Optimal Design of Polyester Thermal Fibre Structure Based on Parametric Geometric Model and Numerical Optimisation

Abstract

This study aims to optimise the cross-sectional shape and branching structure of polyester thermal fibres to enhance their thermal performance through parametric geometric modelling and numerical optimisation methods. Firstly, a parametric geometric model is used to describe the cross-section of different shapes and the heat transfer problem is analysed by finite element analysis (FEA) to select a suitable 2D or 3D model, which is combined with the numerical optimisation algorithm to find out the minimised thermal conductivity of 2.001. Secondly, key parameters such as the number of branches, angles, diameters, and lengths are used in this study, and a recursive algorithm is used to compute the thermal conductivity of each branch, and then to get the total thermal conductivity for the whole branch. The total thermal conductivity of the  entire branch is obtained. A numerical optimisation algorithm is used to optimise the combination of parameters to minimise thermal conductivity and generate a 3D image of the optimal branch structure. The shape parameters of the optimal branching structure, levels [5], base_length1, base diameter1, Angles [30, 30, 30, 30, 30], Lengths [1,1,1,1,1,1], and Diameters [0.5, 0.5, 0.5, 0.5, 0.5, 0.5], were obtained. The model construction in this study provides some theoretical support for generating the properties of polyester thermal fibres.