Optimization of Nozzle Parameters by Investigating the Flow Behavior of Molten Steel in the Mold under a High Casting Speed
Round 1
Reviewer 1 Report
The following key points are to be addressed by authors,
1. It is important to highlight in the abstract, why the nozzle design parameters require optimization and what happens, if there are variations with respect to the flow behaviour.
2. How do you ensure immersion depth has significant influence than impact depth.
3. In the abstract, the authors has not clearly mentioned the method employed to optimize the nozzle design parameters.
4. The specific continuity equation and N-S equation might used the lot of assumptions and how do you ensure to meet those in practical experiments.
5. The mold material may also influence the casting and flow behaviour and it is essential to report the mold material used for your research work.
6. The authors reported the variable ranges in the Table 1 and how those variables are selected for simulation need to be clearly explained.
7. Authors can also test with different melt or pouring temperature, which has direct influence on flow behaviour as a result of change in viscosity, density and so on.
8. Table 3 show the differences is above 10% when tested the simulation results with experimental values. Can you report why such differences are observed with explanation.
9. Does the authors find any gas inclusions during the simulations when the speed is high.
Author Response
Reviewer #1
Comment 1. It is important to highlight in the abstract, why the nozzle design parameters require optimization and what happens, if there are variations with respect to the flow behaviour.
Response: Thanks for the reviewer’s comments. We have added the following contents in the abstract to highlight why the nozzle design parameters require optimization and what happens, if there are variations with respect to the flow behaviour, that is, “A reasonable flow field in the continuous casting mold is beneficial to product high quality billets, and the design of nozzle parameters of mold is the key to regulating the flow behavior of molten steel.”, which is marked in Blue.
Comment 2. How do you ensure immersion depth has significant influence than impact depth.
Response: Thanks for your comments. We investigated the influence of the insertion depth and the inner diameter of the SEN on the impact depth of the molten steel in the mold under different continuous speed. The results show that the impact depth varies greatly in value at varying insertion depth. With the increase of insertion depth, the impact depth of molten steel increases rapidly. In contrast, the influence of nozzle size on the impact depth of molten steel is not obvious. With the increase of the inner diameter of the nozzle, the fluctuation of the impact depth is smooth, and there is no obvious rule. Therefore, based on the above research results, we believe that the immersion depth of the SEN has a greater impact on the flow behavior of molten steel than the inner diameter of the nozzle.
In order to express our findings more clearly, we have revised the relevant content in the manuscript, i. e., “It can be obtained that compared with insertion depth of SEN, the influence of nozzle size on the impact depth of molten steel is not obvious. With the increase of the inner diameter of the nozzle, the fluctuation of the impact depth is smooth, and there is no obvious rule. Under the same casting speed, the difference of impact depth is in the range of 5-30 mm. Therefore, based on the above research results, we believe that the immersion depth of the SEN has a greater influence on the flow behavior of molten steel than the inner diameter of the nozzle”. Please check it on Page 8, Line 223-229, which is marked in Blue.
Comment 3. In the abstract, the authors has not clearly mentioned the method employed to optimize the nozzle design parameters.
Response: Thanks for the review notification. We have added the following in the abstract so clearly mentioned the method employed to optimize the nozzle design parameters, that is “Through combining the numerical simulation and physical experiments, taking SEN immersion depth and inner diameter as indicators, the flow behavior of molten steel in the mold during high speed casting of 160 mm×160 mm billet was investigated in detail, and the nozzle parameters were optimized.”, which is marked in Blue.
Comment 4. The specific continuity equation and N-S equation might used the lot of assumptions and how do you ensure to meet those in practical experiments.
Response: Thanks for your proposal. Due to the limitations of numerical simulation methods, we have to adopt some assumptions to simplify the solution process on the premise of ensuring the accuracy as much as possible. We supplement some basic assumptions in the text, that is “Moreover, we have made appropriate assumptions in the calculation process. â‘ The molten steel is regarded as an incompressible fluid; â‘¡ The influence of a small amount of chemical reactions existing in the mold is ignored, and the molten steel is regarded as a homogeneous medium; â‘¢ The free liquid level of the crystallizer is horizontally stable â‘£ The vibration of the mold is not considered”. Please check it on Page 4, Line 147-151, which is marked in Blue.
In practical experiments, due to the similarity theory, we can directly simulate the flow behavior of molten steel in the mold, and can clearly observe the flow state of the fluid, so it is not necessary to satisfy the lot of assumptions as in the numerical simulation. In this study, we use physical experiments to verify the accuracy and rationality of some numerical simulation results.
Comment 5. The mold material may also influence the casting and flow behaviour and it is essential to report the mold material used for your research work.
Response: Thanks for your suggestion. The mold material did also influence the casting and flow behaviour to some extent. For numerical simulation, we used wall boundary conditions to achieve a closed effect rather than establish a real mold.
Additionally, the ownership of the water molder used in the physical experiments is our laboratory, in which the porous thin shell made of Polymethyl Methacrylate (PMMA) is placed in the inner cavity of the physical hydraulic model of the mold.
We added relevant contents to Page 11, Line 300 and marked it in Blue.
Comment 6. The authors reported the variable ranges in the Table 1 and how those variables are selected for simulation need to be clearly explained.
Response: Thanks for the reviewer’s comments. In the simulation process, we control the casting speed by adjusting the amount of water injected into the mold per unit time. The immersion depth is determined by adjusting the distance that the nozzle is inserted into the mold. In addition, the nozzle size is adjusted by establishing different mathematical models. We added relevant content to Page 4, Line 143-147 and marked it in Blue.
Comment 7. Authors can also test with different melt or pouring temperature, which has direct influence on flow behaviour as a result of change in viscosity, density and so on.
Response: Thanks to the reviewer for your professional suggestions. As you mentioned, the change in viscosity and density will have significant influence on flow behaviour. We will refer to your opinion in our follow-up work, and further investigate the influence of different parameters such as melt and pouring temperature on the melt flow behavior.
Comment 8. Table 3 show the differences is above 10% when tested the simulation results with experimental values. Can you report why such differences are observed with explanation.
Response: Thanks for your comments. As mentioned in comment 4. Due to the limitations of numerical simulation methods, we have to adopt some assumptions to simplify the solution process on the premise of ensuring the accuracy as much as possible. This will also cause the numerical simulation results to deviate from reality. Although there are differences of about 10% between the numerical simulation and the physical experimental results, the overall trend of the two remains consistent, and the error is within an acceptable range. Therefore, on the one hand, it verifies the reliability of our numerical simulation research results, and on the other hand, it also shows that the results can provide theoretical guidance for the optimization of the nozzle structure in actual production to a certain extent.
In addition, we have supplemented the manuscript contents, that is, “Due to the limitations of numerical simulation methods, we have to adopt some assumptions to simplify the solution process on the premise of ensuring the accuracy as much as possible. This will also cause the numerical simulation results to deviate from reality.” Please check it on Page 11, Line 309-312, which is marked in Blue.
Comment 9. Does the authors find any gas inclusions during the simulations when the speed is high.
Response: Thanks for the careful review comment. We did find a little gas inclusion at the beginning of physics experiments. However, it is caused by insufficient sealing of the equipment after inspection. We then sealed the device with plasticine. Therefore, no gas inclusions were observed when re-experimenting. All physical experiments in the manuscript were performed after the device's tightness was improved.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Dear Authors,
The text and form of the presentation were easy to understand. The only mistake I noticed was the caption of Table 3 on lines 299 and 312. There are only two tables in the manuscript.
Author Response
Reviewer #2
Comments.
Dear Authors,
The text and form of the presentation were easy to understand. The only mistake I noticed was the caption of Table 3 on lines 299 and 312. There are only two tables in the manuscript.
Response: Thanks for the reviewer’s comments. We have modified the caption of “Table 3” to “Table 2”, and marked it in Purple.
Author Response File: Author Response.pdf
Reviewer 3 Report
Dear authors, your manuscript should be significantly improved, please find attached Review.
Comments for author File: Comments.pdf
Author Response
Reviewer #3
Comment 1. To start with main concerns regarding the manuscript:
Why do you drop in the presented study to lower casting speeds in comparison to your experimental work in Reference 24? That would be great to compare one-to-one the simulation and experimental predictions to see some possible deviations and explain their origins. Also, that would be great testing of the CFD approach. Please clearly explain your choice and motivation.
Response: Thanks for the reviewer’s comments.
For continuous casting production, high casting speed can significantly improve production efficiency, which is an important development direction of continuous casting production technology. Ultra-high casting speed casting is to further increase the casting speed on the basis of high casting speed, and is now mostly used in the production of thin slabs. Our previous study (Reference 24) mainly used physical experiments to initially focus on the flow behavior of molten steel in the mold at ultra-high drawing speeds. For billets, high-casting speed casting is widely used at present. Although ultra-high casting speed can further improve production efficiency, the current technology is not mature enough, and most are in the research stage of the laboratory. Therefore, the research on high casting speed has more practical significance for the current development of continuous casting technology.
Therefore, this study focuses on the production process of billets with high drawing speed (3.5 -4.5 m/min), and optimizes the SEN parameters by analyzing the flow behavior of molten steel in the mold in detail, which is not only important to realize the reasonable control of the flow in the mold, but also lays a foundation for related research under ultra-high casting speed. Of course, the flow, heat transfer and solidification behavior of molten steel in the mold under ultra-high casting speed will also be one of the key research directions in the future.
Comments 2. Literature review
Check citations style for the Metals journal. It is different than yours. Reference section has some formatting issues. Please correct. Using freeware Zotero (as recommended in the Authors Guide) helps to avoid any reference issues.
Response: Thanks for your comments.
We have modified the formatting of the references in the manuscript according to the citation style of Metals, please check it in “References”.
Reference 11 and 15 from years 1972 and 1973 look a bit outdated. Some other references from 1990’s also have newer and more detailed analogs. At the same time, modern key studies are not presented, please check a description and the reference list below:
A key review papers which emphasize the importance of the presented topic are missing in the introduction. [1–3]
The influence of the melt impact on the growth of the mushy zone was studied early in very details by Wu et al. [4]
Response: Thanks for the review notification.
We have carefully read the newer literature as you mentioned and cited it into the manuscript based on its content to improve the integrity of the manuscript. Additionally, references 11 and 15 are indeed quite old, we have removed these two references and cited newer research results to describe important modern research results. Please check it in “Introduction”, which is marked in Red.
Which SEN materials are used in this study? Is there a protective lining applied against clogging and thermal loss? As it was shown earlier by Barati et al. [5] and later by Vakhrushev et al. [6] the SEN clogging enhanced by the heat flux through the refractory causes the parasitic solidification inside the SEN tube and leads to the reduced superheat of the meld fed into the CC mold. These aspects should be included and discussed since you deal with optimization of the SEN.
Response: Thanks for your proposal.
SEN is hollowed out in this study and not reflected in the model, so there is no material involving in SEN. Besides, SEN takes the WALL boundary and is set to adiabatic and also does not consider a protective lining applied against clogging and thermal loss. Since this study focuses on the influence of SEN parameters (mainly SEN immersion depth and nozzle size) on the flow behavior of molten steel in the mold, and optimizes the SEN parameters on this basis, the material of SEN is not considered too much. In the follow-up research, we will optimize the model and further consider the influence of different SEN materials on the behavior of molten steel flow and heat transfer.
It is also highly recommended to consider important results from very recent publication on experimental and numerical results on the SENs performance to control the meniscus level by Morales and co-authors. [7]
The classical publications on slag entrapment with corresponding criteria should be mentioned. [8–10]
The excellent time resolved PIV measurements were published by Puttinger and Saeedipour [11].
Advanced numerical techniques on your topic were presented by Liu et al. [12,13].
Formation of the asymmetric flow and consequences on the shell formation were detailed in a very recent study [14]
Response: Thanks for the reviewer’s comments.
The literature you mentioned was extremely important in enriching the manuscript. We have properly cited relevant literature, please check it in “Introduction” and “Model description”, which is marked in Red.
Comments 3. Required manuscript corrections
1) English should be significantly improved; authors often use nonstandard structure of the sentences; some ideas become cumbersome and are hard to follow.
Response: Thanks for the review notification.
We have invited a professional English worker to modify the English expressions in the manuscript. Revised portions are marked in Red in the revised manuscript.
2) Line 29: melt is typically fed rather then injected into a mold.
3) Please avoid tautology, many term repeated in the same sentences.
4) Line 31: flow field of the mold change to flow field in the mold.
5) Line 35: involved is a wrong term; did you mean entrapped? What do you mean by
subcutaneous? Once again, make a proofreading of your manuscript.
6) Line 81: which SEN parameters apart of the inner diameter and immerse depth were studied? How would, for example, the ports angle change your results?
Response: Thanks for your professional recommendation.
Sorry for these mistakes. The above spelling and grammar problems have been corrected one by one in the manuscript, please check it in: 2) Page 1, Line 31. 4) Page 1, Line 33, Line 35, Page 2, Line 80, Line 86. 5) Page 1, Line 36, which is marked in Red.
In addition, in our current study, we only considered the influence of the insertion depth and the size of the nozzle on the flow behavior of molten steel, and did not consider other nozzle parameters. However, as you mentioned, parameters such as ports angle also have a critical influence on the flow behavior of molten steel, which is also the focus of our follow-up research.
7) Critical comment: you employ steady state approach with rather dissipative k-epsilon model. Thereby you exclude transient turbulent flow and meniscus waving phenomena from your study. You must strongly motivate your choice, since nowadays transient models with complex sub-grid models are used especially for high casting speeds, see modern studies mentioned in Section II of this review.
Response: Thanks for your suggestion.
I'm sorry for such an error. Due to our negligence, we ignore time effects when writing the governing equations. However, we considered the transient flow behavior of molten steel throughout the calculation, and we set the time step to 0.01 and the total step size to 4000. In addition, in the model description section we supplemented the calculation details, as Page 4, Line 133-134 which is marked in Red. Fig. 1 is a schematic diagram of some calculation parameter settings.
Furthermore, we have corrected the relevant governing equations as Eq. (1)–Eq. (6), please check it on Page3.
Fig 1. Schematic diagram of parameter settings related to transient calculation
8) References 22 and 23 do not relate to the k-epsilon model. Please explain why you refer to them in the modelling section.
Response: Thanks for the reviewer’s comments.
Since these two papers also use numerical simulation to solve some problems in metallurgical production, and discuss the flow characteristics of molten steel in the text, we cite them. But at present, the research content is irrelevant to ours, it is inappropriate. We have therefore removed these two references from the manuscript.
9) It seems that the equations formatting does not corresponds to the Metals style. Please correct!
Response: Thanks for your comments.
We have revised the relevant formulas and made them more in line with Metals style, please check it in Eq. (1)-Eq. (9).
10) In the RHS of the equation (2) there is a buoyancy term. I did not find density variation in your paper. Did you consider buoyancy and how?
Response: Thanks for the careful review comment.
In our study, the density term is constant and does not change during the calculation, so the effect of change of buoyance is not considered.
11) Line 103: the turbulent kinetic energy is defined as k, but on Line 105 another symbol K is used. In equation (4) it is again redefined as K. Please use consistent symbolling in the manuscript.
12) Line 111: u_i, u_j and x_i, x_j are the components, not vectors.
Response: Thanks for your careful comments.
Sorry for these mistakes. The above spelling error has been corrected in the manuscript, please check it on Page 3-4.
13) Line 121, critical comment: you speak about modeling “heat transfer behavior”, however your model (1)-(6) is isothermal as well as your calculations. None of temperature field and shell thickness are presented. That must be explained or corrected!
14) In your boundary conditions your mention temperature again, see previous comment.
Response: Thanks to the reviewer for your professional suggestions.
Since the change of temperature affects the flow behavior of molten steel to a certain extent, we considered the heat transfer between the mold and the molten steel in the research process. However, this study focuses on clarifying the influence of different nozzle parameters on the flow behavior of molten steel in the mold, and does not focus on the changing laws of information such as the temperature field. The variation law of parameters such as relevant temperature and shell thickness was focused on in another study of ours. Therefore, in order to highlight the research focus, we did not involve the temperature field of molten steel and other information in the manuscript, and omitted the relevant heat transfer equations. However, to further improve the completeness and readability of the manuscript, we have supplemented the relevant heat transfer equations and modified some expressions. Please check it on Page 4, Line. 118-28, which is marked in Red.
15) Line 134: model is discretized using finite volume method using hexahedral elements, not meshes. Please check your terminology throughout the manuscript!
16) Lines 159-161: it is not clear what the authors meant by inner and outer recirculation arcs.
17) Line 163: SEN not SEM.
18) Line 168: immersion depth used two times next to each other.
19) Figure 2 and later in the manuscript: what is moir topography? Do you mean contours of the velocity field? Also, the field definition is missing at the color bar, only units are visible. Correct it in the entire manuscript!
Response: Thanks for the careful review comment
The errors in the above expression has been corrected in the manuscript, please check it on: 15) Page 5, Line152. 16) Page 9, Line 255-259. 17) Page 6, Line 177. 18) Page 6, Line 181-184.
19) All expression issues have been revised. Also, all images have field definitions already added, please check it in Fig. 2, 4, 8, 10, 11.
20) General question: why do you get asymmetric flow for example in Figure 10 when you are using a steady state solver and symmetry plane in your domain? Are your results not converged enough or what is an· · explanation?
Response: Thanks to the reviewer for your professional suggestions. Continuous casting includes open pouring, normal pouring and stop pouring process, among which the molten steel is unsteady during the open pouring and stop pouring process, only in a steady state during normal pouring, and the model we established is the steady state. In addition, in the numerical simulation, we established a 1/2 model along the longitudinal section of the mold, and our model is an arc mold rather than straight. Therefore, when the molten steel is injected into the mold from the SEN, the flow environment will be significant different, which will inevitably lead to asymmetric distribution of the molten steel on the surface of the mold.
We added some content in the model description section. Please check it on Page 4, Line 131, which is marked in Red.
21) Figure 12, critical comment: it is not possible to use presented experimental results for the simulation verification by two reasons. Firstly, different view angle is used and now velocity field is presented in the experimental pictures. Secondly, according to your boundary conditions, top of your simulation domain represents a free-slip wall and no waving of the meniscus is modeled. Additionally, due to restriction of the vertical velocity component, you will overpredict submeniscus velocities in the simulation. Figure 12 is misleading and should be removed!
Response: Thanks for the reviewer’s comments.
For the reasons you stated, the two sets of pictures really don't compare well. We want to observe the fluctuation of molten steel on the liquid level of the mold by physical experiments, and at the same time observe whether the liquid slag has defects such as bare and entrained slag, so as to confirm that our optimized nozzle parameters can be well adapted to high-speed production. Therefore, according to your suggestion, we deleted the calculation results of the numerical simulation, leaving the slag film distribution results of the physical simulation. At the same time, we have also made appropriate revisions to the statements in the manuscript, please check it on Page 12, Line 316-326, which is marked in Red.
Author Response File: Author Response.pdf
Reviewer 4 Report
In this paper, authors combining the numerical simulation and physical experiments.
The flow behavior of molten steel in the mold during high speed casting of 160 mm×160 mm billet was investigated in detail, and the nozzle parameters were optimized.
The results obtenaid demonstrate that compared with the inner diameter of nozzle, the immersion depth has a significant influence on the impact depth of molten steel.
The results of physical experiment and numerical simulation reveal that the optimized nozzle parameters can well adapt to the continuous casting process with high casting speed.
Results are clearly presented.
Conclusions are punctual but must be more clearly.
English language and style are fine but are necessary minor spell check.
References are good and appropriate.
Comments for author File: Comments.pdf
Author Response
Reviewer #4
Comments.
In this paper, authors combining the numerical simulation and physical experiments.
The flow behavior of molten steel in the mold during high speed casting of 160 mm×160 mm billet was investigated in detail, and the nozzle parameters were optimized.
The results obtenaid demonstrate that compared with the inner diameter of nozzle, the immersion depth has a significant influence on the impact depth of molten steel.
The results of physical experiment and numerical simulation reveal that the optimized nozzle parameters can well adapt to the continuous casting process with high casting speed.
Results are clearly presented.
Conclusions are punctual but must be more clearly.
English language and style are fine but are necessary minor spell check.
References are good and appropriate.
Response: Thanks for the reviewer’s comments. We have checked and partially revised the conclusions and ensured that they are more clearly, which is marked in Green.
In addition, we have invited a professional English worker to modify the English expressions in the manuscript.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Congratulations for your excellent scientific contribution
Reviewer 3 Report
Dear authors, your manuscript is significantly improved and can be now recommended for publication in Metals. However you have to correct some mistypes throughout the manuscript, for example:
1) Lines 154-162: your list numbering looks nonstandard with numbers in circles. Use standard one.
2) Reference 13: you have strange [C] symbols.
So please check for mistypes and formatting.
