Self-tuning and adaptive controllers for active sound and vibration control

This thesis details the research undertaken to find new self-tuning and adaptive controllers for active sound and vibration control. During the previous 30 years, a lot of research work has been carried out on active attenuation of unwanted noise and vibration. The majority of researchers have concentrated on finding models to describe systems or on the implementation of the well studied filtered reference least mean square algorithm. In this thesis, several new tuning schemes are presented for the active attenuation of tonal disturbing sound or vibration. The new schemes cover control of single input single output (SISO) as well as multiple input multiple output (MIMO) systems. Iterative self-tuning controllers were developed operating in the time domain. In the frequency domain, iterative self-tuning and adaptive model/model free algorithms were researched. Their advantages compared to common controllers are stressed. The algorithms are tested in simulation and the results are analysed in both the time and frequency domains. In simulations, it was possible to show the working mechanism of the proposed controllers. Convergence tracks are shown and the achieved control results were compared. To show the practicability of the presented schemes, the SISO and MIMO controllers were implemented on a laboratory glass vibration and sound control experiments (active controlled headset and speaker/microphone arrays). The achieved results have clarified the usefulness of the approaches taken. Especially for the sound control experiments the attenuation was highly noticeable due to the reduction of around or more I than 20 dB