This analysis demonstrates improved convergence using a reduced-order model in acoustic applications, highlighting efficiency gains in memory usage.
Broadband acoustic scattering with ground reflection poses significant computational challenges due to repeated matrix assembly in conventional boundary element method (BEM). To overcome this, a rapid frequency-sweeping strategy combining the isogeometric boundary element method (IGBEM), Taylor series expansion, and second-order Arnoldi (SOAR) model reduction is developed. The NURBS-based IGBEM ensures geometric accuracy and continuity, while the Burton-Miller formulation maintains numerical stability. The Taylor expansion decouples frequency-dependent terms, enabling fast matrix reassembly, and the SOAR algorithm constructs a compact reduced-order model capturing dominant acoustic modes with greatly reduced computational cost and memory usage. Validations on rigid cylinders and semi-Y-shaped noise barriers, accounting for ground reflection, confirm that increasing Taylor expansion order and using adaptive frequency subdivision enhance accuracy and convergence. Compared with conventional BEM, the proposed method achieves orders-of-magnitude acceleration with minimal accuracy loss, offering an efficient and scalable framework for broadband acoustic scattering of complex geometries and laying the foundation for future 3D and large-scale extensions.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: