KATOsuper is an objective‑agnostic framework that accelerates neural topology optimization by coupling neural‑reparameterized TO with a Sensitivity‑Consistent Fourier Neural Operator (SC‑FNO). It uses a forward_split architecture to ensure that sensitivities derived via automatic differentiation remain consistent with predicted objectives, enabling stable optimization. The method demonstrates significant deployment‑time speedups (15–110×) over MATLAB baselines while preserving optimality across 2D and 3D benchmark problems, including compliance and stress minimization, and supports zero‑shot extrapolation to higher resolutions.
By Shengyu Yan, Jasmin Jelovica
arXiv:2610.02214v1 Announce Type: cross
Abstract: Ship structures govern vessel strength, safety, and manufacturability, but their design must satisfy hundreds of classification society requirements,...
By Noah J. Bagazinski, Md. Ferdous Alam, Jaya Manideep Rebbagondla, Faez Ahmed
arXiv:2609.15001v1 Announce Type: new
Abstract: Density-based topology optimization is typically structured as a nested sequence of material updates, structural analyses, and sensitivity assessments....
By Kangzheng Liu, Uday Kumar Punna, Leixin Ma
arXiv:2607. 14652v1 Announce Type: new Abstract: Topology optimisation (TO) often requires repeated finite element analysis and sensitivity-based material updates, which can be costly when multiple candidate designs are needed under varying physical and design conditions.
By Shusheng Xiao, Jinshuai Bai, Hyogu Jeong, Yunfei Xi, Yilin Gui, YuanTong Gu
The paper introduces DA‑EGO, an efficient global optimization algorithm that dynamically aggregates high‑dimensional design spaces into low‑dimensional subspaces for surrogate‑based search. The algorithm updates subspace variables each iteration using variable‑interaction analyses, perturbation, and ANOVA, and adaptively adjusts search ranges based on previous results. Tests on 21 benchmark functions and real turbomachinery problems demonstrate DA‑EGO’s effectiveness, especially on separable and partially separable problems, while noting case‑dependent performance on non‑separable functions.
By Qineng Wang, Zhendong Guo, Yun Chen, Guangjian Ma, Liming Song, Jun Li
The inverse design of physical systems governed by partial differential equations is computationally demanding due to the high dimensionality and non-convexity of design spaces. Generative models for inverse design often lack robustness and transferability, whereas evolutionary strategies are robust but struggle in high-dimensional spaces.
arXiv:2606. 19921v1 Announce Type: new Abstract: This work proposes an element-based Convolutional Neural Network (CNN) to accelerate density-based Topology Optimization (TO), termed eCNNTO.
By Shengbiao Lu, Xiaodong Wei
arXiv:2607. 07682v1 Announce Type: new Abstract: The inverse design of physical systems governed by partial differential equations is computationally demanding due to the high dimensionality and non-convexity of design spaces.
By Xiangming Huang, Guannan Zhang, Lu Lu, Rapha\"el Pestourie
arXiv:2607. 24777v1 Announce Type: new Abstract: Architected metamaterials derive their functions from structure, creating vast opportunities to program physical responses through topology design.
By Haolin Li, Yuyang Miao, Menglei Li, Jinshuai Bai, Liyuan Wang, Xin Liu, Bo Gao, Jiantao Liu, Danilo Mandic, Zahra Sharif Khodaei, M. H. Aliabadi, Weiqiu Chen
The paper proposes using neural networks to replace the iterative force evaluation in the harmonic balance method for systems with nonlinear contacts and friction. These networks map displacement Fourier coefficients directly to nonlinear force coefficients and supply Jacobians via automatic differentiation, allowing the existing solver and continuation algorithms to remain unchanged. By learning individual nonlinear elements—such as cubic, unilateral, and Jenkins springs—under physics‑based nondimensionalization and phase normalization, a single trained network can handle a wide range of parameters, enabling a reusable library of nonlinear‑element surrogates for complex mechanical systems.
By Miriam Goldack, Johann Gro{\ss}, Malte Krack, Merten Stender
arXiv:2609.07437v1 Announce Type: cross
Abstract: Physics-informed neural networks (PINNs) represent a growing frontier in using artificial intelligence to solve partial differential equations (PDEs)...
By Xing Guo, Hongwei Tang, Zewei Meng, Yidong Zhang, Shaoqiu Xiao, Feng Liu
This study introduces a data‑driven method for multi‑objective optimization in fused filament fabrication (FFF), targeting both geometric accuracy and filament bond quality. Experiments supply part‑quality data, which feed Bayesian neural network models that predict the two objectives while accounting for epistemic and aleatory uncertainties. Using these stochastic predictions, robustness‑based optimization explores nozzle temperature, speed, and layer thickness, producing Pareto surfaces that reveal trade‑offs and are validated through actual part manufacturing.
By Berkcan Kapusuzoglu, Paromita Nath, Matthew Sato, Sankaran Mahadevan, Paul Witherell