Microstructure optimization with constrained design objectives using machine learning-based feedback-aware data-generation

Arindam Paul*, Pinar Acar, Wei-Keng Liao, Alok Nidhi Choudhary, Veera Sundararaghavan, Ankit Agrawal

*Corresponding author for this work

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Microstructure sensitive design has a critical impact on the performance of engineering materials. The safety and performance requirements of critical components, as well as the cost of material and machining of Titanium components, make dovetailing of the microstructure imperative. This paper addresses the optimization of several microstructure design problems for Titanium components under specific design constraints using a feedback-aware data-driven solution methodology. In this study, the microstructure is modeled with an orientation distribution function (ODF) that measures the volumes of different crystallographic orientations. Two algorithms are used to sample the entire microstructure space followed by machine learning-aided identification of a minimal subset of ODF dimensions which is subsequently explored by targeted sampling. Conventional optimization methods lead to a unique microstructure rather than yielding a comprehensive space of optimal or near-optimal microstructures. Multiple solutions are crucial for the deployment of materials design for manufacturing as traditional manufacturing processes can only generate a limited set of microstructures. Our data sampling-based methodology not only outperforms or is on par with other optimization techniques in terms of the optimal property value, but also provides numerous near-optimal solutions, 3–4 orders of magnitude more than previous methods. Consequently, the proposed framework delivers a spectrum of optimal solutions in the microstructure space which can accelerate materials development and reduce manufacturing costs.

Original languageEnglish (US)
Pages (from-to)334-351
Number of pages18
JournalComputational Materials Science
Volume160
DOIs
StatePublished - Apr 1 2019

Fingerprint

machine learning
Learning systems
Microstructure
Machine Learning
Feedback
microstructure
optimization
Optimization
manufacturing
Titanium
Distribution functions
Distribution Function
titanium
Optimal Solution
Manufacturing
Design for Manufacturing
distribution functions
methodology
Sampling
data sampling

ASJC Scopus subject areas

  • Computer Science(all)
  • Chemistry(all)
  • Materials Science(all)
  • Mechanics of Materials
  • Physics and Astronomy(all)
  • Computational Mathematics

Cite this

@article{bf1a545c128d43db8ba21c209ab4807e,
title = "Microstructure optimization with constrained design objectives using machine learning-based feedback-aware data-generation",
abstract = "Microstructure sensitive design has a critical impact on the performance of engineering materials. The safety and performance requirements of critical components, as well as the cost of material and machining of Titanium components, make dovetailing of the microstructure imperative. This paper addresses the optimization of several microstructure design problems for Titanium components under specific design constraints using a feedback-aware data-driven solution methodology. In this study, the microstructure is modeled with an orientation distribution function (ODF) that measures the volumes of different crystallographic orientations. Two algorithms are used to sample the entire microstructure space followed by machine learning-aided identification of a minimal subset of ODF dimensions which is subsequently explored by targeted sampling. Conventional optimization methods lead to a unique microstructure rather than yielding a comprehensive space of optimal or near-optimal microstructures. Multiple solutions are crucial for the deployment of materials design for manufacturing as traditional manufacturing processes can only generate a limited set of microstructures. Our data sampling-based methodology not only outperforms or is on par with other optimization techniques in terms of the optimal property value, but also provides numerous near-optimal solutions, 3–4 orders of magnitude more than previous methods. Consequently, the proposed framework delivers a spectrum of optimal solutions in the microstructure space which can accelerate materials development and reduce manufacturing costs.",
author = "Arindam Paul and Pinar Acar and Wei-Keng Liao and Choudhary, {Alok Nidhi} and Veera Sundararaghavan and Ankit Agrawal",
year = "2019",
month = "4",
day = "1",
doi = "10.1016/j.commatsci.2019.01.015",
language = "English (US)",
volume = "160",
pages = "334--351",
journal = "Computational Materials Science",
issn = "0927-0256",
publisher = "Elsevier",

}

Microstructure optimization with constrained design objectives using machine learning-based feedback-aware data-generation. / Paul, Arindam; Acar, Pinar; Liao, Wei-Keng; Choudhary, Alok Nidhi; Sundararaghavan, Veera; Agrawal, Ankit.

In: Computational Materials Science, Vol. 160, 01.04.2019, p. 334-351.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Microstructure optimization with constrained design objectives using machine learning-based feedback-aware data-generation

AU - Paul, Arindam

AU - Acar, Pinar

AU - Liao, Wei-Keng

AU - Choudhary, Alok Nidhi

AU - Sundararaghavan, Veera

AU - Agrawal, Ankit

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Microstructure sensitive design has a critical impact on the performance of engineering materials. The safety and performance requirements of critical components, as well as the cost of material and machining of Titanium components, make dovetailing of the microstructure imperative. This paper addresses the optimization of several microstructure design problems for Titanium components under specific design constraints using a feedback-aware data-driven solution methodology. In this study, the microstructure is modeled with an orientation distribution function (ODF) that measures the volumes of different crystallographic orientations. Two algorithms are used to sample the entire microstructure space followed by machine learning-aided identification of a minimal subset of ODF dimensions which is subsequently explored by targeted sampling. Conventional optimization methods lead to a unique microstructure rather than yielding a comprehensive space of optimal or near-optimal microstructures. Multiple solutions are crucial for the deployment of materials design for manufacturing as traditional manufacturing processes can only generate a limited set of microstructures. Our data sampling-based methodology not only outperforms or is on par with other optimization techniques in terms of the optimal property value, but also provides numerous near-optimal solutions, 3–4 orders of magnitude more than previous methods. Consequently, the proposed framework delivers a spectrum of optimal solutions in the microstructure space which can accelerate materials development and reduce manufacturing costs.

AB - Microstructure sensitive design has a critical impact on the performance of engineering materials. The safety and performance requirements of critical components, as well as the cost of material and machining of Titanium components, make dovetailing of the microstructure imperative. This paper addresses the optimization of several microstructure design problems for Titanium components under specific design constraints using a feedback-aware data-driven solution methodology. In this study, the microstructure is modeled with an orientation distribution function (ODF) that measures the volumes of different crystallographic orientations. Two algorithms are used to sample the entire microstructure space followed by machine learning-aided identification of a minimal subset of ODF dimensions which is subsequently explored by targeted sampling. Conventional optimization methods lead to a unique microstructure rather than yielding a comprehensive space of optimal or near-optimal microstructures. Multiple solutions are crucial for the deployment of materials design for manufacturing as traditional manufacturing processes can only generate a limited set of microstructures. Our data sampling-based methodology not only outperforms or is on par with other optimization techniques in terms of the optimal property value, but also provides numerous near-optimal solutions, 3–4 orders of magnitude more than previous methods. Consequently, the proposed framework delivers a spectrum of optimal solutions in the microstructure space which can accelerate materials development and reduce manufacturing costs.

UR - http://www.scopus.com/inward/record.url?scp=85060334705&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060334705&partnerID=8YFLogxK

U2 - 10.1016/j.commatsci.2019.01.015

DO - 10.1016/j.commatsci.2019.01.015

M3 - Article

VL - 160

SP - 334

EP - 351

JO - Computational Materials Science

JF - Computational Materials Science

SN - 0927-0256

ER -