Modelling the Effective Thermal Conductivity of Particulate Reinforced Composites using APDL and Genetic Programing

Abstract

The present study is to develop a computational model to find out the effective thermal conductivity of a particulate reinforced composite using genetic programming. To develop the genetic programming model, the values from the numerical simulation of finite element modeling using ANSYS Parametric Design Language (APDL) is used. To implement the concept of genetic programming, DISCIPULUS software is used. The numerical study is conducted in a three-dimensional cube lattice array model with particle fillers having the geometry of sphere. The maximum packing fraction was found out to be 52.4%. To find the effective thermal conductivity values, eleven different Kf/Km ratios from 0.01 to 10000 and concentration ratios ranging from 1 to 52.4% are used in APDL. The values generated in APDL are loaded in the genetic programming software to develop the genetic programming model. The results indicate the efficacy of genetic programming to predict the effect of multiple input parameters on the thermal properties of particulate reinforced composite. The values predicted from Genetic Programming model are compared with existing theoretical and empirical models.

Keywords: Genetic Programming, APDL, Thermal Conductivity, Particulate reinforced composite