Creation of a Hydrodynamic Digital Model of a Laboratory Core Experiment of Surfactant Polymer Impact on Oil Recovery, in Order to Determine Parameters for Further Full-Scale Simulation

Round 1

Reviewer 1 Report

A digital hydrodynamic model to describe the process of waterflooding with surfactant and polymer solutions in cores was created. The problem of determining, validating, and transferring model parameters of the chemical EOR model was solved. It is also extended to further full-scale simulation. The article is meaningful.

However, there are some unclear situations in the text:

1 The references in this article, especially the preface, are too outmoded, and do not express the frontier research progress of this subject.

2 The graphs in the text are not standardized, table 5 appears twice, and the axes labels and figures are blurred, it is recommended to use the original graphs.

3 Figures 10 and 11 compare only the differences in oil recovery and water cut between the different models, and it is recommended to add the actual formation for comparison.

4 the process of obtaining uncertainty parameters in model tuning should be described in detail.

Author Response

Thank you for your comments!!!

1 The references in this article, especially the preface, are too outmoded, and do not express the frontier research progress of this subject.

We add more references in our article

2 The graphs in the text are not standardized, table 5 appears twice, and the axes labels and figures are blurred, it is recommended to use the original graphs.В последней версии статьи что я скидывал всё в редактируемом формате и таблица 5 не дублировалась

We add all graph in new format.

3 Figures 10 and 11 compare only the differences in oil recovery and water cut between the different models, and it is recommended to add the actual formation for comparison.

These calculations were performed only to select optimal size of the cells, data from another experiments were used. Do you suggest adding them?

4 the process of obtaining uncertainty parameters in model tuning should be described in detail.

We add more explanation, from line 78.

Author Response File: Author Response.docx

Reviewer 2 Report

The study focuses on understanding transferring phenomenon of chemical enhanced oil recovery (EOR) from a core experiment to a full-scale simulation model. But following items should be addressed.

1. The labels of figures are two small with a low resolution. Suggest enlarging them.
2. For Figure 4, more data are needed to demonstrate this relationship
3. The figure 7 and 8 are not mentioned in the paper. How do the authors obtain the data?
4. What is the experimental equipment are used for the tests?
5. It is necessary to mention the name of the used commercial simulator. It looks like CMG GEM.
6. A cylinder model is often used to model lab performance. Is the model a cylinder or cartesian one?
7. There are many uncertainties about the simulation model and experiment data as mentioned in the paper, but how could the authors match the lab data. What are the inputs and what the matching targets?

Author Response

Thank you for comments!!

1. The labels of figures are two small with a low resolution. Suggest enlarging them.

We add all graph in new format.

 

1. For Figure 4, more data are needed to demonstrate this relationship –

 

There were only 2 measurement of viscosity. Also, we add some comments from line 111.

 

1. The figure 7 and 8 are not mentioned in the paper. How do the authors obtain the data? –

 

We add mentions on this figures in text.

 

1. What is the experimental equipment are used for the tests? –

 

We add experimental equipment in text. Line – 82.

 

1. It is necessary to mention the name of the used commercial simulator. It looks like CMG GEM.

 

Yes, it was CMG GEM, we remove it from text to avoid hidden advertising.

We add reference (number 22), there is a link to CMG.

 

1. A cylinder model is often used to model lab performance. Is the model a cylinder or cartesian one?

 

There was Cartesian one. We add mention on line 179.

 

1. There are many uncertainties about the simulation model and experiment data as mentioned in the paper, but how could the authors match the lab data. What are the inputs and what the matching targets?

 

We add explanation, line 78-101.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript presents a hydrodynamic digital model to investigate the surfactant polymer impact.  Although the authors acknowledged the limitations of the model (briefly), there are multiple insufficiencies that do not warrant a publication at this stage. 

1. The title is ambiguous. "Surfactant polymer impact" on what? It in not clear from the title what kind of impact is the study about. 
2. The manuscript presented the findings in a rather general and ambiguous manner. Results should be presented in a more quantitative way and choice of words should be more descriptive/clearer. Results should be presented as "% change from X to Y" instead of "quite significantly". As this is a simulation paper, there should be quantitative values supporting the claims and they should come with error bars. 
3. Figures, especially figure legends, should be improved. Grid lines, inconsistent significant figures/decimal places of axes, and compressed legend labels make it difficult to read.
4. Multiple grammatical errors are found. 
5. Solid lines should not be used to interpolating raw data that was collected as discrete data points. 
6. When a power function is used for extrapolation, the power function equation should be provided. 
7. The assumptions made at each step should be presented clearly. 
8. As the simulation involves multiple approximation, error bars should be presented with the results. 
9. Some of the parameters are not well defined, e.g. "cell size by I" in Table 3 and the figure legend labels listed as abbreviations. 
10. A figure or schematic diagram of the simulation to define the parameters or variables will be extremely helpful. 
11. Supporting information was not provided. If the authors do not intend to provide supporting information, please remove the section from the manuscript to avoid confusion to the readers. 

Author Response

1. The title is ambiguous. "Surfactant polymer impact" on what? It in not clear from the title what kind of impact is the study about. 

 

We changed title “impact on oil recovery”

 

1. The manuscript presented the findings in a rather general and ambiguous manner. Results should be presented in a more quantitative way and choice of words should be more descriptive/clearer. Results should be presented as "% change from X to Y" instead of "quite significantly". As this is a simulation paper, there should be quantitative values supporting the claims and they should come with error bars. 

 

We add explanation on line 205 and line 218

 

1. Figures, especially figure legends, should be improved. Grid lines, inconsistent significant figures/decimal places of axes, and compressed legend labels make it difficult to read.

 

We changed all graphs.

 

1. Multiple grammatical errors are found.

 

We tried to improve English in our article. Fixed some error.

 

1. Solid lines should not be used to interpolating raw data that was collected as discrete data points. 

Could you please clarify your comment? There are no solid lines describing interpolation. Could you point out figure in other case.

 

1. When a power function is used for extrapolation, the power function equation should be provided..

 

This is function of hydrodynamic simulator. We add explanation in text. Line 119.

 

1. The assumptions made at each step should be presented clearly. 

 

We didn’t understand your question, could you please clarify? What does it mean assumption on each step? Also, we add in text more explanation. Lines from 78 and lines from 205.

 

1. As the simulation involves multiple approximation, error bars should be presented with the results. 

 

Could you please clarify what is error bars? We had initial model and matched it to real experiment data, during adaptation process we define some parameter, which cannot be obtain directly from result of experiment. In addition, we add more explanation related to errors of adaptation. Lines from 205.

 

1. Some of the parameters are not well defined, e.g. "cell size by I" in Table 3 and the figure legend labels listed as abbreviations. 

We fixed it. Would you suggest describing abbreviations like cP, mN and so on?

 

1. A figure or schematic diagram of the simulation to define the parameters or variables will be extremely helpful

 

We add it.

. 

1. Supporting information was not provided. If the authors do not intend to provide supporting information, please remove the section from the manuscript to avoid confusion to the readers. 

 

We don’t understand this comment. What did you mean – supporting information?

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thank you for revising.

Author Response

Good day,

Thank you for review.

We have corrected some mistakes in the text of article.

 

Author Response File: Author Response.docx

Reviewer 3 Report

1. Solid lines should not be used to interpolating raw data that was collected as discrete data points. 

Could you please clarify your comment? There are no solid lines describing interpolation. Could you point out figure in other case. 

For example, the data points for arithmetic mean error presented in Figure 14 are clearly discrete data points as the author only used a few models for simulation. The data points should not be connected using solid lines. The same applies to Figure 3. 

1. Multiple spelling errors can still be found throughout the text. e.g.,
   -Figure 2 "Cehemicla" concentration?
   -Figure 10 oil "recivery"?
2. subscripts and superscripts are not used appropriately. e.g., 
   -for units, cm3 should be cm³
   -for variables, CNaCl should be C_NaCl to avoid confusion
3. undefined parameters. e.g., 
   -cumulative oil SC appears in a lot of graphs, but it is unclear what SC means. is it an indication that the oil is in supercritical state? is SC referring to some kind of specific capacity? or is it referring to a specific type of oil used in the simulation? the readers can probably guess what "surf. conc. out" means, but figure legends should be clear.
4. as the author states that the current study will be the basis for further work, it is essential that the author and potential readers are aware of the error bars to give statistical significance to the simulation. If the simulation software is unable to provide error estimation, some form of estimation for deviation should be provided. Obviously the viscosity value does not refer to a single number but a range. 

Author Response

Good day,

Thank you for review and your remarks.

1. We have corrected all figures in article and changed type of lines.
2. We have corrected spelling errors through the text and figures.
3. Subscripts and superscripts were corrected.
4. We have added table on line 102, which describes undefined parameters and abbreviations used in text.
5. There are initial measurement errors associated with the equipment, for example, measurements of viscosity, pressure drop, concentrations. Usually, the accuracy of such measurements is quite high and does not exceed 1-2%. Further, when tuning to actual data, a simulator was used, which also has internal errors related to the convergence of the numerical solution, usually the imbalance error is no more than 0.1%. Further, when modeling, there are errors related to the choice of cell sizes (Figure 12) and the accuracy of the adjustment to the actual data (Figure 15, 16). Thus, the basic error cannot exceed the error of the instruments and the accuracy of the adjustment to the actual data. There are also errors that are related to the discreteness of the actual data, as well as mathematical models (correlations) that are embedded in the simulator. Such errors can be more significant and reach 10% (according to our expert estimates).

 

Author Response File: Author Response.docx