Development of Three-Dimensional LES Based Meshless Model of Continuous Casting of Steel

Round 1

Reviewer 1 Report

The paper entitled “Development of three-dimensional LES based meshless model of continuous casting of steel” is having interesting results, novel one, and is having publishable information. The paper is also written well and properly structured. The developed large eddy simulation (LES) model is applied to continuous casting product which is showing the real-time applications. I am recommending this article with minor revision. I have observed the following points.

1. When I read the abstract, first few lines are related to general like introduction. It is suggested to incorporated their contributions in the abstract.

2. Further, the developed model is applied to what kind of product during continuous casting process. Whether the model is applicable to specific metals/alloys during continuous casting process(CCP) or it can be applied to all kind of metals/alloys during CCP. This is missing in the abstract. It has to be incorporated.

3. In introduction part, some more literature related to the use of LES, RANS, and other models in cast products are to be incorporated. In present form, it is lagging.

4. Nomenclature is to be used for various symbols used in their work. For example, k – ε, Ω etc… are not explained.

5. Whether the developed LES model considering steady state condition only  or can it be considered unsteady state also? Need explanation

6. Whether the developed model is validated by any experimental data or not? Need clarification

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work extends to 3D, a local collocation meshless method based on radial basis functions to solve the problem of continuous casting of steel billets already reported in Reference [4]. The manuscript is well written, and the mathematical methods are well presented. However, the authors must address some issues before the manuscript's acceptance. 

 

The predictions of the LES and RANS models are presented using three vertical planes (VA1, VA2, VA3), one horizontal plane (HA1), two vertical lines (VL1, VL2), and two horizontal lines (HL1, HL2). As the authors report, no significant differences exist between the behavior observed in the planes VA2 and VA3 and the lines HL1 and HL2. On the other hand, Figures 11 and 12 show that the RANS model predicts four recirculation zones on each mold corner. These two figures also show that the LES model predicts the existence of recirculation zones on some of the mold corners. The location of the recirculation zones changes as time evolves, and as expected, these zones disappear when the transient numerical results are averaged. Based on these facts, the authors should consider replacing one of the vertical planes, VA2 or VA3, with a plane joining two opposite corners of the mold. Similarly, one of the horizontal lines, HL1 or HL2, should be replaced with a line joining two opposite corners of the mold. 

 

The following general comments must be addressed:

- In Figure 6, the temperature range should be modified to notice the differences between the predictions of the two models.

- Change the comma for a semicolon in line 16.  

- Correct the spelling of Reynolds in line 9.

- For uniformity, change the notation of "low Re" in lines 9 and 46 to "Low-Re." 

- In line 161, the number of the figure was omitted.

- The spelling of "modelling" is a non-American variant. Consider replacing it with "modeling."

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The 3D LES solution is compared with the 3D turbulent models in 3D geometry of continuous casting of steel. This paper demonstrates the ability of this method to solve real industrial 3D cases. Simple adjustment of node density, high accuracy and low numerical diffusivity are the main advantages of this meshless method.

 

1.     The reference of LES method on the numerical simulation of the mold should be supplemented in introduction.

 

2.     Line 220 Heat transfer coefficient in the mould hmo 2000 W/(m2 K).

The Heat transfer coefficient value in the vertical direction of mould is not very suitable. The heat transfer in mold is usually given the heat flux. It is suggested to describe it in the form of piecewise function, so that the heat flow of the crystallizer is more practical, or what is the difference between the heat flow you use and the actual heat flow?

 

3.     The effectiveness or evaluation of the numerical model should be explained in a separate part.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The article deals with the current issue of simulation and modeling of continuous casting billets. The authors use popular methods based on FEM.

I have several questions and comments about the presented research:

In line 66 - there is a repeated word in description

Please add some description about discretization - what software do you use for computation?

The usage of software and HW configuration for numerical implementation is lightly unsuitable (due to computational time). Please add some justification of the used SW and HW configuration. For achieving more accurate and faster outputs, there are non-desktop solutions suitable.

You stated that presented method has a ability to solve realistic industrial problems in the future. Were some comparational measurements or analyses of model outputs with real continuos castin of billet carried out?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors correctly addressed all the comments and suggestions in the revised version of the manuscript.