Application of Discrete Element Method Coupled with Computational Fluid Dynamics to Predict the Erosive Wear Behavior of Arctic Vessel Hulls Subjected to Ice Impacts
Round 1
Reviewer 1 Report
The paper is worth publication after the following revisions:
1. In Line 208-209 “The flow of the route acts as the dominant load on the drift ice. However, the behavior of the drift ice can slightly affect the flow of the route but has no effect on the overall flow.” I disagree with this point of view that the drift ice has no effect on the overall flow. In my opinion, the effect of two-way couple is limited under some conditions, such as low-velocity and small-size ice floe. But the effect of floe’s existence on the flow is still obvious, e.g. the constraint of the free surface’s movement. I suggest the authors add some necessary descriptions and references to support this point.
2. What’s the meaning of “the flow of the route”? Flow chart or fluid flow? It makes me a little confused.
3. Add the references about the values of parameters selected in Table 1 and 2.
4. How to evaluate the ice concentration containing Type 1, 2 and 3? In other words, for type 1 ice particle, how to calculate the area occupied in the ice field? using the cross section?
5. How to validate the feasibility and accuracy of the present numerical model? I suggest the authors to add some necessary validations.
Author Response
Subject: Response to Review of Manuscript ID [JMSE-257962]
Dear Editor
I hope this letter finds you well. We would like to express our sincere gratitude for the time and effort invested by you and the reviewers in reviewing our manuscript titled "Application of DEM coupled with CFD to predict the erosive wear behavior of Arctic vessel hulls subjected to ice impacts." We highly appreciate the insightful and constructive feedback provided by the reviewers, which has been invaluable in improving the quality and robustness of our research.
We are pleased to provide a comprehensive response to each of the reviewer's comments, addressing their concerns and suggestions for enhancement. Our responses are thoroughly outlined in the attached document, where each comment is followed by our point-by-point response and any revisions made to the manuscript based on the feedback received. We have carefully considered all comments and suggestions from the reviewers and have implemented necessary changes to the manuscript accordingly. We are confident that these revisions have strengthened the clarity, accuracy, and overall scientific contribution of our work.
We would like to express our sincere appreciation for the opportunity to revise and resubmit our manuscript. We are hopeful that the modifications made align well with the expectations of the reviewers and the editorial board.
Thank you once again for your time and consideration. We are looking forward to receiving feedback on the revised version of our manuscript.
Sincerely,
Jang Hyun Lee, Ph.D. /Professor
Department of Naval Architecture and Ocean Engineering, INHA University, KOREA
Enclosure: Response to Reviewer Comments and Revised Manuscript
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper under review focuses on the development of a numerical model that combines the Discrete Element Method (DEM) and Computational Fluid Dynamics (CFD) to simulate the ice-induced wear of the hull of Arctic ships. The authors use one-way coupling of DEM and CFD solvers to model the buoyancy of ice particles impacting the ship hull. Modeling of ship hull wear is performed considering the presence of three different particle shapes and sizes simultaneously. The hull wear was estimated using two methods. The first method does not take into account the deformation of the ship hull, which can be used when estimating the initial wear (e.g. hull coating). The second method takes into account the deformation of the hull material itself and, as a consequence, the change of hydrodynamics in the near-wall region.
This study is undoubtedly of high scientific and practical value, and the research involves the use of modern numerical techniques. Although the developed methodology is well justified, the modeling results themselves raise doubts about the reliability of the approach. In particular, the magnitude of ship wear looks too large for the simulated conditions, namely tens of millimeters of wear per 200 m route. After reading this paper I have the following comments (see below):
1. Introduction
1.1. I feel that, in general, the authors have covered the current state of the problem quite well. However, CFD-DEM coupling for modeling ship navigation in broken ice has already been applied earlier in numerous works, which was noted by the authors. In this regard, the rationale of the objective is not clear to me. The authors say that "The purpose of this study was to develop a numerical model to enable the estimation of the wear of a vessel hull undergoing collision with drift ice based on three key considerations." The following considerations are of course important, but have already been discussed earlier by some authors. The authors should focus the reader's attention on a more specific goal than the development of a numerical model. Maybe “…to evaluate the applicability of the combined use of advanced numerical and mathematical techniques for modeling Arctic ship hull wear?”. Of course, this is up to the authors' discretion.
2. State of the Art
2.1. Following up on the previous comment, I would like to add that although the authors have outlined some differences between their work and other papers, I recommend highlighting them more clearly. For example, this can be done by using a table comparing the main features of numerical methods in the current paper and the papers of preceding authors.
I can recommend the following articles to help gather a more complete picture of the state of research in ice-structure interaction:
· Makarov, O., Bekker, A. & Li, L. Comparative analysis of numerical methods for the modeling of ice–structure interaction problems. Continuum Mech. Thermodyn. 34, 1621–1639 (2022). https://doi.org/10.1007/s00161-022-01152-1
· Li F, Huang L. A Review of Computational Simulation Methods for a Ship Advancing in Broken Ice. Journal of Marine Science and Engineering. 2022; 10(2):165. https://doi.org/10.3390/jmse10020165
3. Numerical scheme
3.1. “In Figure 3, the slope of ?? is the loading stiffness, whereas ?? represents the unloading stiffness”. It is an obvious typo, but the unloading stiffness is the slope of BC, not the BC itself.
3.2. It seems that the authors forgot to mark the point C in Fig. 3.
3.3. The use of a combination of HSL and LSCL contact models is described in detail and understandably. However, the authors should provide a small conclusion of the section on the specific parameter values used in the calculation of loading and unloading stiffness. This is important because DEM is typical in that it is often necessary to apply unreasonable values of standard material parameters for ice modeling. For example, even basic characteristics such as elastic modulus and Poisson's ratio have to be assigned unrealistic values in order to obtain the desired results. This obvious disadvantage of DEM is clearly seen in the works of other authors, where DEM was used for modeling the solid ice fields and I would like to obtain justification of the parameters for the current problem.
3.4. As I understand it, the strength of the methodology that the authors apply to their problem is the use of complex particle shapes, which have a strong influence on the magnitude of ice exposure. The downside is that determining the contact condition and the magnitude of penetration of the rigid particles into each other has to become much more difficult. With cylindrical particles everything is clear and simple, where the intersection can be defined unambiguously through the sphere equation. However, in the case of complex particle shapes, such as those presented in Fig. 4 and Fig. 5, it becomes unclear to me how the depth of penetration is calculated. The authors should focus their attention on this issue rather than on describing the general principles of DEM and CFD, since the basic features of numerical methods can be studied by readers elsewhere. Also based on Fig. 4 and Fig. 5 we can see that the particles themselves have some discretization, i.e., nodes, faces? How does this discretization occur and what recommendations can the authors give for tuning the particle generator of such random shapes to get an adequate modeling of broken ice?
3.5. What were the reasons for choosing the k-ε turbulence model? Were the model parameters assumed to be the same as Ansys Fluent defaults?
3.6. Using the Ganser model to determine the CD friction coefficient looks like a very interesting way. Is this tool available by default when setting up the Rocky-DEM and Fluent coupling or did the authors introduce it through UDFs (User defined functions)?
3.7. A question similar to question 3.5 - what are the reasons for choosing Archard Wear Law to calculate the wear?
4. Evaluation of hull wear due to collision with ice floes
4.1. Eq. 26. Allows to estimate the change in material volume due to wear using accelerated analysis with known material parameters Ca and transition coefficient N. First, there is a typo in the description of the notations (the actual material constant is Ca, not Ce). Secondly, just for the sake of interest, can such a transition really be accurately calculated by a simple linear model? What does the value of the transition coefficient N depend on and how can it be quickly estimated?
4.2. In Table 1, the authors give the values of ice and ship material parameters. As I said earlier (see comment 3.3) the ice characteristics (Young's modulus and Poisson's ratio) look unrealistic and very far from the actual values for ice. Why were the values assumed to be so? What would change in the behavior of the numerical model if the actual values of the parameters (e.g. Young's modulus about 5000 MPa and Poisson's ratio 0.33) were used?
4.3. Did the authors take into account the need to refine the computational mesh near the wall (hull) to properly account for turbulent effects? Despite the RANS method is applied, it still requires the existence of some boundary layers of volume elements. Given that the wall (ship's hull) itself is a moving object, it becomes necessary to adaptively reconstruct the mesh in the near-wall region during the calculation process. This will also affect the dynamics of ice particles at the contact with the ship hull and shear work.
4.4. Figures 19-28. Authors should explicitly mention that “60” and “80” are ice concentrations for readability.
4.5. Table 4. What are the dimensions of wear depth provided in this table? Are they given in mm or m, or μm? In any case, values seem to be very large for the 200 m route even if they are given in μm. Need to check this carefully.
4.6. Fig. 29 and Fig. 30 show the wear magnitude in mm. I understand that the wear magnitude evaluated in this study is not representative of actual operational ships as the authors mentioned earlier. Nevertheless, such large values of wear for only 200 m of the route raise doubts about the correctness of the numerical approach. With the linear extrapolation 8 mm turns into 40 cm for 10 km of track, which is already destroying the hull of any ship at the very beginning of navigation in ice. The authors should check all the results in detail and justify such values. Maybe the problem lies in the insufficient accuracy of the boundary layer?
5. Conclusions
5.1. In the conclusion, the authors should describe in more detail the limitations of the developed model and the direction of future research.
Author Response
Subject: Response to Review of Manuscript ID [JMSE-257962]
Dear Editor
I hope this letter finds you well. We would like to express our sincere gratitude for the time and effort invested by you and the reviewers in reviewing our manuscript titled "Application of DEM coupled with CFD to predict the erosive wear behavior of Arctic vessel hulls subjected to ice impacts." We highly appreciate the insightful and constructive feedback provided by the reviewers, which has been invaluable in improving the quality and robustness of our research.
We are pleased to provide a comprehensive response to each of the reviewer's comments, addressing their concerns and suggestions for enhancement. Our responses are thoroughly outlined in the attached document, where each comment is followed by our point-by-point response and any revisions made to the manuscript based on the feedback received. We have carefully considered all comments and suggestions from the reviewers and have implemented necessary changes to the manuscript accordingly. We are confident that these revisions have strengthened the clarity, accuracy, and overall scientific contribution of our work.
We would like to express our sincere appreciation for the opportunity to revise and resubmit our manuscript. We are hopeful that the modifications made align well with the expectations of the reviewers and the editorial board.
Thank you once again for your time and consideration. We are looking forward to receiving feedback on the revised version of our manuscript.
Sincerely,
Jang Hyun Lee, Ph.D. /Professor
Department of Naval Architecture and Ocean Engineering, INHA University, KOREA
Enclosure: Response to Reviewer Comments and Revised Manuscript
Please see the attachment for the reply and revision.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
I am grateful to the authors for such detailed and kind explanations and answers to my questions. I believe that the article deserves to be published and has important practical and scientific significance. I have two minor comments that may improve the readers' understanding of the article.
1. Introduction
1.1. No more comment
2. State of the Art
2.1. No more comment
3. Numerical scheme
3.1. No more comment
3.2. No more comment
3.3. I understand that the bulk modulus of elasticity is used instead of the linear modulus of elasticity (Young's modulus). When I said "unrealistic", I meant that the values differ by a factor of several thousand or tens of thousands. Specifically, your example uses an ice modulus of 61 MPa and a Poisson's ratio of 0.003, which is about 100 times different from the true values. I realize that you have assumed material parameters like Lee's [48]. But the fact is that all such adjustments of parameter values are not strongly justified. In the end, they consist of simple adjustments of "unrealistic" values to fit the modeling to the experiment.
Although this approach is used by many authors, in my opinion, it is poorly applicable in real practice. For example, how can we be sure that in slightly different conditions such values of ice characteristics that you apply will be valid? Yes, the same can be said for models using real parameter values, because sea ice has very variable properties. But if we use real values, we can at least conduct field or laboratory tests of ice properties in a particular region of the Arctic and refer to them. But in your method we have no way to be sure that we are applying the appropriate values of parameters. For example, have you tried to apply other material parameters to the ice, such as an elastic modulus of 100 MPa (instead of 61 MPa), or a different value of Poisson's ratio? How much would the results change? Can we talk about the reliability of your proposed methodology?
How then to apply this methodology for a particular ship/arctic region? After all, we can't use 61 MPa for the elastic modulus and 0.003 for the Poisson's ratio all the time....
3.4. No more comment
3.5. No more comment
3.6. No more comment.
3.7. No more comment
4. Evaluation of hull wear due to collision with ice floes
4.1. No more comment
4.2. See comment 3.3.
4.3. No more comment
4.4. No more comment
4.5. No more comment
4.6. I understand, thank you. I think it would be better to explicitly state in the text that the authors intentionally chose "unrealistic" values of Archard's wear law parameters that resulted in large values of wear to enhance the effect of the case shape change. This is very important because the reader may get the wrong impression of the methodology proposed by the authors.
It also seems that the authors forgot to specify the number of the equation in the revised text: «In this study, the wear amount was evaluated by applying the Archard wear law, where the wear amount is determined by the shear work and the material constant C, as shown in Eq.»
5. Conclusions
5.1. No more comment
Author Response
Dear Reviewer,
I trust this letter finds you well. First and foremost, we extend our heartfelt gratitude to you and all the reviewers for the continued constructive feedback during the review process of our manuscript, titled "Application of DEM coupled with CFD to predict the erosive wear behavior of Arctic vessel hulls subjected to ice impacts."
First and foremost, Thank you so much for your sincere and valuable comments. Your insight into ice mechanics has been invaluable to us. We were able to learn very well from you the fundamental shortcomings of our work, what we need to improve, and what needs to be explained in detail.
We have taken great care to address each comment raised by you. These are detailed in the accompanying document, and the necessary amendments have been made in the manuscript.
We are confident that these focused revisions serve to clarify specific aspects of our research, thereby elevating the quality and readability of our manuscript to meet the high standards set by JMSE. We have incorporated a detailed justification for our choice of material properties, expanded our discussion on the limitations of the Bulk Young's modulus values we employed, and clarified how our methodology may be tailored for various Arctic conditions and vessel types.
We hope that the changes we have implemented will fully address your concerns and meet the approval of the editorial board, thus advancing our work towards contribution to the broader scientific community.
We eagerly await your feedback on this revised submission.
Thank you once again for your time, attention, and invaluable contributions to the refinement of our work.
Author Response File: Author Response.pdf
