Topology Optimization Using Hyper Radial Basis Function Network

Abstract

In this paper we present the application of a hyper radial basis function network as a topology description function. A hyper radial basis function network is used to parameterize the material distribution (density) for topology optimization. The density of each finite element in the continuum topology optimization domain is governed by a single network of hyper radial function bases. Thus, the topology optimization problem is to determine the parameters governing the hyper radial basis function network to satisfy certain design criteria. Here we present the solution to a minimum compliance topology design. A hyper radial basis function network is used to parameterize the material density during the finite element analysis stage. An efficient optimization algorithm that makes use of perturbation and sequential linear programming is developed to obtain the hyper radial basis function network parameters. Examples are presented to demonstrate the proposed approach and to compare it with the traditional solid isotropic microstructures with penalization topology optimization. Some of the advantages of the proposed approach are that it can yield checkerboard-free, manufacturable topologies using coarse-mesh finite element analysis models as opposed to the traditional approach, which requires fine meshes. This results in a reduction in the solution time for finite element analysis while conserving the ability to yield a smooth topology.