A parametric study of Element-Free Galerkin Method for room acoustics numerical simulation

This study investigates the influence of parameter settings on the computational results of the Element-Free Galerkin Method (EFGM) for room acoustics numerical simulation. Building upon previous work that confirmed its applicability to simple geometric problems, this research conducts numerical experiments to clarify the impact of various parameter configurations. The study compares EFGM calculations against analytical solutions and results of the Finite Element Method (FEM). The benchmark problems provided by the Architectural Institute of Japan were utilized for the analysis. Numerical experiments focused on the influence radius and background cell divisions. The influence radius significantly affected the frequency deviation of peaks from the analytical solution, suggesting an optimal value for accuracy exists. While reducing background cell divisions did not impact the frequency deviation of peaks, it considerably shortened the computational time for the global matrix assembly (e.g., from approximately 180 min for 100 divisions to 24 min for 50 divisions). This indicates that appropriately adjusting background cell divisions can improve computational efficiency without sacrificing accuracy. Additionally, the application of Modified Integration Rules showed improved accuracy in FEM but did not yield a significant improvement in EFGM. Further investigation into parameter settings is expected to improve the performance of EFGM for room acoustics numerical simulations.