Prediction of nucleation lag time from elemental composition and temperature for iron and steelmaking slags using deep neural networks

Corey Adam Myers, Takao Nakagaki

研究成果: Article

1 引用 (Scopus)

抄録

A prediction of the nucleation lag time of iron and steelmaking melts solely from elemental composition and temperature was produced via deep neural networks trained on data available in the literature. To the best of our knowledge, this constitutes the first published instance of prediction of nucleation lag time that does not require composition specific empirical data. Control of the nucleation process is critical for the production of ground granulated blast furnace slag, control of slag properties for heat recovery or utilization, and the optimization of slag for CO 2 mineralization. The deep neural network achieved an average absolute scaled error (AASE) over a testing set of 947 points covering 7 orders of magnitude of 39.9%. Performance was further improved by bootstrapping with a prediction of liquidus temperature from a separate deep neural network (AASE = 33.4%). Bootstrapping using DNN-generated viscosity data did not increase prediction accuracy. The negligible calculation load of the trained deep neural networks allows for rapid design, analysis, and optimization of novel slag compositions and treatment methods. This ability was demonstrated by calculating the necessary continuous cooling rate to generate amorphous slag across all CaO–Al 2 O 3 –SiO 2 and CaO–FeO–SiO 2 compositions and the potential to use additives to alter said cooling rate.

元の言語English
ページ(範囲)687-696
ページ数10
ジャーナルISIJ International
59
発行部数4
DOI
出版物ステータスPublished - 2019 4 1

Fingerprint

Steelmaking
Slags
Nucleation
Iron
Chemical analysis
Temperature
Cooling
Waste utilization
Waste heat utilization
Carbon Monoxide
Deep neural networks
Viscosity
Testing

ASJC Scopus subject areas

  • Mechanics of Materials
  • Mechanical Engineering
  • Metals and Alloys
  • Materials Chemistry

これを引用

@article{666bfce4d9984b698a2e8ca5b1b35f3f,
title = "Prediction of nucleation lag time from elemental composition and temperature for iron and steelmaking slags using deep neural networks",
abstract = "A prediction of the nucleation lag time of iron and steelmaking melts solely from elemental composition and temperature was produced via deep neural networks trained on data available in the literature. To the best of our knowledge, this constitutes the first published instance of prediction of nucleation lag time that does not require composition specific empirical data. Control of the nucleation process is critical for the production of ground granulated blast furnace slag, control of slag properties for heat recovery or utilization, and the optimization of slag for CO 2 mineralization. The deep neural network achieved an average absolute scaled error (AASE) over a testing set of 947 points covering 7 orders of magnitude of 39.9{\%}. Performance was further improved by bootstrapping with a prediction of liquidus temperature from a separate deep neural network (AASE = 33.4{\%}). Bootstrapping using DNN-generated viscosity data did not increase prediction accuracy. The negligible calculation load of the trained deep neural networks allows for rapid design, analysis, and optimization of novel slag compositions and treatment methods. This ability was demonstrated by calculating the necessary continuous cooling rate to generate amorphous slag across all CaO–Al 2 O 3 –SiO 2 and CaO–FeO–SiO 2 compositions and the potential to use additives to alter said cooling rate.",
keywords = "Basic oxygen furnace, Blast furnace, CCT diagram, CNT, Crystallization, Electric arc furnace, Ggbs, TTT diagram",
author = "Myers, {Corey Adam} and Takao Nakagaki",
year = "2019",
month = "4",
day = "1",
doi = "10.2355/isijinternational.ISIJINT-2018-338",
language = "English",
volume = "59",
pages = "687--696",
journal = "ISIJ International",
issn = "0915-1559",
publisher = "Iron and Steel Institute of Japan",
number = "4",

}

TY - JOUR

T1 - Prediction of nucleation lag time from elemental composition and temperature for iron and steelmaking slags using deep neural networks

AU - Myers, Corey Adam

AU - Nakagaki, Takao

PY - 2019/4/1

Y1 - 2019/4/1

N2 - A prediction of the nucleation lag time of iron and steelmaking melts solely from elemental composition and temperature was produced via deep neural networks trained on data available in the literature. To the best of our knowledge, this constitutes the first published instance of prediction of nucleation lag time that does not require composition specific empirical data. Control of the nucleation process is critical for the production of ground granulated blast furnace slag, control of slag properties for heat recovery or utilization, and the optimization of slag for CO 2 mineralization. The deep neural network achieved an average absolute scaled error (AASE) over a testing set of 947 points covering 7 orders of magnitude of 39.9%. Performance was further improved by bootstrapping with a prediction of liquidus temperature from a separate deep neural network (AASE = 33.4%). Bootstrapping using DNN-generated viscosity data did not increase prediction accuracy. The negligible calculation load of the trained deep neural networks allows for rapid design, analysis, and optimization of novel slag compositions and treatment methods. This ability was demonstrated by calculating the necessary continuous cooling rate to generate amorphous slag across all CaO–Al 2 O 3 –SiO 2 and CaO–FeO–SiO 2 compositions and the potential to use additives to alter said cooling rate.

AB - A prediction of the nucleation lag time of iron and steelmaking melts solely from elemental composition and temperature was produced via deep neural networks trained on data available in the literature. To the best of our knowledge, this constitutes the first published instance of prediction of nucleation lag time that does not require composition specific empirical data. Control of the nucleation process is critical for the production of ground granulated blast furnace slag, control of slag properties for heat recovery or utilization, and the optimization of slag for CO 2 mineralization. The deep neural network achieved an average absolute scaled error (AASE) over a testing set of 947 points covering 7 orders of magnitude of 39.9%. Performance was further improved by bootstrapping with a prediction of liquidus temperature from a separate deep neural network (AASE = 33.4%). Bootstrapping using DNN-generated viscosity data did not increase prediction accuracy. The negligible calculation load of the trained deep neural networks allows for rapid design, analysis, and optimization of novel slag compositions and treatment methods. This ability was demonstrated by calculating the necessary continuous cooling rate to generate amorphous slag across all CaO–Al 2 O 3 –SiO 2 and CaO–FeO–SiO 2 compositions and the potential to use additives to alter said cooling rate.

KW - Basic oxygen furnace

KW - Blast furnace

KW - CCT diagram

KW - CNT

KW - Crystallization

KW - Electric arc furnace

KW - Ggbs

KW - TTT diagram

UR - http://www.scopus.com/inward/record.url?scp=85065800191&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065800191&partnerID=8YFLogxK

U2 - 10.2355/isijinternational.ISIJINT-2018-338

DO - 10.2355/isijinternational.ISIJINT-2018-338

M3 - Article

VL - 59

SP - 687

EP - 696

JO - ISIJ International

JF - ISIJ International

SN - 0915-1559

IS - 4

ER -