Co-Optimization of Structure and Manufacturable Semi-Continuous Layers for Laminated Composites

Tao Liu, Aoran Lyu, Yongxue Chen, Yu Jiang, Michael Petty, Charlie C.L. Wang†

†Corresponding author
The University of Manchester/ Digital Manufacture Lab

Paper Coming soon Code Coming soon Video

Co-optimized topology and curved semi-continuous layers for a woven fabric–reinforced laminated composite boat structure. The proposed optimization pipeline supports both complete and partial layers through the computation of semi-continuous periodic scalar fields, resulting in improved mechanical performance (reduced strain energy) and enhanced manufacturability, with well-controlled Gaussian curvature and layer thickness. For comparison, co-optimization results generated by NeuralTOMO using both (i) curved continuous layers and (ii) planar & parallel layers are also shown.

Abstract

To enable the design and manufacturing of optimized composite structures using fabric plies, we propose a field-driven optimization framework that jointly optimizes structural topology and manufacturable layers. A central challenge in this setting is the modeling and optimization of partial fabric layers with near-uniform thickness, which we formulate as a semi-continuous periodic scalar field parameterized by a continuous implicit neural vector field. Within this concurrent structure–layer optimization framework, we further derive a formulation of inter-layer anisotropic mechanical behavior that enables effective modeling of mechanical property transitions induced by partial-layer boundaries, together with additional objectives for manufacturability and field regularization. We validate the effectiveness of our approach through both numerical simulations and physical experiments, demonstrating that the optimized fabric-reinforced laminated composites achieve up to 43.8% higher stiffness compared to counterparts fabricated using planar fabric plies.

Method Overview

Computational pipeline of the proposed framework. Two continuous fields -- a density field and a vector field -- are parameterized by neural networks (NN) and jointly optimized in a self-learning loop via backpropagation. Mechanical performance is evaluated through FEA, while manufacturability is enforced using geometry-based loss terms. The optimized fields are then used to extract a collection of curved semi-continuous layers, which are flattened into planar panel layouts for the fabrication of laminated composites.

Results

Co-optimization results for the interior core structure of a laminated-composite propeller: (a) loading conditions and design domain of the propeller, (b) optimized interior structure and curved semi-continuous layers generated by our approach, (c) the ablation study result obtained by disabling the Gaussian curvature loss L_gc, i.e., using ω_gc = 0.0, and (d) convergence curves of the stiffness loss L_stf and histograms of Gaussian curvature for cases with and without the L_gc loss.

Results

Overview Figure

Co-optimized topology and curved semi-continuous layers for a woven fabric–reinforced laminated composite boat structure, resulting in improved mechanical performance and enhanced manufacturability with well-controlled Gaussian curvature and layer thickness.

Fabrication / Validation

The fabrication of laminated composites are taken on 3D printed moulds -- see an example of fabrication sequence, and all the physically fabricated models.

Tensile Test

Tensile test results for the Shell model. Specimens fabricated using our curved semi-continuous layers and the conventional planar & parallel layers are tested under compression.

Video

BibTeX

@article{LIU2026cooptimization,
  title   = {Co-Optimization of Structure and Manufacturable Semi-Continuous Layers for Laminated Composites},
  author  = {TAO LIU, AORAN LYU, YONGXUE CHEN, YU JIANG, MICHAEL PETTY,and CHARLIE C.L. WANG},
  journal = {ACM Transactions on Graphics},
  year    = {2026}
}

Contact

Tao LIU (tao.liu@manchester.com)

Charlie C.L. WANG (charlie.wang@manchester.ac.uk)