GM-TOuNN: Graded Multiscale Topology Optimization using Neural Networks

TL;DR

This paper introduces GM-TOuNN, a neural network-based framework for graded multiscale topology optimization that reduces computational costs and efficiently manages multiple microstructures while satisfying design constraints.

Contribution

It presents a novel neural network approach for GM-TO that handles numerous microstructures efficiently and enforces partition of unity without manual sensitivity analysis.

Findings

Reduces computational cost compared to traditional M-TO

Ensures partition of unity and discourages microstructure mixing

Supports automatic differentiation for sensitivity analysis

Abstract

Multiscale topology optimization (M-TO) entails generating an optimal global topology, and an optimal set of microstructures at a smaller scale, for a physics-constrained problem. With the advent of additive manufacturing, M-TO has gained significant prominence. However, generating optimal microstructures at various locations can be computationally very expensive. As an alternate, graded multiscale topology optimization (GM-TO) has been proposed where one or more pre-selected and graded (parameterized) microstructural topologies are used to fill the domain optimally. This leads to a significant reduction in computation while retaining many of the benefits of M-TO. A successful GM-TO framework must: (1) be capable of efficiently handling numerous pre-selected microstructures, (2) be able to continuously switch between these microstructures during optimization, (3) ensure that the partition of unity is satisfied, and (4) discourage microstructure mixing at termination. In this paper, we propose to meet these requirements by exploiting the unique classification capacity of neural networks. Specifically, we propose a graded multiscale topology optimization using neural-network (GM-TOuNN) framework with the following features: (1) the number of design variables is only weakly dependent on the number of pre-selected microstructures, (2) it guarantees partition of unity while discouraging microstructure mixing, and (3) it supports automatic differentiation, thereby eliminating manual sensitivity analysis. The proposed framework is illustrated through several examples.