Building an active noise control demonstrator

Hekman, T. (2016) Building an active noise control demonstrator.

[img] PDF
Restricted to Restricted

12MB
Abstract:During measurements on an offshore production platform, problematic noise levels were detected in the sleeping quarters. Not only had the inhabitants of the platform expressed their annoyance with the noise generated by the compressors on the platform, but the measured values also exceeded 45 dB(A), which is a commonly used maximum acceptable level in bedrooms. The low frequent nature of the noise makes it almost impossible to reduce the levels using passive means. To be able to reduce the noise levels in the bedrooms without having to make structural changes to the rig, an active noise control system was tested. A model of a bottom bunk was built and an ANC algorithm with online system identification was implemented on an EVAL-21489-EZKIT DSP board augmented with an EVAL-ADAU1772 audio codec board. Experiments were done to evaluate the performance of the ANC system, both in terms of maximum noise reduction, the size of the area of silence and the functionality of the online system identification. It was found that the ANC system was capable of reducing the noise levels by 14.59 dB(A), from 58.65 dB(A) to 44.06 dB(A). This showed that the system was capable of ful�filling its main objective: reducing the noise to acceptable levels (anything below 45 dB(A)). Running the system using the ADAU1772 codec, instead of the built-in AD1939 codec, yielded an improvement in performance of more than 2 dB(A). The area of silence, in which the noise levels were below the 45 dB(A) threshold, was found to spread out between 30 and 45 cm away from the head end of the cabin. This is large enough to contain the head of a sleeping person. Finally, it was verified that, thanks to its online system identification, the algorithm could remain stable when changing the secondary path in a way that would have been likely to induce instability in a system with offline system identification. The tests indicate that ANC is a viable solution for the noise problem on the offshore rig.
Item Type:Internship Report (Master)
Faculty:ET: Engineering Technology
Subject:52 mechanical engineering
Programme:Mechanical Engineering MSc (60439)
Link to this item:http://purl.utwente.nl/essays/72017
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page