (PIV) e visualizzazioni del flusso superficiale. Le simulazioni numeriche, che vanno dalla fluidodinamica computazionale (CFD) ad alta fedeltà Reynolds-Averaged Navier-Stokes (RANS) a modelli a bassa e media fedeltà come il metodo dei pannelli e il modello della linea portante non lineare, completano gli sforzi sperimentali per analizzare le soluzioni del flusso e comprendere i complessi fenomeni aerodinamici. La tesi affronta tre obiettivi chiave: Primo, confronta le simulazioni RANS e i modelli a bassa e media fedeltà utilizzando i risultati DUST per valutarne l’accuratezza e l’affidabilità nella previsione dei coefficienti aerodinamici per impennaggi a freccia con deflessione delle superfici di controllo e angoli di diedro variabili. Secondo, esplora la Quantificazione dell’Incertezza tramite il Metodo di Perturbazione degli Autovalori (EPM) per identificare le regioni nello spazio dei parametri dove le simulazioni RANS mancano di affidabilità, guidando gli esperimenti in galleria del vento per la selezione delle geometrie. Infine, indaga la fattibilità di impiegare un database multi-fedeltà per addestrare un modello ridotto, migliorando le previsioni in condizioni operative diverse utilizzando un framework di calibrazione bayesiana. Attraverso queste indagini, questa tesi avanza la nostra comprensione dei comportamenti aerodinamici in configurazioni complesse e contribuisce allo sviluppo di strumenti di previsione efficienti e affidabili per la progettazione futura di aeromobili, allineandosi con l’obiettivo generale della mobilità aerea sostenibile.
Multi-fidelity experimental-numerical calibration of a reduced-order model for empennage configurations
RAUSA, ANDREA
2023/2024
Abstract
Sustainable air mobility is a critical pursuit in modern aeronautical research, aimed at reducing aircraft emissions and advancing environmental goals. The Clean Sky 2 Joint Undertaking (CSJU), in collaboration with the European Commission and the European aeronautic industry, is dedicated to addressing this challenge. Within this context, this PhD thesis focuses on developing a physics-based, low-order numerical model to predict the nonlinear aerodynamic characteristics of lifting surfaces equipped with control surfaces typical of commercial aircraft tails. By enhancing aerodynamic efficiency and reducing aircraft weight, such innovations contribute significantly to the evolution of future aircraft generations. This thesis conducts experimental and numerical investigations into the aerodynamics of tail surfaces equipped with control surfaces. The wind tunnel campaign includes various measurement techniques, such as force and moment assessments, boundary-layer transition detection via infrared thermography, Particle Image Velocimetry (PIV) surveys, and surface flow visualizations. Numerical simulations, ranging from high-fidelity Reynolds-Averaged Navier-Stokes (RANS) computational fluid dynamics (CFD) to low-to-mid fidelity models like the panel method and non-linear lifting line model, complement experimental efforts to analyze flow solutions and understand complex aerodynamic phenomena. The thesis addresses three key objectives: Firstly, it compares RANS simulations and low-to-mid fidelity models using DUST results to evaluate their accuracy and reliability in predicting aerodynamic coefficients for swept tails with control surface deflection and varying dihedral angles. Secondly, it explores Uncertainty Quantification via the Eigenvalue Perturbation Method (EPM) to identify regions in the parameter space where RANS simulations lack reliability, guiding wind tunnel experiments for selecting geometries. Lastly, it investigates the feasibility of employing a multi-fidelity database to train a reduced model, enhancing predictions across diverse operating conditions using a Bayesian calibration framework. Through these investigations, this thesis advances our understanding of aerodynamic behaviors in complex configurations and contributes to the development of efficient and reliable prediction tools for future aircraft design, aligning with the overarching goal of sustainable air mobility.File | Dimensione | Formato | |
---|---|---|---|
PhD_Thesis___Andrea_Rausa.pdf
solo utenti autorizzati a partire dal 21/05/2025
Dimensione
102.75 MB
Formato
Adobe PDF
|
102.75 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/10589/221372