Microstructure-Conditioned Surrogate Models for Graded Multiscale Optimization of Mycelium Composites

DOI10.48550/arXiv.2607.13688MaRDI QIDQ7284132arXivFDO

Authors J. Storm, F. P. van der Meer, I. B. C. M. Rocha, S. Schyck, K. Masania

Publication date 15 July 2026

Published in arXiv

Tldr We train a hypernetwork that conditions a hybrid physics-data surrogate on microstructural variables to accelerate graded multiscale simulations of mycelium-woodchip composites. We validate our approach against full FE^2 simulations, demonstrating a 42% reduction in peak stress when optimizing a graded structure.

Full work available at URL https://doi.org/10.48550/arXiv.2607.13688

arXiv classification

cond-mat.mtrl-scics.LGmath.NAphysics.comp-ph

This page was built for publication: Microstructure-Conditioned Surrogate Models for Graded Multiscale Optimization of Mycelium Composites

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284132)