Diffusion Transformers with Hybrid Conditioning for Structural Optimization

2026-05-03

AI Summary

This work proposes a hybrid conditional diffusion Transformer model to overcome the computational inefficiency and limited scalability of traditional topology optimization methods such as SIMP, which rely on repeated finite element analyses. The model uniquely integrates spatially distributed stress–strain fields with global scalar conditions—including applied loads and volume fraction—to directly generate near-optimal topologies from problem specifications without iterative refinement during inference. Leveraging AdaLN conditioning, concatenated field inputs, and deterministic DDIM sampling, the model is trained on 30,000 two-dimensional examples and produces high-quality designs in seconds with only five denoising steps. The generated structures exhibit compliance errors below 1% relative to SIMP ground truths, precise adherence to target volume fractions, and excellent connectivity.

Abstract

This work presents a diffusion transformer framework for data-driven structural topology optimization that combines the accuracy of physics-based methods with the efficiency of generative deep learning. Conventional approaches such as the Solid Isotropic Material with Penalization (SIMP) method require repeated finite element analyses at every iteration, making large-scale or real-time optimization computationally expensive. We propose a hybrid conditioning diffusion transformer (DiT) model that learns to generate near-optimal topologies directly from problem definitions, eliminating iterative analysis during inference. The model integrates spatially distributed conditioning through concatenated stress and strain fields and global conditioning via adaptive layer normalization (AdaLN) using scalar descriptors such as load position, magnitude, and prescribed volume fraction. A dataset of 30,000 two-dimensional SIMP-optimized structures was generated for training and evaluation. Results demonstrate that the proposed DiT achieves less than 1% compliance errors relative to ground-truth SIMP solutions while maintaining accurate volume fractions and structural connectivity. Deterministic DDIM sampling enables high-fidelity topology generation in seconds using as few as five denoising steps, enabling near-real-time performance. The hybrid conditioning diffusion transformer thus provides an efficient and scalable alternative to traditional topology optimization methods, with strong potential for integration into interactive computer-aided design workflows.

Problem

• structural topology optimization
• computational efficiency
• real-time optimization
• finite element analysis
• SIMP

Innovation

• Diffusion Transformer
• Hybrid Conditioning
• Topology Optimization
• AdaLN
• DDIM Sampling