Study of active noise control system for a commercial HVAC unit.

摘要

Acoustic noise is a common problem in everyday life. If the appliances that are present in the work and living areas generate noise then it's a serious problem. One such appliance is the Heating, Ventilation and Air-conditioning system (HVAC) in which blower fan and compressor units are housed together. Operation of a HVAC system creates two kinds of noise. One is the noise due to the air flow and the other is the result of the compressor. Both of them exhibit different signal properties and need different strategies to control them. There has been previous efforts in designing noise control systems that can control noise from the HVAC system. These include passive methods which use sound absorption materials to attenuate noise and active methods which cancel noise by generating anti-noise. Passive methods are effective in limiting the high frequency noise, but are inefficient in controlling low frequency noise from the compressor. Compressor noise is one of the strong low frequency components that propagate through the walls, therefore there is need for deploying active signal processing methods that consider the signal properties into consideration to cancel the noise acoustically. The quasi periodic nature of the compressor noise is exploited in noise modeling which aids in implementing an adaptive linear prediction filter in estimating the anti noise [12]. In this thesis, a multi channel architecture has been studied for a specific HVAC system in order to improve noise cancellation by creating larger quiet zone. In addition to the multi-channel architecture, a real time narrow band Active Noise Control (ANC) was employed to cancel noise under practical conditions.