Neural parametric representations for thin-shell shape optimisation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Xiao, Xiao, Cirak, Fehmi
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913013739552768
author Xiao, Xiao
Cirak, Fehmi
author_facet Xiao, Xiao
Cirak, Fehmi
contents Shape optimisation of thin-shell structures requires a flexible, differentiable geometric representation suitable for gradient-based optimisation. We propose a neural parametric representation (NRep) for the shell mid-surface based on a neural network with periodic activation functions. The NRep is defined using a multi-layer perceptron (MLP), which maps the parametric coordinates of mid-surface vertices to their physical coordinates. A structural compliance optimisation problem is posed to optimise the shape of a thin-shell parameterised by the NRep subject to a volume constraint, with the network parameters as design variables. The resulting shape optimisation problem is solved using a gradient-based optimisation algorithm. Benchmark examples with classical solutions demonstrate the effectiveness of the proposed NRep. The approach exhibits potential for complex lattice-skin structures, owing to the compact and expressive geometry representation afforded by the NRep.
format Preprint
id arxiv_https___arxiv_org_abs_2604_06612
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Neural parametric representations for thin-shell shape optimisation
Xiao, Xiao
Cirak, Fehmi
Numerical Analysis
Machine Learning
Shape optimisation of thin-shell structures requires a flexible, differentiable geometric representation suitable for gradient-based optimisation. We propose a neural parametric representation (NRep) for the shell mid-surface based on a neural network with periodic activation functions. The NRep is defined using a multi-layer perceptron (MLP), which maps the parametric coordinates of mid-surface vertices to their physical coordinates. A structural compliance optimisation problem is posed to optimise the shape of a thin-shell parameterised by the NRep subject to a volume constraint, with the network parameters as design variables. The resulting shape optimisation problem is solved using a gradient-based optimisation algorithm. Benchmark examples with classical solutions demonstrate the effectiveness of the proposed NRep. The approach exhibits potential for complex lattice-skin structures, owing to the compact and expressive geometry representation afforded by the NRep.
title Neural parametric representations for thin-shell shape optimisation
topic Numerical Analysis
Machine Learning
url https://arxiv.org/abs/2604.06612