Diverse Topology Optimization using Modulated Neural Fields

Andreas Radler, Eric Volkmann, Johannes Brandstetter, Arturs Berzins

Track: Track 1: Original Research/Position/Education/Attention Track

Keywords: Geometry, Topology Optimization, Solver-in-the-loop, Theory-informed learning

TL;DR: Topology Optimization using Modulated Neural Fields (TOM) is a novel, data-free method using neural networks to generate diverse, structurally near-optimal shapes.

Abstract: Topology optimization (TO) is a family of computational methods that derive near-optimal geometries from formal problem descriptions. Despite their success, established TO methods are limited to generating single solutions, restricting the exploration of alternative designs. To address this limitation, we introduce *Topology Optimization using Modulated Neural Fields* (TOM) – a data-free method that trains a neural network to generate structurally compliant shapes and explores diverse solutions through an explicit diversity constraint. The network is trained with a solver-in-the-loop, optimizing the material distribution in each iteration. The trained model produces diverse shapes that closely adhere to the design requirements. We validate TOM on 2D and 3D TO problems. Our results show that TOM generates more diverse solutions than any previous method, all while maintaining near-optimality and without relying on a dataset. These findings open new avenues for engineering and design, offering enhanced flexibility and innovation in structural optimization.

Submission Number: 127