Smart foam is a composite noise control treatment that consists of a distributed piezoelectric actuator, known as polyvinylidene fluoride (PVDF), embedded within a layer of partially-reticulated polyurethane foam. The principal function of smart foam is to yield broadband sound attenuation. Passive acoustic foams are a very reliable high-frequency sound reduction method. With regard to smart foam, the embedded piezoelectric actuator is introduced to overcome the limitations of the passive foam in the low-frequency region. The piezoelectric actuator excites the structural and acoustic phases of the foam when driven by an externally supplied control voltage. This generates a secondary acoustic field which destructively interacts with the acoustic field created by a primary noise source. Initial experiments employ the composite "active/passive" treatment to yield attenuation of piston sound radiation. For this simple source, the global farfield pressure is minimized according to the feedforward, Filtered-x LMS control algorithm using one error sensor. Significant broadband sound attenuation is obtained. A more advanced noise control problem is investigated which minimizes plate radiation. The vibrating plate has a distributed modal response requiring a collective array of independently-phased smart foam actuators to yield reduction of the radiated sound power. This is accomplished by minimizing the sound pressure at an array of nearfield microphones. Good broadband sound power reduction is obtained using a MIMO (multiple-input/multiple-output) Filtered-x LMS control scheme. Various techniques for improving smart foam's acoustic control authority are identified during manufacturing and finite element modeling. of the actuator. These improved smart foam actuators are employed as an active/passive liner to suppress the transverse propagating acoustic modes within an anechoically-terminated rectangular duct. A section of a duct wall is lined with an array of smart foam and the sound downstream of the control actuators is minimized at several error microphones. Successful harmonic and broadband noise control is achieved. A full-scale numerical model of the duct acoustic control application is presented based on the finite element method. The purpose of the model is to study the sensitivity of this active/passive control approach relative to the spatial distribution of control channels and error sensors. A comparison of the numerical and experimental results yields similar trends. / Ph. D.

In this work, the potential for a smart foam actuator in controlling interior cabin noise due to a turbulent boundary layer excitation has been experimentally demonstrated. A smart foam actuator is a device comprised of sound absorbing foam with an embedded distributed piezoelectric layer (PVDF) designed to operate over a broad range of frequencies. The acoustic foam acts as a passive absorber and targets the high frequency content, while the PVDF serves as the active component and is used to overcome the limitations of the acoustic foam at low frequencies. The fuselage skin of an aircraft was represented by an experimental test panel in an anechoic box mounted to the side of a wind tunnel. The rig was used to simulate turbulent boundary layer noise transmission into and aircraft cabin. An active noise control (ANC) methodology was employed by covering the test panel with the smart foam actuators and driving them to generate a secondary sound field. This secondary sound field, when superimposed with the panel radiation, resulted in a reduction in overall sound in the anechoic box. An adaptive feedforward filtered-x Least-Mean-Squared (LMS) control algorithm was used to drive the smart foam actuators to reduce the sound pressure levels at an array of microphones. Accelerometers measured the response of the test panel and were used as the reference signal for the feedforward algorithm. A detailed summary of the smart foam actuator control performance is presented for two separate low speed wind tunnel facilities with speeds of Mach 0.1 and Mach 0.2 and a single high speed tunnel facility operating at Mach 0.8 and Mach 2.5. / Master of Science

Les matériaux absorbants ont une efficacité très faible aux basses fréquences (<500Hz). Les mousses dites adaptatives ont été développées pour combler ce manque d'efficacité. Une mousse adaptative est la réunion d'un matériau absorbant et d'un actionneur commandé par un système de contrôle actif. L'objectif de cette thèse est de mener une étude numérique et expérimentale approfondie du concept de mousse adaptative et d'en dégager les mécanismes physiques et les limitations technologiques pour le contrôle de l'absorption acoustique. Le matériau absorbant est ici une mousse de mélamine et l'actionneur est un film piézoélectrique de PVDF. Le contrôle actif de l'absorption acoustique est réalisé en incidence normale avec l'hypothèse d'onde plane sur la plage de fréquence [0-1500Hz]. Les résultats font apparaître la possibilité d'absorber une pression incidente de 1Pa à 100Hz avec 100V et un bruit large bande de 94dB avec une centaine de Vrms à partir de 250Hz. Ces résultats ont été obtenus avec un prototype de mousse adaptative ayant une épaisseur moyenne de 4cm. Un frein important au contrôle large bande provient du fort niveau de distorsion des prototypes dans le bas (<500Hz) et le haut (<1500Hz) du spectre de fréquences. Un modèle éléments finis 3D couplant les domaines poroélastiques, acoustiques, élastiques et piézoélectriques a été proposé et validé expérimentalement. L'utilisation de ce modèle, complété par une étude analytique, a permis de mettre en lumière le mode d'action et les mécanismes de dissipation dans la mousse adaptative.