Laboratoire de Mécanique des Fluides et d'Acoustique - UMR 5509

LMFA - UMR 5509
Laboratoire de Mécanique des Fluides et d’Acoustique

Nos tutelles

Nos partenaires

Accueil > Actualités > Thèses - Habilitations à diriger des recherches

Soutenance de thèse ECL - VKI

Alessandro Zarri

Lundi 28 juin 2021, 15h00, visioconférence

Alessandro Zarri

Aerodynamic and acoustic investigation of automotive fan-driven cooling systems

Composition du jury :

Pr. Thomas CAROLUS, Universität Siegen, Allemagne, rapporteur.
Pr. Marlène SANJOSE, École de Technologie Supérieure, Québec (CA), rapporteur.
Pr. Michel ROGER, École Centrale de Lyon, directeur de thèse.
Pr. Christophe SCHRAM, von Karman Institute for Fluid Dynamics, Belgique, coencadrant.
Pr. Christophe BAILLY, École Centrale de Lyon, examinateur.
Dr. Michaela HERR, Centre aérospatial allemand (DLR), Germany, examinatrice.
Dr. Manuel HENNER, Valéo Systèmes Thermiques, France, examinateur.
Pr. Stéphane MOREAU, Université de Sherbrooke, Québec (CA), examinateur.
Dr. Julien GRILLIAT, ebm-papst Sankt Georgen & Co, France, examinateur.

Lien pour la visioconférence
Id : 732-339-949
This meeting is locked with a password : Phd_A-Zarri28

The mitigation of urban noise pollution, due to automobiles among other things, is a growing international concern, as exposure to noise at high levels or for long periods of time can lead to physical and mental health problems. This research work focuses on broadband noise emitted by fan cooling modules, used for engine temperature control. Several mechanisms responsible for noise generation are present in such applications due to the aerodynamic interaction between the airflow and the fan blades. Three broadband noise mechanisms dominating the acoustic spectra are identified in the literature review, namely : trailing-edge noise, turbulence-interaction noise, and tip-clearance noise. First, this work proposes to estimate the relative importance between these mechanisms, especially in the presence of the radiator upstream of the propeller. Indeed, its influence on the modification of the flow through the radiator and on the resulting aerodynamic sound sources is not yet clearly defined. Direct sound measurement and rotating acoustic beamforming techniques are being implemented at the von Karman Institute for Fluid Dynamics in Belgium and at the University of Sherbrooke in Canada. The obtained sound-localization maps indicate that the radiator has a negligible effect on the source position, which is always located at the leading edge of the blade tip when the operating conditions of the fan used vary. This experimentally validates the high-fidelity simulations available in the literature and extends the importance of the tip-clearance noise to high frequencies. Since the effect of the radiator is not restricted to its influence on the acoustics, an experimental campaign was conducted to characterize the turbulent flow downstream of the fan, exploiting a stereoscopic PIV technique. For the analyzed cases, a maximum turbulence intensity production of about 15% is measured, decreasing to 6% at the position corresponding to the fan location. Significant levels of anisotropy and non-homogeneity are found here, requiring an analysis of the applicability of the isotropy-based von Kármán turbulence model, typically used in noise prediction methods. The results indicate that this model approaches reasonably well the two-dimensional turbulence spectra in most cases. In the second part of the thesis, a review of previous work shows that high-fidelity simulations, although accurate, are still too expensive when aeroacoustics is taken into account in industrial pre-design optimization processes. A low-order technique capable of predicting the broadband noise of fans is developed here. In order to understand the influence of the blade curvature on the acoustic prediction, the case of the isolated fan is chosen to be studied numerically. Stationary 3D RANS simulations are computed as input data for a semi-analytical approach, based on Amiet’s theory, adapted to the rotational case, and extended to include blade sweep effects. The results obtained are in agreement with the experimental data at high frequencies. However, the stationary approach used does not allow the simulation of the large turbulent structures that develop at the tip of the blade, thus underestimating the low and medium frequencies. The sweep angle is a parameter that should be considered in the early stages of design, as it attenuates the noise emissions and changes the dominant sources along the blade.

Lien pour la visioconférence
Id : 732-339-949
This meeting is locked with a password : Phd_A-Zarri28


Ajouter un événement iCal