For the adverse effects of a large amount of background noise on the test results in the wind tunnel test, a noise separation method based on double arrays is proposed. Firstly, all back-ground noise is equivalent to an interference sound source, which is located on two planes re-spectively with the target sound source. Two planar arrays are inserted between the target sound source and the interference sound source. Next, by calculating the sound pressure measured on two arrays jointly, the data of the target sound source radiated to an array sur-face alone is obtained. Finally, the nearfield acoustic holography algorithm in moving fluid medium is used to reconstruct the target sound source, achieving the separation of the target sound source and background noise. In addition, a hologram pressure extrapolation technique using back propagation neural network is introduced to reduce the window effect by the finite aperture and the wraparound error due to the discrete microphone array. Compared with the traditional noise separation methods, the proposed method does not require a priori knowledge and is easy to use, yet of strong robustness and high accuracy. Numerical simulations further verify the feasibility and effectiveness of this method.
