Active noise control, an alternative to conventional passive control, duplicates an existing noise field in antiphase so that when the noise and its antidote are added together the result is silence. This dissertation describes work aimed at implementing active control techniques on various noise problems. As each one was successfully completed confidence grew that the technique is on the verge of becoming a potential reality. The first project demonstrated the cancellation of the discrete frequency sound produced by a pair of large Electricity board transformers; the noise in a nearby office was reduced by some 20 decibels. Initial attempts at broadband cancellation demonstrated the necessity for an adaptable form of filtering. Digital techniques were developed to optimally control one-dimensional sound waves in an air-conditioning duct: some 20 decibels than three octaves of random sound. Their application to a practical problem at the Royal Signals and Radar Establishment, Malvern demonstrated their simplicity and. flexibility: it took only a day to design, and construct a device that stopped the low-frequency sound produced by a Diesel generator from entering an anechoic chamber. Further development has led to. control algorithms for a broadband adaptive controller. Its rapid convergence was demonstrated in an experiment which stopped the sound of a wind tunnel fan from disturbing sensitive measurements of turbulence in the test section. The control of three-dimensional sound presents many practical problems. One possible approach to these problems was presented by Olson in 1953 who described a device for cancelling local noise fields in order to produce a quiet zone in a noisy environment. We built an optimal form of that device and demonstrated some of its potential but its direct incorporation as one of many in an array was shown to result in an unstable system. Slight modification of that system has lead to the design of a stable array that effectively reflects sound. An investigation was carried out for Rolls-Royce into the possibility of using such an array to stop the broadband, lowfrequency sound, produced by two RB211 engine test-beds, from disturbing people in a large office 250 Metres away. The scale of the. solution was beyond the scope of a PhD. project. Finally a combination of all the techniques developed have been used to assess the possible performance of an active control system in a Fleet Street press room. The predictions are likely to be tested in a demonstration which is currently being planned. This research has helped reveal the major problems inhibiting the application of active noise control methods. Most of these have been overcome so that it should now be possible for the solutions outlined in this thesis to be applied relatively easily to design reliable and effective noise control systems.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:236368 |
| Date | January 1980 |
| Creators | Ross, C. F. |
| Publisher | University of Cambridge |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0026 seconds