Boundary element method virtual goniometer to predict the diffusion and scattering coefficients

Abstract

The experimental measurement of the diffusion coefficient according to ISO 17497-2 is very time consuming and requires several sample periods to evaluate the effect of diffraction lobes, an anechoic or large reflection-free volume and far-field conditions. Wave-based BEM methods have predicted diffusion and correlation scattering coefficients very accurately [Hargreaves et al., J. Acoust. Soc. Am. 108 (4), 1710–1720 (2000)]. This presentation will describe a new Python Virtual Goniometer program, called VIRGO, which predicts the free-field and surface-mounted diffusion and correlation scattering coefficients for any shaped surface that can be successfully meshed from a three-dimensional file. The program accurately predicts the periodic diffraction grating lobes of a reference one-dimensional hemicylinder and a two-dimensional hemisphere, both in the free-field and surface mounted on a boundary. The diffusion coefficient and three-dimensional polar responses of additional number-theoretic and optimized profiled and curvilinear shapes will also be compared with scale model boundary-plane goniometer measurements. The results will verify that it is possible to precisely predict the diffusion and correlation scattering coefficients, without having to fabricate or 3D print scale models or full-scale samples.