In EHV power transmission

the value of audible noise under transmission lines is one of the main parameters for selection of conductor

so prediction of this audible noise is rather important in planning and constructing EHV transmission lines. In this paper

thirteen factors influencing the audible noise

such as environmental factor

geographical parameter and conductor’s structural parameter

are simplified using the grey-correlation analysis. The grey GM（1

9） and GM（0

9） prediction models based on the grey-correlation analysis are established through selecting main factors from the thirteen ones.Calculations indicate that the GM（1

9） model possesses higher prediction accuracy and smaller error

and its prediction accuracy is higher than that of BP neural network. Compared with other methods

the grey model is more effective with a higher degree of precision for audible noise prediction of EHV transmission lines.