University of Twente Student Theses

Login

Active noise reduction in double-panel structures : decentralized adaptive feedforward control

Kalverboer, J. (2012) Active noise reduction in double-panel structures : decentralized adaptive feedforward control.

[img] PDF
2MB
Abstract:In this thesis an adaptive harmonic decentralized feedforward controller is implemented on an experimental setup. The experimental setup consists of a double-panel structure which is mounted on top of an acrylic box with thick walls. In this box a speaker is installed which can produce disturbance signals. On each panel of the double-panel structure five actuator sensor pairs are installed which are controlled and observed by their own decentralized controller .The goal of this thesis is to minimize the sound transmission of the double-panel structure. The damping of the panel is increased by adding a low-authority feedback controller. The disturbance signal will contain only a few deterministic frequency components. Each frequency component is processed independently, by the harmonic decentralized controllers on the panels. The stability and convergence rate of decentralized feedforward control is analysed. In simulations it is demonstrated that a decentralized controller can reduce the noise transmission of the structure. There are multiple feedforward control configurations possible, for example, one strategy might only control the incident panel. The most noise reduction is obtained by controlling only the radiant panel. This configuration can reduce the sound pressure level above the panel by 10 to 20 dB for frequencies below 250 Hz. Above this frequency almost no reduction of the disturbance signal is possible. The addition of feedback control improves the robustness of the system. It reduces the cross-coupling between the decentralized controllers. There are also multiple feedback control configurations possible. The combination of feedback pressure speakers in the cavity and decentralized feedforward control of the radiant panel reduces the disturbance signal the most in the simulations. In practise this configuration could not be tested, because the pressure speaker feedback controller has not yet been realized. For optimal control of the double-panel structure the transfer functions between all actuators and sensors in the system must be known. Therefore, a system identification technique is developed. This technique uses the computational components available in the decentralized controllers. The communication with a centralized controller is minimized.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Electrical Engineering MSc (60353)
Link to this item:http://purl.utwente.nl/essays/69662
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page