Modelling of the Steel High-Temperature Deformation Behaviour Using Artificial Neural Network

Round 1

Reviewer 1 Report

The manuscript entitled "Modelling of the Steel High-Temperature Deformation Behavior Using Artificial Neural Network" is a piece of interesting research in the modelling of material hot deformation behaviors.  The paper is recommended to be accepted after minor corrections. Please see the comments as follows:

1) Please give more details about the algorithm, and discuss advantages and characteristics those make it suitable for modelling hot deformation of metals.

2) The modelling fitting is not that good, the difference 20MPa roughly is quite common for Gleeble high temperature testing. How many tests have been performed for each condition? How is the repeatability? Would be better if the error can be discussed.

3) How the test conditions are chosen ? For figure 5, a comparison of the experimental and predicted hot deformation curves at different deformation conditions, the results show that the model error is large at temperatures lower than 980℃, could the authors make the fitting better by optimizing the structural form of the equations of the constitutive model?

Author Response

Dear Reviewer!

Thank you for your thorough consideration of our paper “Modelling of the Steel High-Temperature Deformation Behavior Using Artificial Neural Network”, your kind response and valuable comments. The manuscript was modified accordingly to your advice.

Please, find attached file with the detailed answers.

Author Response File: Author Response.pdf

Reviewer 2 Report

1. Authors investigated the construction and approbation of the ANN-based model relating rheological properties with the chemical composition and hot deformation parameters of high-alloyed corrosion-resistant steels. But which software was used for development of artificial neural networks? Is this author's own software or commercial software?
2. Authors constructed the artificial neural network model for predicting flow stress of the high-alloyed corrosion-resistant steel during the hot deformation. The model can be used to determine the possible influence of chemical composition variations of the steel on the stress value. But it is not clear how another readers will be able to use this model? If you have formula it is simple to share with other readers. But if you have ANN-based model it is not simple to share with other readers and researchers.
3. The structure of the constructed ANN had 16 neurons for input layer, 30 and 15 neurons for hidden layers, 1 neuron for output layer. The dataset size was 7912. The dataset size depends on the number of layers, neurons and the number of connections between them. Authors should provide a formula or rule showing the relationship between the minimum required dataset size and structure of the ANN.
4. The results presented in the article have the practical value. Usage of the ANN-based model lets optimization of the industrial hot deformation technology depending on the real chemical composition of the steel. But the authors should also to formulate the scientific novelty.

Author Response

Dear Reviewer!

Thank you for your thorough consideration of our paper “Modelling of the Steel High-Temperature Deformation Behavior Using Artificial Neural Network” your kind response and valuable comments. The manuscript was modified accordingly to your advice.

Please, find attached file with our detailed answers.

Author Response File: Author Response.pdf

Reviewer 3 Report

The main idea behind the paper is somewhat interesting: presenting an artificial neural network used for predicting with high accuracy the steel flow stress of the Cr12Ni3 Cu steel in hot deformation. The advantage of the ANN model consists in obtaining a high accuracy for predicting models of the flow stress of the high-alloyed corrosion-resistant steel during the hot deformation.

The title and the intentions declared in the abstract correspond to the contents of the paper. Some of the references could be more related to the subject of the paper.

The authors have serious contributions in the last 5 years in the field of the hot deformation:

1. Churyumov, A.Y.; Medvedeva, S.V.; Mamzurina, O.I.; Kazakova, A.A.; Churyumova, T.A. United approach to modelling of the hot deformation behavior, fracture, and microstructure evolution of austenitic stainless AISI 316Ti steel. Applied Sciences (Switzerland) 2021, 11, doi:10.3390/app11073204.
2. Shaikh, A.; Churyumov, A.; Pozdniakov, A.; Churyumova, T. Simulation of the hot deformation and fracture behavior of reduced activation ferritic/martensitic 13CrMoNbV Steel. Applied Sciences (Switzerland) 2020, 10, doi:10.3390/app10020530.
3. Renault, C.; Churyumov, A.Y.; Pozdniakov, A.V.; Churyumova, T.A. Microstructure and hot deformation behavior of FeMnAlCMo steel. J. Mater. Res. Technol. 2020, 9, 4440-4449, doi:10.1016/j.jmrt.2020.02.069.
4. Churyumov, A.Y.; Pozdnyakov, A.V.; Churyumova, T.A.; Cheverikin, V.V. Hot plastic deformation of heat-resistant austenitic aisi 310s steel. Part 1. simulation of flow stress and dynamic recrystallization. Chernye Metally 2020, 2020, 48-55.
5. Churyumov, A.Y.; Pozdnyakov, A.V.; Churyumova, T.A.; Cheverikin, V.V. Hot plastic deformation of heat-resistant austenitic aisi 310s steel. Part 2. tensile torsional fracture simulation. Chernye Metally 2020, 2020, 32-38.
6. Churyumov, A.Y.; Pozdniakov, A.V. Simulation of Microstructure Evolution in Metal Materials under Hot Plastic Deformation and Heat Treatment. Phys. Met. Metall. 2020, 121, 1064-1086, doi:10.1134/S0031918X20110034.
7. Aripov, G.R.; Bazlov, A.I.; Churyumov, A.Y.; Polkin, V.I.; Luzgin, D.V.; Prokoshkin, S.D. Study of the Change in the Structure and Properties of High-Entropic Alloys during Thermal and Thermomechanical Processing. Russ. J. Non-Ferrous Met. 2020, 61, 413-420, doi:10.3103/S1067821220040021.
8. Prosviryakov, A.; Mondoloni, B.; Churyumov, A.; Pozdniakov, A. Microstructure and hot deformation behaviour of a novel Zr-alloyed high-boron steel. Metals 2019, 9, doi:10.3390/met9020218.
9. Churyumov, A.Y.; Pozdniakov, A.V.; Prosviryakov, A.S.; Loginova, I.S.; Daubarayte, D.K.; Ryabov, D.K.; Korolev, V.A.; Solonin, A.N.; Pavlov, M.D.; Valchuk, S.V. Microstructure and mechanical properties of a novel selective laser melted Al-Mg alloy with low Sc content. Mater. Res. Express 2019, 6, doi:10.1088/2053-1591/ab5bea.
10. Churyumov, A.Y.; Pozdniakov, A.V.; Mondoloni, B.; Prosviryakov, A.S. Effect of boron concentration on hot deformation behavior of stainless steel. Results Phys. 2019, 13, doi:10.1016/j.rinp.2019.102340.
11. Churyumov, A.Y.; Pozdniakov, A.V.; Bazlov, A.I.; Mao, H.; Polkin, V.I.; Louzguine-Luzgin, D.V. Effect of Nb Addition on Microstructure and Thermal and Mechanical Properties of Fe-Co-Ni-Cu-Cr Multiprincipal-Element (High-Entropy) Alloys in As-Cast and Heat-Treated State. JOM 2019, 71, 3481-3489, doi:10.1007/s11837-019-03644-z.
12. Churyumov, A.Y. Deformation and Fracture of 13CrMoNbV Ferritic-Martensitic Steel at Elevated Temperature. Phys. Met. Metall. 2019, 120, 1228-1232, doi:10.1134/S0031918X19120032.
13. Churyumov, A.Y.; Spasenko, V.V.; Hazhina, D.M.; Mikhaylovskaya, A.V.; Solonin, A.N.; Prosviryakov, A.S. Study of the Structural Evolution of a Two-Phase Titanium Alloy during Thermodeformation Treatment. Russ. J. Non-Ferrous Met. 2018, 59, 637-642, doi:10.3103/S1067821218060032.
14. Churyumov, A.Y.; Mikhaylovskaya, A.V.; Bazlov, A.I.; Tsarkov, A.A.; Kotov, A.D.; Aksenov, S.A. Influence of Al3Ni crystallisation origin particles on hot deformation behaviour of aluminium based alloys. Philos. Mag. 2017, 97, 572-590, doi:10.1080/14786435.2016.1273557.

Remarks:

1. The paper contains an abstract and introduction which should be in fact a critical review of the state of the art. Please elaborate more the Introduction chapter.
2. Many references are only related to ANN analysis. Please reduce the number of these references and only keep a few essential ones.
3. In the Table 2 the authors mentioned that the temperature range is 700 – 1250⁰ C and in the abstract and the text of the paper they mentioned a range of 900 – 1200 (which I consider correct). The same remarks for the strain rate. Does the table only refer to the training dataset? It is a little bit confusing.
4. Did the authors use an open access dataset to build the ANN-based model? Or did they collect the results from scientific papers? It is hard to believe that they introduced all the 7912 results only from scientific papers.
5. In the Materials and Methods chapter, please present some images with the experimental layout used for the hot compression tests and some images with the specimens.
6. The text presented between lines 149 to 164 (the equation of constitutive modelling using the Zener-Hollomon parameter) was also presented in the paper Renault, C.; Churyumov, A.Y.; Pozdniakov, A.V.; Churyumova, T.A. Microstructure and hot deformation behavior of FeMnAlCMo steel. J. Mater. Res. Technol. 2020, 9, 4440-4449, doi:10.1016/j.jmrt.2020.02.069. Please eliminate or reformulate those paragraphs.
7. I consider that the paper needs a Discussion chapter where the results of this study must be compared with the results obtained by other researchers. There are some discussions in the results chapter (you could split this chapter and extend it).
8. Limitations of the study must be pointed out in the Discussion section. The main problem of the study is that the number of the experimental tests is not presented. How many times has each experiment been replicated? As it is, the study is not statistically relevant.
9. The Conclusion number 4 “The approbation of the constructed model shows that the hot deformation behavior may be significantly influenced by the fluctuation of the chemical composition within the grade of the steel” is obvious. Of course, the changes in the chemical composition influence the mechanical behaviour of the steel, not only on hot forming!

I will provide the opportunity to the authors to extensively modify their manuscript.

Author Response

Thank you for your thorough consideration of our paper “Modelling of the Steel High-Temperature Deformation Behavior Using Artificial Neural Network” your kind response and valuable comments. The manuscript was modified accordingly to your advice.

Please, find attached file with our detailed answers.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report