Page 54 - IJAMD-1-2

International Journal of AI for
Materials and Design

ORIGINAL RESEARCH ARTICLE
Data-driven prediction of strain fields in auxetic

structures and non-contact validation with
mechanoluminescence for structural health

monitoring

Junheui Jo † , Minwoo Park † , Sukheon Kang † , Hugon Lee † ,
Chang-Yeon Gu , Taek-Soo Kim , and Seunghwa Ryu*

Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology,
Yuseong-gu, Daejeon, Republic of Korea

Abstract

Recent advancements in 3D printing technology have significantly enhanced
the potential of auxetic structures, which are notable for their negative Poisson’s
ratio, particularly in applications such as sensor technology and structural health
monitoring. Central to the performance of these structures is the accurate estimation
of the effective strain parameter, a critical metric for assessing structural integrity.
However, as structural complexity increases, estimating this parameter becomes
increasingly challenging. The fabrication and real-world validation of these structures
† These authors contributed equally are equally important challenges. This paper introduces two key innovations for
to this work. the practical application of auxetic structures. First, we present a multi-kernel
*Corresponding author: hierarchical deep neural network that leverages finite element simulation data
Seunghwa Ryu to accurately predict effective strain fields in complex auxetic configurations. This
(ryush@kaist.ac.kr) model architecture not only reduces the number of parameters requiring training
Citation: Jo J, Park M, Kang S, but also enhances feature learning and generalization capabilities, achieving over
et al. Data-driven prediction of 90% accuracy in predicting strain fields. Second, we validate these predictions using
strain fields in auxetic structures
and non-contact validation with a 3D-printed specimen embedded with mechanoluminescent (ML) particles. This
mechanoluminescence for structural approach enables direct, non-contact visualization of strain in real-time, offering
health monitoring. Int J AI Mater high spatial and temporal resolution. The alignment observed between predicted
Design. 2024;1(2):3539.
doi:10.36922/ijamd.3539 and observed strain concentration areas demonstrates the efficacy of integrating ML
technology into auxetic designs. This integration significantly improves the reliability
Received: April 30, 2024
and diagnostic capabilities of advanced structural systems.
Accepted: June 18, 2024
Published Online: July 30, 2024 Keywords: Mechanical metamaterials; Auxetic structure; Structural health monitoring;
Copyright: © 2024 Author(s). 3D printing; Mechanoluminescence; Deep learning
This is an Open-Access article
distributed under the terms of the
Creative Commons Attribution
License, permitting distribution,
and reproduction in any medium, 1. Introduction
provided the original work is
properly cited. Mechanical auxetic structures have gained significant attention in engineering applications,
Publisher’s Note: AccScience driven by advancements in additive manufacturing technologies that enable the production
Publishing remains neutral with of complex shapes. These structures exhibit a unique ability to expand laterally under tensile
1-4
regard to jurisdictional claims in
published maps and institutional stress, characterized by a negative Poisson’s ratio (NPR), which confers superior mechanical
affiliations. properties. Notably, auxetic structures are recognized for their enhanced shear resistance,
5,6
Volume 1 Issue 2 (2024) 48 doi: 10.36922/ijamd.3539