Active control of sound radiated from a vibrating plate

This dissertation is concerned with the design of active noise control systems. In particular the response of a plate with different boundary conditions with PZT actuators and distributed sensors is discussed. The calculation of the sound power radiated from a plate with different boundary conditions, the design of a distributed sensor system which can sense the sound power radiated from the vibrating plate, the optimal placement of the PZT actuators and the control algorithms are considered. In describing the development of active noise control systems, many techniques are reviewed in some detail.; Based on the controllers and sensors used, active noise control systems can be classified into two main approaches. One approach is to use a microphone and loudspeaker pair to control the noise. The other approach is to use a PZT actuator and PVDF sensor pair to control the vibration of the structure and then reduce the noise.; For the first approach, an inverse FXRLS algorithm was developed. This algorithm was implemented in a feedback control system. Traditionally, FXLMS algorithms are used, and the secondary path is measured with a feedback system. In this study, the FXLMS algorithm was replaced with an FXRLS algorithm. Instead of measuring the secondary path, we measured the inverse of the secondary path. The simulation and experimental results showed that the inverse FXRLS algorithm outperformed the traditional FXLMS algorithm.; For the second approach, a control system was designed. The disturbance is assumed to contain multiple frequency components. The objective is to minimize the sound power radiated from a vibrating plate with different specified boundary conditions. The response of the plate with different boundary conditions to a rectangular PZT actuator is derived. The optimal placement of the PZT actuator is discussed. A technique to design a sensor system which can be used to measure the sound power from a vibrating structure is presented. A multiple-channel control algorithm with sound power as a cost function was implemented. Finally a noise control system was designed.