Search

During takeoff, the harmonic noise of the turbofan is the main acoustic nuisance for people near airports. Much research has been done to achieve active noise control of turbofans with loudspeakers or piezo actuators as noise suppression sources. However, the required fragility, and weight and volume penalty make them unsuitable for engine nacelle applications. On the other hand, electro-pneumatic sound sources (sirens or air modulators) are good candidates for generating the required high noise control sound pressure level, but they are not designed for active noise control applications. Therefore, an alternative solution, called the Harmonic Acoustic Pneumatic Source (HAPS), has been designed to generate a high harmonic noise level controllable in amplitude, phase and frequency. The specificity of the HAPS is to use a flow chopper to generate a pulsed jet at a given frequency and a servo-valve to control its amplitude. This study presents the HAPS composed of a high-pressure pneumatic air source, a flow chopper using a rotating perforated cage and a dedicated exhaust. The fundamental equation of the pneumatic speaker is used to define the time-varying orifice as the generator of a periodic pressure source with a built-in internal impedance. To implement phase control of the generated sound, a phase-locked loop (PLL) controls the instantaneous angle of the rotating cage. Thus, control of the HAPS requires 2 slow-time signals: one for amplitude (position of the servo-valve) and the second for phase (via the PLL). However, for active noise control, the HAPS control is presented as a mechanical modulator driven by two slowly varying signals (the real and imaginary parts of the complex value envelope of the command). The mechanical time responses of the servo valve and the rotating cage motor imply a narrow bandwidth centered on the modulation frequency. The experimental results presented are obtained on a dedicated test bench for different frequencies. Final, "Résumé de la communication présentée lors du congrès international tenu conjointement par Canadian Society for Mechanical Engineering (CSME) et Computational Fluid Dynamics Society of Canada (CFD Canada), à l’Université de Sherbrooke (Québec), du 28 au 31 mai 2023.

La reproduction de champs sonores est un domaine qui naquit avec l'émergence du téléphone il y a près de 150 ans et qui est en constante évolution depuis. Ce domaine s’intéresse aujourd’hui à des réseaux complexes de capteurs ou de sources, permettant d’enregistrer ou de reproduire des scènes sonores en 3-D. Selon le contexte d'étude, qu'il soit médiatique, artistique ou scientifique, les utilisateurs et les concepteurs de ces technologies peuvent privilégier : soit une reproduction subjective, qui favorise l'auditeur ; soit une reproduction objectives, basée sur des critères physiques tel que la pression acoustique. De plus, le nombre de ces techniques de reproduction ne cesse de croitre, de même que le nombre de canaux audio les constituant. On parle aujourd'hui de systèmes multicanaux massifs pour certains réseaux dépassant parfois la centaine de sources de reproduction. Dans un contexte d’ingénierie, on s’intéresse principalement à la reproduction de critères objectifs. Les deux principales méthodes proposées dans la littérature sont la Wave Field Synthesis et l’Ambisonie d’ordre élevé ; cette dernière étant la méthode étudiée tout au long de ce manuscrit. Il est observé dans la littérature que l'ambisonie permet une reproduction exacte au centre du réseau de sources. De plus, la taille de cette région valide, pour un nombre de source donné et donc à un ordre ambisonique précis, diminue lorsque la fréquence à reproduire augmente. Ainsi, pour augmenter la taille de la RRP, il est de coutume d’augmenter l’ordre de la reproduction ambisonique et intrinsèquement, d’augmenter le nombre de sources de reproduction. Cette thèse a pour vocation de minimiser le nombre de sources additionnelles, dans le cas où l'on souhaiterait étendre cette RRP. Pour cela, des travaux portant sur la captation ambisonique, l’évaluation de champs sonores et la reproduction de champs sonores, sont présentés. Tout d’abord, on présente une méthode d’extrapolation de la vitesse particulaire à p, The reproduction of sound fields is a domain that was born with the invention of the telephone nearly 150 years ago and has been in constant evolution ever since. Today, this field focuses on complex arrays of sensors or sources enabling not only to record, but also to reproduce 3-D sound scenes. According to the context, be it media, artistic or scientific, users as well as designers of these technologies can prefer either a reproduction that favors the preferences of the listeners or a reproduction that rather respects objective criteria; the most used criterion in this case being, of course, the sound pressure at any point in space. Moreover, the number of these reproduction techniques is constantly increasing, as is the number of audio channels required. We then speak of massive multichannel systems for certain arrays exceeding tens or even hundred distinct reproduction sources. In engineering contexts, we are mainly interested in the reproduction of objective criteria. The two main methods proposed in the literature are Wave Field Synthesis and Higher-Order Ambisonic; the latter being the method studied throughout this manuscript. The literature suggests that this method allows exact sound pressure reproduction at the center of the source array ; or at least reproduction below a certain error threshold. This region is called accurate reproduction area. Moreover, the size of this region is inversely proportional to the reproduced frequency. Thus, in order to increase the size of this precise reproduction region, it is common to increase the order of the ambisonic reproduction and, intrinsically, to increase the number of reproduction sources. This thesis aims to minimize the number of additional sources, in the case where one wants to extend this accurate reproduction area. For this purpose, works on ambisonic capture, sound field evaluation and sound field reproduction are presented. First of all, a method of extrapolation of particle velocity from t