Fast HdBNM for large-scale thermal analysis of CNT-reinforced composites

Abstract

Because of their high thermal conductivities, carbon nanotubes (CNT) have promising potential in development of fundamentally new composites. To study the influence of CNTs distribution on the overall properties of a composite, the modeling of a Representative Volume Element (RVE) including a large number of CNTs that are randomly distributed and oriented is necessary. However, analysis of such a RVE using standard numerical methods faces two severe difficulties, namely the discretization of the geometry and a very large computational scale. In this paper, the first difficulty is alleviated by employing the Hybrid Boundary Node Method (HdBNM), which is a form of the boundary type meshless methods. To overcome the second difficulty, the Fast Multipole Method (FMM) is combined with the HdBNM to solve a simplified mathematical model. RVEs containing various numbers of CNTs with different lengths, shapes and alignments have been analyzed, resulting in valuable insights gained into the thermal behavior of the composite material.