Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling

This study addresses the usage of data from industrial plate mills to calculate the mean flow stress of different steel grades. Accurate flow stress values may optimize rolling technology, but the existing literature often provides coefficients like those in the Hensel–Spittel equation for a limited...

Full description

Saved in:
Bibliographic Details
Main Authors: Alexey G. Zinyagin, Alexander V. Muntin, Vadim S. Tynchenko, Pavel I. Zhikharev, Nikita R. Borisenko, Ivan Malashin
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Metals
Subjects:
Online Access:https://www.mdpi.com/2075-4701/14/12/1329
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846103675427094528
author Alexey G. Zinyagin
Alexander V. Muntin
Vadim S. Tynchenko
Pavel I. Zhikharev
Nikita R. Borisenko
Ivan Malashin
author_facet Alexey G. Zinyagin
Alexander V. Muntin
Vadim S. Tynchenko
Pavel I. Zhikharev
Nikita R. Borisenko
Ivan Malashin
author_sort Alexey G. Zinyagin
collection DOAJ
description This study addresses the usage of data from industrial plate mills to calculate the mean flow stress of different steel grades. Accurate flow stress values may optimize rolling technology, but the existing literature often provides coefficients like those in the Hensel–Spittel equation for a limited number of steel grades, whereas in modern production, the chemical composition may vary by thickness, customer requirements, and economic factors, making it necessary to conduct costly and labor-intensive laboratory studies. This research demonstrates that leveraging data from industrial rolling mills and employing machine learning (ML) methods can predict material rheological behavior without extensive laboratory research. Two modeling approaches are employed: Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) architectures. The model comprising one GRU layer and two fully connected layers, each containing 32 neurons, yields the best performance, achieving a Root Mean Squared Error (RMSE) of 7.5 MPa for the predicted flow stress of three steel grades in the validation set.
format Article
id doaj-art-cd3a3db4fdfb4e78aa8a779791a72f07
institution Kabale University
issn 2075-4701
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Metals
spelling doaj-art-cd3a3db4fdfb4e78aa8a779791a72f072024-12-27T14:39:45ZengMDPI AGMetals2075-47012024-11-011412132910.3390/met14121329Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial RollingAlexey G. Zinyagin0Alexander V. Muntin1Vadim S. Tynchenko2Pavel I. Zhikharev3Nikita R. Borisenko4Ivan Malashin5Artificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaArtificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaArtificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaArtificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaArtificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaArtificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, RussiaThis study addresses the usage of data from industrial plate mills to calculate the mean flow stress of different steel grades. Accurate flow stress values may optimize rolling technology, but the existing literature often provides coefficients like those in the Hensel–Spittel equation for a limited number of steel grades, whereas in modern production, the chemical composition may vary by thickness, customer requirements, and economic factors, making it necessary to conduct costly and labor-intensive laboratory studies. This research demonstrates that leveraging data from industrial rolling mills and employing machine learning (ML) methods can predict material rheological behavior without extensive laboratory research. Two modeling approaches are employed: Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) architectures. The model comprising one GRU layer and two fully connected layers, each containing 32 neurons, yields the best performance, achieving a Root Mean Squared Error (RMSE) of 7.5 MPa for the predicted flow stress of three steel grades in the validation set.https://www.mdpi.com/2075-4701/14/12/1329flow stressmachine learningrolling millGRULSTMmaterial rheology
spellingShingle Alexey G. Zinyagin
Alexander V. Muntin
Vadim S. Tynchenko
Pavel I. Zhikharev
Nikita R. Borisenko
Ivan Malashin
Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
Metals
flow stress
machine learning
rolling mill
GRU
LSTM
material rheology
title Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
title_full Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
title_fullStr Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
title_full_unstemmed Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
title_short Recurrent Neural Network (RNN)-Based Approach to Predict Mean Flow Stress in Industrial Rolling
title_sort recurrent neural network rnn based approach to predict mean flow stress in industrial rolling
topic flow stress
machine learning
rolling mill
GRU
LSTM
material rheology
url https://www.mdpi.com/2075-4701/14/12/1329
work_keys_str_mv AT alexeygzinyagin recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling
AT alexandervmuntin recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling
AT vadimstynchenko recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling
AT pavelizhikharev recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling
AT nikitarborisenko recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling
AT ivanmalashin recurrentneuralnetworkrnnbasedapproachtopredictmeanflowstressinindustrialrolling