Nowadays rotary turbomachines are widely used in various kinds of industry. However, it brings convenience as well as problems. The turbomachine makes great noise when it works, which badly effects people's life and living environment. There are mainly two kinds of noise of turbomachines – mechanical noise and flow noise. At presents, researches on flow noise received little attention. So researchers of Institute of Acoustics, Chinese Academy of Sciences, in collaboration with the researchers of Beijing Technology and Business University and CAS Institute of Engineering Thermophysics carried out a series of studies and proposed a noise source identification method to separate the noise source function and acoustic response function of the flow noise of tuibomachines enclosed by ducts.

Based on the acoustic similarity law, they obtain the noise source function and acoustic response function by least square multi linear regression method and Gauss filter respectively and conducte numerical simulation. The proposed method is proved to be effective. The simulation results show that the proposed method is much better than existing methods in terms of the relative errors of noise source function and acoustic response function. Two cases of a computer cooling fan working in free field and in dust are studied subsequently. They calculate the noise source function, acoustic response function and sound source tip speed exponent number by this method and analyze the main components of the sound source finally.