Non-Intrusive polynomial chaos expansion for
aerodynamic uncertainty quanti cation & robust design
with manufacturing imperfections

Villette, Sergios; Βιλλέτ, Σέργιος

dc.contributor.author	Villette, Sergios	en
dc.contributor.author	Βιλλέτ, Σέργιος	el
dc.date.accessioned	2022-06-27T11:21:55Z
dc.date.available	2022-06-27T11:21:55Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/55342
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.23040
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	CFD	en
dc.subject	niPCE	en
dc.subject	KLT	en
dc.subject	Robust design optimization	en
dc.subject	Uncertainty quantification	en
dc.subject	CFD	en
dc.subject	KLT	en
dc.subject	Στιβαρός σχεδιασμός	el
dc.subject	Ανάπτυγμα πολυωνυμικού χάους	el
dc.subject	Ποστικοποίηση αβεβαιότητας	el
dc.title	Non-Intrusive polynomial chaos expansion for aerodynamic uncertainty quanti cation & robust design with manufacturing imperfections	en
dc.title	Μη-επεμβατικό ανάπτυγμα πολυωνυμικού χάους για αεροδυναμικό στιβαρό σχεδιασμό υπό κατασκευαστικές ατέλειες	el
heal.type	bachelorThesis
heal.classification	CFD	en
heal.classification	Robust design optimization	en
heal.classification	Uncertainty quantification	en
heal.language	el
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-02
heal.abstract	In the field of aerodynamics, the geometrical and ow conditions of a certain shape are usually considered to be constants, while in reality they exhibit some stochasticity, which can have a varying effect on its performance. This thesis, stresses aerodynamic cases in which the geometrical-manufacturing uncertainties of a certain shape are taken into account, by proposing a computational process capable to, firstly, evaluate the stochasticity of their performance (uncertainty quantification) and, secondly, to optimize their stochastic performance (robust design). Therefore, this thesis presents the development of software, implementing the non-intrusive Polynomial Chaos Expansion and the Karhunen-Loeve Transform theories, in order to perform aerodynamic uncertainty quantification and robust design optimization on 2D shapes with manufacturing uncertainties. The Karhunen-Loeve Transform theory is used to simulate the real-time uncertainties that may occur during the manufacturing of aerodynamic shapes. The theory of Polynomial Chaos is based on the use of orthogonal polynomials to model the stochasticity of a certain phenomena, by analyzing its stochastic input and quantifying its stochastic output, though the form of its statistical moments. The Karhunen-Loeve Transform software developed as well as the OpenFOAM© Computational Fluid Dynamics solvers are coupled to an in-house non-intrusive Polynomial Chaos Expansion code, so as to quantify the stochastic aerodynamic performance of 2D imperfect geometries. Additionally, robust design is performed on such imperfect geometries, parameterized through Volumetric B-Splines, by optimizing the statistical moments of their performance, with respect to the design variables controlling the parameterized shape. This is achieved through the incorporation of the continuous adjoint optimization algorithm, developed by PCOpt/NTUA in the OpenFOAM environment, into the aforementioned Karhunen-Loeve Transform and non-intrusive Polynomial Chaos coupled algorithm. The Karhunen-Loeve Transform code is designed to recreate imperfect perturbations on any 2D geometry and when combined the generalist nature of the non-intrusive Polynomial Chaos Expansion mathematical tool, it grants the ability to the proposed method to cope with a wide variety of aerodynamic cases with shape uncertainties. Simultaneously, the deterministic adjoint optimization method greatly mitigates the computational cost needed to perform the uncertainty quantification and robust design processes, when compared to other stochastic methods often employed in literature, such as the Evolutionary Algorithms.	en
heal.advisorName	Γιαννάκογλου, Κυριάκος Χ.	el
heal.committeeMemberName	Γιαννάκογλου, Κυριάκος Χ.	el
heal.committeeMemberName	Μαθιουδάκης, Κωνσταντίνος	el
heal.committeeMemberName	Αρετάκης, Νικόλαος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Μηχανολόγων Μηχανικών. Τομέας Ρευστών. Εργαστήριο Θερμικών Στροβιλομηχανών	el
heal.academicPublisherID	ntua
heal.numberOfPages	170 σ.	el
heal.fullTextAvailability	false