Neural model speeds material-efficient 3D lattice refining by 5000 times
SLIM: Learning an Operator for Material-Efficient 3D Lattice Refinement
Graphics
Summary
Making 3D lattice structures lighter without losing their strength usually takes a long time because it involves complex step-by-step adjustments for each design. The authors created SLIM, a neural network that learns from previous optimized designs to quickly predict how to remove excess material while keeping the lattice stiff. This approach cuts down the time from over 40 minutes to about half a second per design. Their model also keeps most of the original strength and reduces material by nearly 16% on average.
What this means in practice
- •For mechanical design engineers: Rapidly generate lightweight yet stiff 3D lattice components without costly per-design optimization iterations.
- •For additive manufacturing teams: Produce optimized lattice infills for 3D printed parts faster by predicting material reductions accurately.
Authors
Yueze Zhu, Yu Xing, Tianyang Xue, Peng-Shuai Wang, Jikai Liu, Haisen Zhao, Lin Lu
Abstract
Lattice inverse-design methods can generate cells that match prescribed effective properties, but property feasibility does not imply material efficiency. Removing redundant material from an existing lattice while preserving its homogenized elasticity ordinarily requires solving a separate constrained topology optimization problem for each cell, making large-scale refinement computationally expensive. We introduce SLIM (Structural Lightweighting via Implicit Mapping), a neural lattice-refinement model that replaces this per-instance iterative optimization with a single forward prediction. SLIM uses a dual-octree implicit network to predict the signed-distance field of a lightweight lattice and is trained on input--output pairs generated by property-constrained topology optimization. To improve stiffness preservation, we use homogenization sensitivities computed on the optimized target to estimate how residual voxel errors in a prediction affect its homogenized stiffness, and penalize this estimated change during training. Once trained, SLIM requires no topology optimization, homogenization analysis, or sensitivity evaluation during prediction. On the held-out test set of 906 lattices, SLIM achieves a mean relative volume-fraction reduction of 15.90%, a lightweighting success rate of 90.51%, and a mean normalized aggregate TO-target error of 0.636%, while reducing the average per-lattice generation time from 2516.0 s with direct TO to approximately 0.5 s. These results show that SLIM provides a fast neural surrogate for repeated topology-optimization-based refinement of 3D lattices.