CBF Computation Acceleration of CBFM Using k-Means Clustering Algorithm

Abstract

This study utilized the clustering-based CBF(characteristic basis function) generation algorithm to reduce the computing time of CBFM(characteristic basis function method), which is an accelerated version of moment methods in electromagnetics. As an efficient optimization of block generation, the meshes are grouped using the clustering algorithm, resulting in a reduction in the CBF computing time. In this study, we optimized the mesh distribution using the K-means clustering algorithm and verified the performance of the clustering-based method.