An efficient adaptive frequency sampling scheme for large-scale transient boundary element analysis

TL;DR

This paper introduces an adaptive frequency sampling scheme that significantly reduces computational time in large-scale transient boundary element analysis by decreasing the number of sampling frequencies without sacrificing accuracy.

Contribution

A novel adaptive frequency sampling algorithm that automatically selects sampling frequencies, halving the computational cost of frequency-domain boundary element methods.

Findings

Reduced sampling frequencies by over 50%.

Achieved fourfold reduction in total computational time.

Maintained high accuracy in large-scale porous solid models.

Abstract

The frequency-domain approach (FDA) to transient analysis of the boundary element method, although is appealing for engineering applications, is computationally expensive. This paper proposes a novel adaptive frequency sampling (AFS) algorithm to reduce the computational time of the FDA by effectively reducing the number Nc of sampling frequencies. The AFS starts with a few initial frequencies and automatically determines the subsequent sampling frequencies. It can reduce Nc by more than 2 times while still preserving good accuracy. In a porous solid model with around 0.3 million unknowns, 4 times reduction of Nc and the total computational time is successfully achieved.