Physics-Informed Transformer for Real-Time High-Fidelity Topology Optimization

TL;DR

This paper introduces a physics-informed transformer model that predicts optimal structural topologies in real-time, significantly reducing computational costs compared to traditional iterative methods.

Contribution

It presents a novel transformer architecture that directly maps physical inputs to optimized topologies, capturing long-range interactions and extending to dynamic scenarios.

Findings

Achieves higher fidelity than diffusion models with a single forward pass.

Eliminates iterative finite element analyses, enabling real-time topology optimization.

Extends to dynamic loading using frequency-domain encoding and transfer learning.

Abstract

Topology optimization is used for the design of high-performance structures but remains fundamentally limited by its iterative nature, requiring repeated finite element analyses that prevent real-time deployment and large-scale design exploration. In this work, we introduce a physics-informed transformer architecture that directly learns a non-iterative mapping from boundary conditions, loading configurations, and derived physical fields to optimized structural topologies. By leveraging global self-attention, the proposed model captures long-range mechanical interactions that govern structural response, overcoming the locality limitations of convolutional architectures. A conditioning-token mechanism embeds global problem parameters, while spatially distributed stress and strain energy fields are encoded as patch tokens within a Vision Transformer framework. To ensure physical realism and manufacturability, we incorporate auxiliary loss functions that enforce volume constraints, load adherence, and structural connectivity through a differentiable formulation. The framework is further extended to dynamic loading scenarios using frequency-domain encoding and transfer learning, enabling efficient generalization from static to time-dependent problems. Comprehensive benchmarking demonstrates that the proposed model achieves fidelity beyond that of diffusion models, while requiring only a single forward pass, thereby eliminating iterative inference entirely. This establishes topology optimization as a real-time operator-learning problem, enabling high-fidelity structural design with significant reductions in computational cost.