Active control of sounds, vibrations and fluid-structure interactions

点击翻译按钮获取中文摘要

The domain of my research program will be in the design of devices and controllers for active sound control (ASC), active vibration control (AVC) and control of fluid-structure interaction (FSI). The term control means attenuation of sound and vibration, but it also can be used for the enhancement of the oscillation. The axis on the devices will be on the development of a new high level subband acoustic pneumatic source (e.g. a pneumatic loudspeaker working with compressed air) able to generate a perfectly controlled time-varying amplitude and phase at a given carrier frequency. The challenge will be to drive the instantaneous phasor of the pneumatic source in order to generate a narrowband stochastic signal. The main practical interest of such pneumatic loudspeaker will be to attenuate high level narrow-band sounds in harsh environment (e.g. turbofan). Such new source can also be used for the study of the fluid-structure interaction in mufflers, the characterization of liners, the observation of the mouth noise generation, or for active control of fluid. The axis on the controller will focus on the command of multiple actuators in order to attenuate a pressure sound in far field, but with microphone measurements at a close distance of the actuators. Typically, the control objective will be to minimize the radiated acoustic power which is not equivalent to minimize a norm of outpu. The objective is to develop a generic experimental method to learn (or identify) the internal model that will allow to use the outputs (y) to control the inputs (u) in order to minimize a norm of z. The research will define the limitations of the method in order to provide guidelines about the number and the location of actuators and sensors, and the impact of uncertainties. The method will be first limited to harmonic signal. In practical applications, this research will give knowledge for the design of ASC/AVC with multiples actuators and sensors. Because it will be a generic method, it could be translated in the design of active meta-materials. The axis of FSI will concern the enhancement, or the attenuation, of the oscillatory (or unstable) response of a flexible tube filled with a liquid. The manipulated variable is the unstationnary flow rate, and the observed variable is the parietal pressure. An experimental set-up will be developed to generate different dynamics. A grey-box non-linear modeling approach will be developed to identify the nonlinear model, and to design observers. Finally, an extremum seeking controller will be designed to enhance or to attenuate the instability. The practical main motivation for such active control of FSI is the study of the collapse of the trachea during expiration in total liquid ventilation (when the lungs is ventilated with a breathable liquid). However, the developed knowledge and the experimental set-up will be used in other studies related to similar physiological problems.