Enhancing the accuracy in reconstruction of vibro-acoustic responses of a complex structure using Helmholtz equation least squares based nearfield acoustical holography

摘要

Traditional nearfield acoustical holography uses the acoustic pressure measured at very close range to a source surface as input data. The source characteristics are determined by reconstructing the acoustic quantities on the source surface. The accuracy in reconstruction depends critically on the accuracy and the completeness of input data. In practice, constraints in measurements in terms of microphone spacing and standoff distance, and background noise can make input data neither complete nor accurate. Although the accuracy in reconstruction may be improved by using regularization, the incompleteness in input data cannot be adequately compensated. This paper demonstrates that this difficulty may be circumvented by supplementing the acoustic pressure with the normal surface velocity measured on a few points on a source surface. Using a combined acoustic pressure and normal surface velocity as input data to the Helmholtz equation least squares method based NAH, we may significantly improve the accuracy in reconstructing all vibro-acoustic responses on the surface of any vibrating structure. Experimental validations of this approach were conducted on a baffled rectangular, thin plate with free-free boundary conditions subjected to random excitations. Improvement in reconstruction accuracy was shown by comparing the reconstructed vibro-acoustic responses with the measured quantities.