Investigation into Bias Errors of the Two-Microphone Cross-Spectral Method to Measure Acoustic Intensity.

摘要

The practical application of the two-microphone cross-spectral method of measuring acoustic intensity has recently gained great popularity in the field of noise control and soon will become the standard method of measuring the sound power of a noise source in-situ. The cross-spectral intensity technique contains inherent limitations which must be thoroughly understood before applying the measuring method in the field. These errors of the cross-spectral intensity method can be classified in terms of both bias and random errors, and if the coherence between the two receivers approaches unity and the random errors are sufficiently reduced by an appropriate amount of averaging by the FFT analyzer, the bias errors are then considered to be the major contributor to the deficiencies of the two-microphone intensity technique. These bias errors of the cross-spectral intensity technique have been theoretically analyzed in the literature but not completely verified experimentally. Therefore, the purpose of this work is to experimentally investigate the bias errors of the cross-spectral method of measuring acoustic intensity and also provide a current review of the technique. (Author)