This thesis presents a computational framework for the preliminary aerodynamic optimization of horizontal-axis wind turbine (HAWT) blades with power ratings below 200 kW. The optimization approach is divided into two stages: first, a Kriging surrogate model explores the design space of 4-digit NACA airfoils, parameterized through continuous variables, in order to identify an optimal initial profile; subsequently, a gradient-free optimization algorithm based on Class-Shape Transformation (CST) parametrization refines the airfoil geometry. The aerodynamic analysis couples XFOIL, used for the generation of two-dimensional airfoil polars, with the BEM solver CCBlade for rotor performance evaluation. A dual-airfoil strategy enables the sequential optimization of the root and tip sections, while the intermediate sections are obtained through interpolation of CST coefficients and aerodynamic polars. Chord and twist distributions are derived from Betz optimal relations, ensuring geometric and aerodynamic consistency among the candidate profiles. The framework, developed in Julia with an open-source perspective, is intended for applications in the sub-200 kW range and aims to provide a computationally efficient tool for the preliminary design of wind turbine rotors.
Questa tesi presenta un framework computazionale per l’ottimizzazione aerodinamica preliminare delle pale di turbine eoliche ad asse orizzontale (HAWT) con potenza inferiore a 200 kW. L’approccio di ottimizzazione si divide in due fasi: inizialmente, un modello surrogato di Kriging esplora lo spazio dei profili NACA a 4 cifre, parametrizzati tramite variabili continue, al fine di individuare un profilo iniziale ottimale; successivamente, un algoritmo gradient-free basato su parametrizzazione CST (Class-Shape Transformation) raffina la geometria del profilo. L’analisi aerodinamica accoppia XFOIL, impiegato per la generazione delle polari bidimensionali, con il solutore BEM CCBlade per la valutazione delle prestazioni del rotore. Una strategia a doppio profilo consente l’ottimizzazione sequenziale delle sezioni di root e tip, mentre le sezioni intermedie sono ottenute tramite interpolazione dei coefficienti CST e delle polari aerodinamiche. Le distribuzioni di corda e svergolamento derivano dalle relazioni ottimali di Betz, garantendo coerenza geometrica e aerodinamica tra i profili candidati. Il framework, sviluppato in Julia con finalità open-source, è progettato per applicazioni nel range sub-200 kW e mira a fornire uno strumento computazionalmente efficiente per la progettazione preliminare di rotori eolici.
Rotor blade optimization for sub-200 kW wind turbines
POLETTO, PIETRO
2025/2026
Abstract
This thesis presents a computational framework for the preliminary aerodynamic optimization of horizontal-axis wind turbine (HAWT) blades with power ratings below 200 kW. The optimization approach is divided into two stages: first, a Kriging surrogate model explores the design space of 4-digit NACA airfoils, parameterized through continuous variables, in order to identify an optimal initial profile; subsequently, a gradient-free optimization algorithm based on Class-Shape Transformation (CST) parametrization refines the airfoil geometry. The aerodynamic analysis couples XFOIL, used for the generation of two-dimensional airfoil polars, with the BEM solver CCBlade for rotor performance evaluation. A dual-airfoil strategy enables the sequential optimization of the root and tip sections, while the intermediate sections are obtained through interpolation of CST coefficients and aerodynamic polars. Chord and twist distributions are derived from Betz optimal relations, ensuring geometric and aerodynamic consistency among the candidate profiles. The framework, developed in Julia with an open-source perspective, is intended for applications in the sub-200 kW range and aims to provide a computationally efficient tool for the preliminary design of wind turbine rotors.| File | Dimensione | Formato | |
|---|---|---|---|
|
Poletto_Pietro.pdf
accesso aperto
Dimensione
5.43 MB
Formato
Adobe PDF
|
5.43 MB | Adobe PDF | Visualizza/Apri |
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/20.500.12608/109936