Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials
Metamaterials have received extensive attention in fundamental and applied research over the past two decades due to their unique mechanical behavior. This paper presents an interpretable machine learning (ML) approach for efficient response prediction of three-dimensional (3D)-printed metamaterials...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-07-01
|
Series: | Journal of Materials Research and Technology |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2238785423012851 |
_version_ | 1797745252279255040 |
---|---|
author | Bolaji Oladipo Helio Matos N.M. Anoop Krishnan Sumanta Das |
author_facet | Bolaji Oladipo Helio Matos N.M. Anoop Krishnan Sumanta Das |
author_sort | Bolaji Oladipo |
collection | DOAJ |
description | Metamaterials have received extensive attention in fundamental and applied research over the past two decades due to their unique mechanical behavior. This paper presents an interpretable machine learning (ML) approach for efficient response prediction of three-dimensional (3D)-printed metamaterials. However, developing such an ML-based model requires a large consistent, representative, balanced, and complete dataset. To this extent, an experimentally validated finite element analysis (FEA) approach is implemented to generate 8096 non-self-intersecting re-entrant honeycomb structures by varying the mesoscale geometrical features to obtain the corresponding Poisson's ratios. This dataset is leveraged to develop a feed-forward multilayer perceptron-based predictive model. The developed ML model shows excellent predictive efficacy on the unseen test dataset. Shapely additive explanation (SHAP) is then used for model interpretation. SHAP results show that the slant cell length is the dominant input feature dictating the model output whereas cell angle and vertical cell length show mixed trends signifying that other input features influence their effect on the model output. Moreover, cell thickness does not significantly influence the model output when compared to other input features. Overall, the integrated numerical simulation-experiment-interpretable ML-based predictive approach presented here can be leveraged to design and develop metamaterials for a wide range of engineering applications. |
first_indexed | 2024-03-12T15:20:44Z |
format | Article |
id | doaj.art-55c206a052824e33862bd8d66d148b0d |
institution | Directory Open Access Journal |
issn | 2238-7854 |
language | English |
last_indexed | 2024-03-12T15:20:44Z |
publishDate | 2023-07-01 |
publisher | Elsevier |
record_format | Article |
series | Journal of Materials Research and Technology |
spelling | doaj.art-55c206a052824e33862bd8d66d148b0d2023-08-11T05:33:19ZengElsevierJournal of Materials Research and Technology2238-78542023-07-012516121625Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterialsBolaji Oladipo0Helio Matos1N.M. Anoop Krishnan2Sumanta Das3Department of Civil and Environmental Engineering, University of Rhode Island, Kingston, RI, 02881, United StatesDepartment of Mechanical, Industrial and Systems Engineering, University of Rhode Island, Kingston, RI, 02881, United StatesDepartment of Civil Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi, 110016, India; Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi, Hauz Khas, New Delhi, 110016, IndiaDepartment of Civil and Environmental Engineering, University of Rhode Island, Kingston, RI, 02881, United States; Department of Mechanical, Industrial and Systems Engineering, University of Rhode Island, Kingston, RI, 02881, United States; Corresponding author.Metamaterials have received extensive attention in fundamental and applied research over the past two decades due to their unique mechanical behavior. This paper presents an interpretable machine learning (ML) approach for efficient response prediction of three-dimensional (3D)-printed metamaterials. However, developing such an ML-based model requires a large consistent, representative, balanced, and complete dataset. To this extent, an experimentally validated finite element analysis (FEA) approach is implemented to generate 8096 non-self-intersecting re-entrant honeycomb structures by varying the mesoscale geometrical features to obtain the corresponding Poisson's ratios. This dataset is leveraged to develop a feed-forward multilayer perceptron-based predictive model. The developed ML model shows excellent predictive efficacy on the unseen test dataset. Shapely additive explanation (SHAP) is then used for model interpretation. SHAP results show that the slant cell length is the dominant input feature dictating the model output whereas cell angle and vertical cell length show mixed trends signifying that other input features influence their effect on the model output. Moreover, cell thickness does not significantly influence the model output when compared to other input features. Overall, the integrated numerical simulation-experiment-interpretable ML-based predictive approach presented here can be leveraged to design and develop metamaterials for a wide range of engineering applications.http://www.sciencedirect.com/science/article/pii/S2238785423012851Machine learningNeural networkRe-entrantAuxetic behaviorSHAPFinite element analysis |
spellingShingle | Bolaji Oladipo Helio Matos N.M. Anoop Krishnan Sumanta Das Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials Journal of Materials Research and Technology Machine learning Neural network Re-entrant Auxetic behavior SHAP Finite element analysis |
title | Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials |
title_full | Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials |
title_fullStr | Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials |
title_full_unstemmed | Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials |
title_short | Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D printed re-entrant metamaterials |
title_sort | integrating experiments finite element analysis and interpretable machine learning to evaluate the auxetic response of 3d printed re entrant metamaterials |
topic | Machine learning Neural network Re-entrant Auxetic behavior SHAP Finite element analysis |
url | http://www.sciencedirect.com/science/article/pii/S2238785423012851 |
work_keys_str_mv | AT bolajioladipo integratingexperimentsfiniteelementanalysisandinterpretablemachinelearningtoevaluatetheauxeticresponseof3dprintedreentrantmetamaterials AT heliomatos integratingexperimentsfiniteelementanalysisandinterpretablemachinelearningtoevaluatetheauxeticresponseof3dprintedreentrantmetamaterials AT nmanoopkrishnan integratingexperimentsfiniteelementanalysisandinterpretablemachinelearningtoevaluatetheauxeticresponseof3dprintedreentrantmetamaterials AT sumantadas integratingexperimentsfiniteelementanalysisandinterpretablemachinelearningtoevaluatetheauxeticresponseof3dprintedreentrantmetamaterials |