Data Knowledge Dual-Driven Rate of Penetration Prediction Method for Horizontal Wells

Round 1

Reviewer 1 Report

Dear authors,

I read carefully your manuscript, named “Data Knowledge Dual-Driven Rate of Penetration Prediction Method for Horizontal Wells“, and I find it a decent paper.

A lot of your literature is domestic, so you should supplement it with international references, given that this topic is discussed on a global level. Generally, Introduction part must be improved, you must provide more information because enough relevant data has been published. I suggest you to read following articles, and add some useful information in the first part of Introduction of your article:

El Sabeh, K., Gaurina-Međimurec, N., Mijić, P., Medved, I. & Pašić, B. (2023) Extended-Reach Drilling (ERD) - The main problems and current achievements. Applied Sciences, 13(7), 4112.

 

Gaurina-Međimurec, N., Pašić, B., Mijić, P., & Medved, I. (2021) Drilling fluid and cement slurry design for naturally fractured reservoirs. Applied Sciences, 11(2), 767.

 

Also, Discussion part of your article must be covered in more detail that you present it in this manuscript. Add more information in Conclusion section.

Plagiarism software detected 13,3% of matching with other published materials, so this number should be lower. Although, I do not consider this figure to be inappropriate.

Here is the list of some specific objections:

Line 9 – always write vertical wells instead straight wells;

Line 44 – add comma (,) between reduction[10], kriging difference[11], outlier rejection;

Line 52 – when you write abbreviation for the first time in paper, like MW-SVR, write what does it stands for;

Line 56 – samo as last comment, this time for ANN;

Line 61 – what scholars? Please do not use this term, write scientists or any other synonym;

Line 74 – don't write well borehole, this two words do not fit together. Write wellbore;

Line 79 – leave space between bit and (WOB);

Line 80 – leave space between WOB and (SWOB);

Line 83 – cuttings instead cutting;

Lines 86-90 – you must describe in more detail the part about trajectory control, since there are also some other ways and tools that serve to achieve trajectory control;

Line 103 – (p. 987-992) shouldn't be written in text, leave it to references;

Line 104 – write some other synonym instead of discretizing;

Lines 108 and 109 – uniform font size and write what every mark in equations means and what are their units;

Line 126 – leave space between load and (HL);

Line 129 – is Figure 3 your original scheme, or is used from some other paper or book?;

Line 132 – the definition [28]  instead the[28] definition;

Lines 137 and 138 – uniform font size;

Line 140 – you wrote; shale gas cluster horizontal well platform in Sichuan. What do you mean by term platform, what does it stand for? Jacket, production platform, offshore drilling rig…? You must specify this;

Line 153 – make a neat table in which each column is separated by a line, this is unreadable;

Line 160 – Figure 5 instead Figure5;

Line 164 – please explain why there is not much difference between surface WOB and corrected WOB at about 3500-3550 m of measured depth, but is significant difference at 3550 m. On greater depth than this there are also intervals with smaller and greater differences between this two parametres. White your explanation and add it to your paper above or below Figure 4.

Line 196 – in Figure 7. you have written knowadge, correct it;

Lines 200 and 201 – Horizontal wells are the same as straight wells in that rock fragmentation is achieved by… - bolded part of the sentence cannot be written like that;

Line 202 – pump flow rate (Flow rate) - the text in brackets is unnecessary;

Line 204 – state variables? What do you mean by term state?;

Line 210 – leave space between depth and (MD);

Line 211 – greater would be better term than larger;

Line 218 – write what is D&T model;

Line 222 and 223 – You have written: As shown in Step 3 in Figure 7, a data knowledge dual-driven ROP prediction model 222 is constructed in this section considering horizontal well characteristics.

I don't see this in step 3;

Line 241 – specify term platform;

Line 243 and 244 – The actual drilling trajectories are confined within a 3.5-meter box…

Elaborate this;

Line 290 – instead Figure 11-12, write Figures 11 and 12;

 

Best regards,

Reviewer

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

 

I carefully read your manuscript related to the rate of penetration prediction in horizontal wells. Rate of penetration prediction is one of the main tasks for drilling engineers, and good prediction results in drilling process optimization as well as cost reduction. In your work, you singled out all relevant parameters that influence the drilling process and rate of penetration in horizontal wells. Also, your work represents a good practical approach to eliminating the influence of horizontal well characteristics on ROP prediction. Although your valuable work, there are a few vagueness and small that need to be clarified before manuscript publishing:

-  In the manuscript title you have the word “dual-driven” but you don’t have any explanation for it in the text of the manuscript;

- Line 25: please replace “tools” with the more appropriate word “ways”;

- General: each abbreviation must be fully defined at its first appearance in the text of the manuscript;

- Line 79: please replace “pressure” with the more appropriate word “weight”, because we usually talk about Weight On the Bit (WOB);

- Lines 77-80: the sentences are a bit unclear, please revise them;

- Line 81: “The drill string in the horizontal part of the wellbore is eccentric to the lower side…”;

- Line 103: when you refer to the specific paper and author(s) in the text of the manuscript, please use their surnames (for example proposed by Johancsik et. al not proposed by [27]);

- General: all physical quantities in the equations need to be fully defined (for example αi, βi, N, Li); Furthermore, on lines 108 to 109 you have F0, but I cannot find it in equations 1 or 2.

- Line 128: In Figure 3. you show an order of the operations you have undertaken to calculate the friction coefficient. This friction coefficient you use for further calculation, but I can't find it anywhere exact calculated data;

- Line 153 (table 1): Can you explain more in detail, which data you collected during drilling wells A1 to A4;

- Line 158: can you explain “factoring in drag”;

- Lines 155 to 163: in the last paragraph on page 5 you talk about figures 4. and 5., and the differences between sliding and rotating drilling modes, but the differences between these two modes aren’t visible in the mentioned figures;

- Line 160: Please add space between Figure and 5 as well as show and case;

- Line 166 (fig. 5.): in my humble opinion, it is better to say real downhole RPM instead of calculated RPM;

- Line 171: t (in the middle of the sentence)?

- Line 181: Can you explain how you calculated the values presented in Figure 6.?

- Line 186: Can you explain the terms “operation variable, state variable, and formation variable”; especially because on page 8 (line 203.) you have “characteristic variable”;

- Line 218: D&T?

- Line 219: Can you explain, how you correct the drilling parameters for the drilling characteristics of horizontal wells;

- Line 241: the platform; did you mean drilling rig?

- Line 248 – 249: “The simulation includes data from…”

- Line 253: From Figure 10 we can see that each considered well has a 1500 m horizontal section length, but in Table 1 you present information for only 700 m per well;

- Lines 295 to 303: this part of the text is a bit unclear, and I strongly recommended revising it;

- General: Can you analyse separate data for sliding and rotating modes? It is very important for the impact assessment of the drilling mode on the ROP prediction of the proposed new model.

Best regards,

reviewer

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The advantage of horizontal wells

At present, there is a slow depletion of reserves in the oil industry, and an increasing part of them falls on hard-to-reach fields.

The complexity of production is that they are characterized by high viscosity of oil and sea shelves. The analysis and effectiveness of the use of horizontal wells is confirmed by oil reserves.

The use of horizontal technologies greatly increases the efficiency of reserve development. They imply the drilling process and, in fact, the horizontal wells themselves. They have the most significant extended zone.

The main indicator in the construction of these wells is efficiency. 

Generalization of the conditions of application and effectiveness of the use of horizontal wells in the development of gas and oil fields proved that current technologies and special hardware make it possible to drill wells of almost any trajectory with a possible backlash of no more than 2 m. The use of such wells increases with a decrease in reservoir thickness and an increase in the heterogeneity of its structure. One horizontal well can replace 5 vertical wells, and if we take into account the factor of heterogeneity, the ratio can be 1:20.

 

The topic raised by the authors is relevant and contemporary. This paper couples the drag and torque (D&T) model and the data-driven method to propose a data knowledge dual-driven horizontal well ROP prediction model for horizontal wells. This paper proposes a data knowledge dual-driven horizontal well ROP prediction model, which provides a basis for intelligent ROP optimization and management during the drilling process of horizontal wells.

 

A special feature is the calculation

Once the drill string description, survey data, and friction coefficient are specified, the calculation starts at the top of the drill string and proceeds stepwise down. The tension at  the top of the first unit is known to be the hook load, then the tension at the end of the  first unit can be calculated according to equations (1) - (2). The pulling force at the end of  the previous unit is used as the pulling force at the top of the next unit in turn, and the  calculation is iterated to the bottom of the drill string, then the pulling force at the end of  the last unit is the downhole WOB.

The difference is, in this paper, a D&T model is introduced to correct the drilling parameters for the drilling characteristics of horizontal wells, which is also part of the data pre-processing.

The simulated dataset includes real-time mudlogging data, drill bit records, drill string description, sur- vey data, drilling logs, and more. Notably, for the testing dataset, all four wells (A1 to A4) are equipped with the same type of drill bit.

ROP prediction method is proposed, and the prediction accuracy of the method is improved by 30% compared with the conventional data-driven ROP prediction method. It shows that the method proposed in this paper can provide the basis for the intelligent optimization and management of ROP in the horizontal well drilling process.

Figures, tables, introduction, abstract, keywords, literature correspond to the research topic. There are no comments .

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

I am satisfied with the revised version of your paper after the review.

Best regards,

Reviewer

Reviewer 2 Report

I reviewed the revised version of the manuscript, and I can confirm that the authors made an extra effort and improved their manuscript significantly. They answered point-by-point all my observations and comments as well as comments posted by other reviewers, and I am certainly satisfied with the revised version.

Minor editing of English language required