Stochastic full-range multiscale modeling of thermal conductivity of Polymeric carbon nanotubes composites: A machine learning approach

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

  • Bokai Liu
  • Nam Vu-Bac
  • Xiaolong Fu
  • Xiaoying Zhuang
  • Timon Rabczuk

Externe Organisationen

  • Bauhaus-Universität Weimar
  • Xi'an Modern Chemistry Research Institute
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Aufsatznummer115393
Seitenumfang1
FachzeitschriftComposite structures
Jahrgang289
Frühes Online-Datum12 März 2022
PublikationsstatusVeröffentlicht - 1 Juni 2022

Abstract

Based on a stochastic full-range multiscale model, we propose a data-driven approach to predict the thermal conductivity of CNT reinforced polymeric nano-composites (PNCs). Uncertain input parameters at different scales are propagated from nano- to macro-scale within a bottom-up multi-scale framework. Atomistic models are employed at the nano-scale while continuum mechanics approaches are used at the micro-, meso- and macro-scale. Representative volume elements in the context of finite element modeling (RVE-FEM) are used to finally obtain the homogenized thermal conductivity. To connect the micro and mesoscale and simplify the computation, we take advantage of the equivalent fiber theory. The input parameters are selected by a top-down scanning method and subsequently are converted as uncertain inputs. The length of single-walled carbon nanotube (SWCNT), the chirality of SWCNT, the thermal conductivity of the fibers, the thermal conductivity of the matrix, the Kapitza resistance, aspect ratio, agglomeration index, dispersion index and volume fraction are assumed as random-parameters. The Regression-tree-based (Random Forest and Gradient Boosting Machine) and Neural networks-based (Artificial neural networks and Deep neural networks) approaches are exploited for computational efficiency, where Particle Swarm Optimization (PSO) and 10-fold Cross Validation (CV) are employed for hyper-parameter tuning. Our machine learning prediction results agree well with published experimental data, which can provide a versatile and efficient method to design new PNCs.

ASJC Scopus Sachgebiete

Zitieren

Stochastic full-range multiscale modeling of thermal conductivity of Polymeric carbon nanotubes composites: A machine learning approach. / Liu, Bokai; Vu-Bac, Nam; Fu, Xiaolong et al.
in: Composite structures, Jahrgang 289, 115393, 01.06.2022.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Liu B, Vu-Bac N, Fu X, Zhuang X, Rabczuk T. Stochastic full-range multiscale modeling of thermal conductivity of Polymeric carbon nanotubes composites: A machine learning approach. Composite structures. 2022 Jun 1;289:115393. Epub 2022 Mär 12. doi: 10.1016/j.compstruct.2022.115393
Download
@article{ea4351c4f447450087d3900d4eaa1d45,
title = "Stochastic full-range multiscale modeling of thermal conductivity of Polymeric carbon nanotubes composites: A machine learning approach",
abstract = "Based on a stochastic full-range multiscale model, we propose a data-driven approach to predict the thermal conductivity of CNT reinforced polymeric nano-composites (PNCs). Uncertain input parameters at different scales are propagated from nano- to macro-scale within a bottom-up multi-scale framework. Atomistic models are employed at the nano-scale while continuum mechanics approaches are used at the micro-, meso- and macro-scale. Representative volume elements in the context of finite element modeling (RVE-FEM) are used to finally obtain the homogenized thermal conductivity. To connect the micro and mesoscale and simplify the computation, we take advantage of the equivalent fiber theory. The input parameters are selected by a top-down scanning method and subsequently are converted as uncertain inputs. The length of single-walled carbon nanotube (SWCNT), the chirality of SWCNT, the thermal conductivity of the fibers, the thermal conductivity of the matrix, the Kapitza resistance, aspect ratio, agglomeration index, dispersion index and volume fraction are assumed as random-parameters. The Regression-tree-based (Random Forest and Gradient Boosting Machine) and Neural networks-based (Artificial neural networks and Deep neural networks) approaches are exploited for computational efficiency, where Particle Swarm Optimization (PSO) and 10-fold Cross Validation (CV) are employed for hyper-parameter tuning. Our machine learning prediction results agree well with published experimental data, which can provide a versatile and efficient method to design new PNCs.",
keywords = "Data-driven modeling (DDM), Machine Learning, Polymeric Nanotube composites (PNCs), Stochastic multi-scale modeling, Thermal properties",
author = "Bokai Liu and Nam Vu-Bac and Xiaolong Fu and Xiaoying Zhuang and Timon Rabczuk",
note = "Funding Information: We gratefully acknowledge the support of the China Scholarship Council (CSC) .",
year = "2022",
month = jun,
day = "1",
doi = "10.1016/j.compstruct.2022.115393",
language = "English",
volume = "289",
journal = "Composite structures",
issn = "0263-8223",
publisher = "Elsevier BV",

}

Download

TY - JOUR

T1 - Stochastic full-range multiscale modeling of thermal conductivity of Polymeric carbon nanotubes composites

T2 - A machine learning approach

AU - Liu, Bokai

AU - Vu-Bac, Nam

AU - Fu, Xiaolong

AU - Zhuang, Xiaoying

AU - Rabczuk, Timon

N1 - Funding Information: We gratefully acknowledge the support of the China Scholarship Council (CSC) .

PY - 2022/6/1

Y1 - 2022/6/1

N2 - Based on a stochastic full-range multiscale model, we propose a data-driven approach to predict the thermal conductivity of CNT reinforced polymeric nano-composites (PNCs). Uncertain input parameters at different scales are propagated from nano- to macro-scale within a bottom-up multi-scale framework. Atomistic models are employed at the nano-scale while continuum mechanics approaches are used at the micro-, meso- and macro-scale. Representative volume elements in the context of finite element modeling (RVE-FEM) are used to finally obtain the homogenized thermal conductivity. To connect the micro and mesoscale and simplify the computation, we take advantage of the equivalent fiber theory. The input parameters are selected by a top-down scanning method and subsequently are converted as uncertain inputs. The length of single-walled carbon nanotube (SWCNT), the chirality of SWCNT, the thermal conductivity of the fibers, the thermal conductivity of the matrix, the Kapitza resistance, aspect ratio, agglomeration index, dispersion index and volume fraction are assumed as random-parameters. The Regression-tree-based (Random Forest and Gradient Boosting Machine) and Neural networks-based (Artificial neural networks and Deep neural networks) approaches are exploited for computational efficiency, where Particle Swarm Optimization (PSO) and 10-fold Cross Validation (CV) are employed for hyper-parameter tuning. Our machine learning prediction results agree well with published experimental data, which can provide a versatile and efficient method to design new PNCs.

AB - Based on a stochastic full-range multiscale model, we propose a data-driven approach to predict the thermal conductivity of CNT reinforced polymeric nano-composites (PNCs). Uncertain input parameters at different scales are propagated from nano- to macro-scale within a bottom-up multi-scale framework. Atomistic models are employed at the nano-scale while continuum mechanics approaches are used at the micro-, meso- and macro-scale. Representative volume elements in the context of finite element modeling (RVE-FEM) are used to finally obtain the homogenized thermal conductivity. To connect the micro and mesoscale and simplify the computation, we take advantage of the equivalent fiber theory. The input parameters are selected by a top-down scanning method and subsequently are converted as uncertain inputs. The length of single-walled carbon nanotube (SWCNT), the chirality of SWCNT, the thermal conductivity of the fibers, the thermal conductivity of the matrix, the Kapitza resistance, aspect ratio, agglomeration index, dispersion index and volume fraction are assumed as random-parameters. The Regression-tree-based (Random Forest and Gradient Boosting Machine) and Neural networks-based (Artificial neural networks and Deep neural networks) approaches are exploited for computational efficiency, where Particle Swarm Optimization (PSO) and 10-fold Cross Validation (CV) are employed for hyper-parameter tuning. Our machine learning prediction results agree well with published experimental data, which can provide a versatile and efficient method to design new PNCs.

KW - Data-driven modeling (DDM)

KW - Machine Learning

KW - Polymeric Nanotube composites (PNCs)

KW - Stochastic multi-scale modeling

KW - Thermal properties

UR - http://www.scopus.com/inward/record.url?scp=85126582297&partnerID=8YFLogxK

U2 - 10.1016/j.compstruct.2022.115393

DO - 10.1016/j.compstruct.2022.115393

M3 - Article

VL - 289

JO - Composite structures

JF - Composite structures

SN - 0263-8223

M1 - 115393

ER -