Transport Performance Improvement of a Multiphase Pump for Gas–Liquid Mixture Based on the Orthogonal Test Method
Round 1
Reviewer 1 Report
Unfortunately, the introduction part is not acceptable due to the lack of wide discussion about the state-of-art. Reader can't find such works like 10.13140/2.1.3936.4804 or
IMPELLER-DIFFUSER INTERACTION IN CENTRIFUGAL COMPRESSORS - Proceedings of ASME Turbo Expo 2012 GT2012 June 11-15, 2012, Copenhagen, Denmark
Author Response
Dear reviewer：
Think you for your comments ,those comments are all valuable and very helpful for revising and improving our paper,as well as important guiding significance to our researches.we have studied comments carefully and made correction.The main corrections in the paper and the responds to your comments are as following:
Point1:
Unfortunately, the introduction part is not acceptable due to the lack of wide discussion about the state-of-art. Reader can't find such works like 10.13140/2.1.3936.4804 or IMPELLER-DIFFUSER INTERACTION IN CENTRIFUGAL COMPRESSORS - Proceedings of ASME Turbo Expo 2012 GT2012 June 11-15, 2012,
Response 1: We have reorganized the introduction and made a discussion about the state-of-art.
Reviewer 2 Report
Recommendation: Major Revision
Comments to the Authors:
Ms. Ref. No.: processes-1284947
Title: Transport performance of multiphase pump on gas-liquid mixture is improved based on the orthogonal test method
Overview and general recommendation:
The present manuscript address optimization of the transport performance of rotodynamic multiphase pump on the gas-liquid mixture. The article is interesting, and despite addressing a topic that is not new, it presents new insights, the orthogonal test design method and CFD technology are used to optimize the head and efficiency indexes underrated flow rate with 15% inlet gas volume fraction. The paper is well organized, but the description of the numerical methodology is incomplete. The computational setup is too generic and fails to give appropriate insight into the numerical setup, in what is a CFD study. Relevant equations for the present CFD study should be added, including detailed information about the SST k-w turbulence model. The flow configuration/computational domain is completed, and the boundary conditions are well described. The mesh information is incomplete; the mesh independence study is referred to but is not shown. The study of y+ is ignored. In general, not enough information is provided for the replication of the obtained data. A corresponding description of predictive/ corrective steps taken in the numerical solver would provide the reader with a full extent of the difficulties associated with the simulation of these types of phenomena. Without a full extent description of the numerical solver, it is not possible to analyze the obtained data, as such an analysis is dependent upon the numerics used, as well as their limitations. However, the paper is worth to be published but only after implementing the reviews addressing the comments/questions detailed before.
Author Response
Dear reviewer:
Think you for your comments ,those comments are all valuable and very helpful for revising and improving our paper,as well as important guiding significance to our researches.we have studied comments carefully and made correction.The main corrections in the paper and the responds to your comments are as following:
Point1:detailed information about the SST k-w turbulence model.
Response1:SST k-w turbulence model have been added to the relevant content of the paper.
Point2:the mesh independence study is referred to but is not shown.
Response2:Grid independent data has been added in the paper.
Point3:The study of y+ is ignored.
Response3:The y plus of the impeller and diffuser blades has been displayed in the paper.
Response4:the information on solver Settings is supplemented.
Reviewer 3 Report
The article “Transport performance of multiphase pump on gas-liquid mixture is improved based on the orthogonal test method” covers details of an optimization process of a pump. The described process involves several steps: identification of objective function and independent variables, choice of numerical algorithm and model settings for simulation of fluid flow in the considered pump, analysis of results. The work may pose some interest to a reader, who specializes in optimization of technical devices. However, it suffers from two major drawbacks:
- grammar and style of English require thorough editing;
- description of numerical model and mesh independence study needs clarification, namely:
a) why air bubbles diameter is constant?
b) how this diameter is assessed?
c) why mesh size (in terms of number of elements) is specified instead of resolution (grid points per some element of pump)?
d) how mesh resolution is preserved for different geometries?
e) how mesh independence is estimated for optimal designs of pump?
f) final configuration for water provides 1.47% increase of head – it is only 3 times larger than the variation of parameters due to mesh resolution (0.5%). In other words, is it correct to claim that changes are meaningful, when final result and variation due to grid are of the same order of magnitude?
Author Response
Dear reviewer:
Think you for your comments ,those comments are all valuable and very helpful for revising and improving our paper,as well as important guiding significance to our researches.we have studied comments carefully and made correction.The main corrections in the paper and the responds to your comments are as following:
Point1：Extensive editing of English language and style required.
Response 1:We have improved the grammar and language style of this paper.
Point2:why air bubbles diameter is constant?
Response 2: In order to simplify the study variables, the bubble diameter is constant.
Point3:how this diameter is assessed?
Response3: on the one hand ,comprehensive selection were made through experimental observation and literature research .
Point4:why mesh size (in terms of number of elements) is specified instead of resolution (grid points per some element of pump)?
Response4:Different models do not have the same number of elements as the base pump.
Point5:how mesh resolution is preserved for different geometries?
Point6:how mesh independence is estimated for optimal designs of pump?
Response5,6:In the independent verification of the model grid, only the reliability of the base model were verified. Since the base model differs little from other models, therefore. Based on the verified mesh of the base model, the mesh generation parameters of other models were set in TurboGird, which ensure the mesh of other models meets the accuracy requirements.
Point7:final configuration for water provides 1.47% increase of head – it is only 3 times larger than the variation of parameters due to mesh resolution (0.5%). In other words, is it correct to claim that changes are meaningful, when final result and variation due to grid are of the same order of magnitude?
Response7:Since the grid error is guaranteed to be within 0.5% and the increment of head is 1.47%, this increment will not be caused by error, so I think it makes sense.
Round 2
Reviewer 1 Report
The draft of publication has been improved. In the present form it can be accepted for publication.
Author Response
Thank you for your approval of this article.
Reviewer 2 Report
Recommendation: Major Revision
Comments to the Authors:
Ms. Ref. No.: processes-1284947 (round 2)
Title: Transport performance of multiphase pump on gas-liquid mixture is improved based on the orthogonal test method
Overview and general recommendation:
The draft of the manuscript has been improved, but the comments were not appropriately addressed. The reviewer suggests rewriting chapter "3. Numerical method" to include all the information to reproduce the numerical simulations. All relevant equations must be added to the manuscript (RANS equations). All variables must be described in the text (e.g., equation (2) F1?), and the turbulence model constants should be added. The study of y+ is mentioned but should be explained what means "scalable wall function" and the implications to the flow development. What is the meaning of the sentence "The advection scheme and turbulence numerics are selected with high resolution, which makes the solution more accurate."? As mentioned in the previous revision, without a full extent description of the numerical solver, it is impossible to analyze the obtained data, as such an analysis depends on the numerics setup used and their limitations. The paper is worth publishing, but only after implementing the reviews addressing the comments/questions detailed.
Author Response
Dear reviewer:
Think you again for your comments .The main corrections in the paper and the responds to your comments are as following:
Point1：All relevant equations must be added to the manuscript (RANS equations). All variables must be described in the text (e.g., equation (2) F1?), and the turbulence model constants should be added.
Response1：Related information has been added in the corresponding section of this paper.
Point2：The study of y+ is mentioned but should be explained what means "scalable wall function" and the implications to the flow development.
Response2：Related information has been added in the corresponding section of this paper.
Point3:What is the meaning of the sentence "The advection scheme and turbulence numerics are selected with high resolution, which makes the solution more accurate."?
Response3:High Resolution is the second-order precision, while the accuracy of the first-order upwind scheme is not so accurate. Therefore, High Resolution is relatively more accurate.
Reviewer 3 Report
Actually, comments were not addressed properly, for example:
1) Let's take line No. 117 (just as one example among many others): "...Therefore, numerical simulation model is consisted of ..." -- passive voice is used instead of active voice "consists of"
Lines 120-121: The order of words must be reverted in the sentence to get sensible statement.
I would strongly recommend to use proofreading service.
2) OK
3) But still there is no justification for the choice of bubbles diameters. Value of this quantity is also missing.
4) OK
5,6) contradicts with 4), because I see:
in response to 4): "Different models do not have the same number of elements "
in response to 5,6): "which ensure the mesh of other models meets the accuracy requirements"
7) Is it possible to provide actual value of grid-associated error instead of it's threshold (0.5%)?
Author Response
Dear reviewer:
Think you again for your comments .The main corrections in the paper and the responds to your comments are as following:
Point1:Let's take line No. 117 (just as one example among many others): "...Therefore, numerical simulation model is consisted of ..." -- passive voice is used instead of active voice "consists of"Lines 120-121: The order of words must be reverted in the sentence to get sensible statement.I would strongly recommend to use proofreading service.
Response1:To be honest,we have had professionals improve the style and grammar of this paper. Therefore, it was really difficult for us to revise this paper again. This time, we have revised this article according to your comments. I hope you can forgive us if there are still any mistakes.
Poimt2：But still there is no justification for the choice of bubbles diameters. Value of this quantity is also missing.
Response2：Although our team has not conducted experiments on bubble diameter measurement of the pump in this paper, other teams have conducted visual experiments on pump of the same type and very similar size, and measured the diameter of bubbles in the pump under actual working conditions.The bubble diameter used in this paper is based on the experimental results of other team.
Poimt5、6：contradicts with 4), because I see:in response to 4): "Different models do not have the same number of elements "in response to 5,6): "which ensure the mesh of other models meets the accuracy requirements"
Response5、6：The optimized geometric parameters are mainly the geometric shape of the blade, the size of the pump itself changes very little. Therefore, the same mesh generation parameters can ensure the mesh accuracy of different models.
point7: Is it possible to provide actual value of grid-associated error instead of it's threshold (0.5%)?
Response7：The actual operating condition of mutiphase pump is mainly multiphase flow. If you really think that the performance improvement in pure water conditions may be caused by grid errors, then,I choose to delete these inaccurate descriptions.
