Embedded One-Dimensional Orifice Elements for Slosh Load Calculations in Volume-Of-Fluid CFD

Round 1

Reviewer 1 Report

Based on previous study for constant cross-sectional orifices, this work presents a method of replacing the fine computational mesh elements within orifices with large one-dimensional mesh elements with the intended advantage of reducing the overall computational cost of CFD simulations. This allows one to simply modify the mesh in orifice areas by drastically reducing the amount of mesh elements in these regions. Moreover, the 1D model was further extended to account for two-phase loss effects by applying existing semi-empirical correlations for flow losses. To ensure sharp tracking of the liquid-gas interface within the 1D orifice elements, careful attention to the VOF advection should be paid. The contents of the paper are in line with the topic of the journal. The document is well written and structured. Before being accepted, however, the following improvements are still needed. 

 

Questions:

1.      For the violent flow of sloshing, the flow pattern in the long orifice is very complex. It may be difficult to capture the hydraulic characteristics of the violent flow in the long office by using one-dimensional model instead of three-dimensional one. This method may be more meaningful when applied to the numerical simulation of water conveyance structure with relatively stable flow pattern.

 

2.      When using one-dimensional model in long office, how to consider the loss of water head due to wall friction? This needs further explanation and illustration.

 

3.      As we know, for liquid sloshing, turbulence plays an import role. Therefore, a proper turbulence model should be used in three-dimensional CFD simulations. Now the liquid within the long office is simplified to be one dimensional. How was the head loss caused by turbulence considered? This needs further explanation and illustration.

 

4.      In the simulation of sloshing process, how to consider the influence of air mass entering the long office on its hydraulic characteristics (e.g., flow velocity, pressure, etc. in the pipe) and head loss? Please give detailed explanation and illustration.

 

5.      In the 3D simulation of tank sloshing, to meet the accuracy requirement, the number of vertical grid layers in the long office should be sufficiently large. This is the main reason why a small grid size has to be used for the whole domain, which brings a heavy computational load. If the grid size in the cross-sectional area of the long orifice remains unchanged, and longer grid size is arranged along the length direction of the office, whether significant accuracy degeneration will be caused? If not, can this strategy be used?

6.      There are many inconsistencies in the references. Please check all of them.

Author Response

Please see our response in the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this work, an integrated numerical/empirical/experimental approach is presented for slosh modelling and investigation. Authors have presented that Implementing of the 1D orifice element can result with significant reduction of overall CPU time while preserving accuracy of more demanding computational models.

The introduction provides sufficient background and includes relevant references. The research design is appropriate and well structured. Investigation approach and used methods are adequately described. The conclusions are well supported by the clearly presented results.

Authors are kindly advised to edit text in order to maintain text consistency (e.g. beginning of the chapter 2.3)

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear authors,

I have found your paper very interesting and it gives a sound proof that the methodology proposed is solid. I would like you to address very minor comments and I have some questions for you.

Line 54: repetition of the word "work". suggestion: In this work, we further the methodology of Ozhan [...].

The correlations proposed are valid for smooth surfaces. Could you address or comment how correlations are affected by roughness. Has this an impact and can be this addressed separately on a dedicated CFD study? Are there any limitations of the correlations proposed based on the properties of the fluids?

Figure 7: for sake of simplicity I would explain first the middle plot and then the bottom plot... in order of appeareance from top to down. Instead in the caption you switch them explaining first the bottom and then the middle plot.

I could also suggest that the paper could have been shortened if the editor require so, combining figure 10 and figure 15 and figure 11 and 16... The information are redundant.

Could you argument better after the 25% of the simulations what is deviating? You show a contour comparison of the void fraction in figure 22 at the 20% of the period. Could you maybe show as well similar contour for the part of the period in which you consider the simulations not reliable?
Are the full resolution and the mesh C giving similar results?

It is not clear to me which Mesh is better at the end. Could you enhance your conclusions on this?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf