Artificial Neural Network (ANN) based microstructural prediction model for 22MnB5 boron steel during tailored hot stamping

Prasun Chokshi, Richard Dashwood, Darren J. Hughes

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Because of demand for lower emissions and better crashworthiness, the use of hot stamped 22MnB5 boron steel has greatly increased in manufacturing of automobile components. However, for many applications it is required that only certain regions in hot stamped parts are fully hardened whereas other regions need be more ductile. The innovative process of tailored hot stamping does this by controlling the localized microstructures through tailored cooling rates by dividing the tooling into heated and cooled zones. A barrier to optimal application of this technique is the lack of reliable phase distribution prediction model for the process. We present a novel Artificial Neural Network (ANN) based phase distribution prediction model for tailored hot stamping. The model was developed and validated using data generated from extensive thermo-mechanical physical simulation experiments and instrumented nanoindentation based phase quantification method. Advanced statistical techniques were used for preventing overfitting, for making the optimal use of available experimental data and for quantification of prediction uncertainty. The final predictions made by the ANN model during its independent validation have shown good agreement with the experimentally generated data and have a RMS prediction error of just 7.7%, which is a significant improvement over the existing models.

Original languageEnglish
Pages (from-to)162-172
Number of pages11
JournalComputers and Structures
Volume190
Early online date15 Jun 2017
DOIs
Publication statusPublished - 1 Oct 2017

Fingerprint

Boron
Stamping
Steel
Prediction Model
Artificial Neural Network
Neural networks
Quantification
Crashworthiness
RMS Errors
Physical Simulation
Nanoindentation
Prediction
Overfitting
Automobile
Prediction Error
Neural Network Model
Simulation Experiment
Cooling
Microstructure
Manufacturing

Keywords

  • 22MnB5 boron steel
  • Artificial Neural Network
  • Microstructure
  • Modelling
  • Nanoindentation
  • Tailored hot stamping

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Modelling and Simulation
  • Materials Science(all)
  • Mechanical Engineering
  • Computer Science Applications

Cite this

Artificial Neural Network (ANN) based microstructural prediction model for 22MnB5 boron steel during tailored hot stamping. / Chokshi, Prasun; Dashwood, Richard; Hughes, Darren J.

In: Computers and Structures, Vol. 190, 01.10.2017, p. 162-172.

Research output: Contribution to journalArticle

@article{0a20a2d60a9b4ae883a6c8c62fb56ee2,
title = "Artificial Neural Network (ANN) based microstructural prediction model for 22MnB5 boron steel during tailored hot stamping",
abstract = "Because of demand for lower emissions and better crashworthiness, the use of hot stamped 22MnB5 boron steel has greatly increased in manufacturing of automobile components. However, for many applications it is required that only certain regions in hot stamped parts are fully hardened whereas other regions need be more ductile. The innovative process of tailored hot stamping does this by controlling the localized microstructures through tailored cooling rates by dividing the tooling into heated and cooled zones. A barrier to optimal application of this technique is the lack of reliable phase distribution prediction model for the process. We present a novel Artificial Neural Network (ANN) based phase distribution prediction model for tailored hot stamping. The model was developed and validated using data generated from extensive thermo-mechanical physical simulation experiments and instrumented nanoindentation based phase quantification method. Advanced statistical techniques were used for preventing overfitting, for making the optimal use of available experimental data and for quantification of prediction uncertainty. The final predictions made by the ANN model during its independent validation have shown good agreement with the experimentally generated data and have a RMS prediction error of just 7.7{\%}, which is a significant improvement over the existing models.",
keywords = "22MnB5 boron steel, Artificial Neural Network, Microstructure, Modelling, Nanoindentation, Tailored hot stamping",
author = "Prasun Chokshi and Richard Dashwood and Hughes, {Darren J.}",
year = "2017",
month = "10",
day = "1",
doi = "10.1016/j.compstruc.2017.05.015",
language = "English",
volume = "190",
pages = "162--172",
journal = "Computers & Structures",
issn = "0045-7949",
publisher = "Elsevier",

}

TY - JOUR

T1 - Artificial Neural Network (ANN) based microstructural prediction model for 22MnB5 boron steel during tailored hot stamping

AU - Chokshi, Prasun

AU - Dashwood, Richard

AU - Hughes, Darren J.

PY - 2017/10/1

Y1 - 2017/10/1

N2 - Because of demand for lower emissions and better crashworthiness, the use of hot stamped 22MnB5 boron steel has greatly increased in manufacturing of automobile components. However, for many applications it is required that only certain regions in hot stamped parts are fully hardened whereas other regions need be more ductile. The innovative process of tailored hot stamping does this by controlling the localized microstructures through tailored cooling rates by dividing the tooling into heated and cooled zones. A barrier to optimal application of this technique is the lack of reliable phase distribution prediction model for the process. We present a novel Artificial Neural Network (ANN) based phase distribution prediction model for tailored hot stamping. The model was developed and validated using data generated from extensive thermo-mechanical physical simulation experiments and instrumented nanoindentation based phase quantification method. Advanced statistical techniques were used for preventing overfitting, for making the optimal use of available experimental data and for quantification of prediction uncertainty. The final predictions made by the ANN model during its independent validation have shown good agreement with the experimentally generated data and have a RMS prediction error of just 7.7%, which is a significant improvement over the existing models.

AB - Because of demand for lower emissions and better crashworthiness, the use of hot stamped 22MnB5 boron steel has greatly increased in manufacturing of automobile components. However, for many applications it is required that only certain regions in hot stamped parts are fully hardened whereas other regions need be more ductile. The innovative process of tailored hot stamping does this by controlling the localized microstructures through tailored cooling rates by dividing the tooling into heated and cooled zones. A barrier to optimal application of this technique is the lack of reliable phase distribution prediction model for the process. We present a novel Artificial Neural Network (ANN) based phase distribution prediction model for tailored hot stamping. The model was developed and validated using data generated from extensive thermo-mechanical physical simulation experiments and instrumented nanoindentation based phase quantification method. Advanced statistical techniques were used for preventing overfitting, for making the optimal use of available experimental data and for quantification of prediction uncertainty. The final predictions made by the ANN model during its independent validation have shown good agreement with the experimentally generated data and have a RMS prediction error of just 7.7%, which is a significant improvement over the existing models.

KW - 22MnB5 boron steel

KW - Artificial Neural Network

KW - Microstructure

KW - Modelling

KW - Nanoindentation

KW - Tailored hot stamping

U2 - 10.1016/j.compstruc.2017.05.015

DO - 10.1016/j.compstruc.2017.05.015

M3 - Article

VL - 190

SP - 162

EP - 172

JO - Computers & Structures

JF - Computers & Structures

SN - 0045-7949

ER -