Combining Knowledge and a Data Driven Method for Identifying the Gas Kick Type in a Fractured Formation
Round 1
Reviewer 1 Report
The paper presents the investigation of gas kick type in the fractured formation for differentiating between kick form either due to the underbalanced condition and gravity displacement. A model presenting the wellbore and formation to simulate well kick is developed and the artificial intelligence (AI) technique dynamic time wrapping was applied for distinguishing the kick form.
The paper presents important aspect of the drilling industry and is required especially drilling fractured formation. The model formulation is appropriate although there are several simplifications are adopted for representing the system. As the study is focused on the application of the fluid mechanics and AI equation for the evaluation of the kick form the work is weak in terms of scientific contribution. However, the applied part is important for addressing the important part of the drilling industry and adding value to the application.
The conclusion from studies are more general.The following point is required to be addressed in the revised manuscript-
The phenomenon of the changes in the bottomhole pressure is not captured. The results show that the bottomhole pressure is constant with time however in reality it will change with time as gas volume in the two phase mixture will increase with time.
Comments for author File: 
Comments.pdf
Author Response
Point 1: The phenomenon of the changes in the bottomhole pressure is not captured. The results show that the bottomhole pressure is constant with time however in reality it will change with time as gas volume in the two phase mixture will increase with time.
Response 1: Figure 4 in the article shows the change curve of bottomhole pressure with time during overflow. It can be seen from Figure 4 that the bottomhole pressure gradually decreases with time, as the reviewer said, the bottomhole pressure gradually decreases with time as gas volume in the two phase mixture will increase with time.
Special thanks to you for your good comments.
Author Response File: 
Author Response.docx
Reviewer 2 Report
In this study, the authors tried to combine the knowledge and data-driven method to identify the type of gas kick in the fractured formation. The manuscript is not good enough to be acceptable for publication.
Author Response
Point 1: In this study, the authors tried to combine the knowledge and data-driven method to identify the type of gas kick in the fractured formation. The manuscript is not good enough to be acceptable for publication.
Response 1: First of all, thank you very much for reviewing our paper. Please allow us to introduce the work of this article to you again.
The paper presents the investigation of gas kick type in the fractured formation for differentiating between kick form either due to the underbalanced condition and gravity displacement. A model presenting the wellbore and formation to simulate well kick is developed and the artificial intelligence (AI) technique dynamic time wrapping was applied for distinguishing the kick form. Our paper presents important aspect of the drilling industry and is required especially drilling fractured formation. The model formulation is appropriate although there are several simplifications are adopted for representing the system. As the study is focused on the application of the fluid mechanics and AI equation for the evaluation of the kick form the work is weak in terms of scientific contribution. However, the applied part is important for addressing the important part of the drilling industry and adding value to the application. Compared with existing method, the method proposed in this paper does not need to wait for a lag time, and there is no need to adjust wellhead backpressure, also the fine outlet flow measurement also not required, which is not only suitable for managed pressure drilling but also for conventional drilling.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper is well written. However, the author needs to clearly mention the novelty of this paper in the introduction section.
The results should be compared with other conventional methods.
Model initialization needs to be added
Author Response
Point 1: The paper is well written. However, the author needs to clearly mention the novelty of this paper in the introduction section.
Response 1: we have clearly mentioned the novelty of this paper in the introduction section. Lines 84-92.
Point 2:The results should be compared with other conventional methods.
Response 2: Comparison with other conventional methods has been added in the Discussion section of the article.
Point 3: Model initialization needs to be added
Response 3: Model initialization has been added.
Special thanks to you for your good comments.
Author Response File: Author Response.docx
Reviewer 4 Report
This paper presents a method to detect and identify the type of gas kick occurring during wellbore formation. The process is based on artificial intelligence techniques, particularly the Dynamic Time Warping method. The paper is well written, and it is clear about its novelty and significance. However, I suggest the following, prior to a possible publication:
1. Please improve Figure 1 by adding labels and identifications to represent the wellbore section better. Consider that Applied Sciences is a multidisciplinary journal, and the papers must be addressed to a broad audience.
2. I strongly recommend including a confident reference to help the reader understand the continuity equations (1) and (2).
3. In Figure 3, I suggest placing the x-axis label of the graph at its bottom. Also, please consider that the lines are not visible at the selected scale. Please improve this Figure for more precise observation.
4. I consider that the theoretical background of the gas kick phenomenon and the basic explanation of the DTW algorithm should be presented before the experimental section. Please refer to the journal recommendations and paper examples to perform it accordingly.
5. In the Results section, only one measured time series is compared to the simulation to compute the dynamic distance. Why do the authors consider only one comparison enough to validate their proposed method?
Author Response
Point 1: Please improve Figure 1 by adding labels and identifications to represent the wellbore section better. Consider that Applied Sciences is a multidisciplinary journal, and the papers must be addressed to a broad audience.
Response 1: We have added labels and identifications in Figure 1 to represent the wellbore section better.
Point 2: I strongly recommend including a confident reference to help the reader understand the continuity equations (1) and (2).
Response 2: We have given a confident reference to help the reader understand the continuity equations (1) and (2).
Point 3: In Figure 3, I suggest placing the x-axis label of the graph at its bottom. Also, please consider that the lines are not visible at the selected scale. Please improve this Figure for more precise observation.
Response 3: We have redrawn the Figure 3 according to the reviewer's suggestion.
Point 4: I consider that the theoretical background of the gas kick phenomenon and the basic explanation of the DTW algorithm should be presented before the experimental section. Please refer to the journal recommendations and paper examples to perform it accordingly.
Response 4: We have presented the theoretical background of the gas kick phenomenon and the basic explanation of the DTW algorithm before the experimental section. Lines 28-31, lines 37-39, and lines 75-79.
Point 5: In the Results section, only one measured time series is compared to the simulation to compute the dynamic distance. Why do the authors consider only one comparison enough to validate their proposed method?
Response 5: In the Restults and Discussion section, we have added calculations at different times and compared them with drilling examples in references. The applicability of the method proposed in this paper is discussed, and the advantages of the proposed method over existing methods are discussed.
Special thanks to you for your good comments.
Author Response File: Author Response.docx
Reviewer 5 Report
The study presents a methodology combining previous data and field measurements to identify the type of gas kick. Overall, this is a very relevant topic with great significance for drilling engineering. The methodology and results presentation are easy to follow; however, the introduction and discussion could be expanded and improved. Therefore, this reviewer has only a few suggestions and observations before publication.
1. The Introduction is good but could be enriched by emphasizing the importance of this study. For example, explaining the consequences if the gas kick is not controlled properly and expanding on the difficulty of establishing the formation pore pressure during drilling for carbonated formations with well-developed fractures and caves.
2. Lines 61-78. This paragraph starts by stating the importance of AI technology, followed by the paper's organization and then explaining the AI algorithm. This reviewer suggests separating the paragraph in two, with the last paragraph describing how the paper is organized.
3. The paper ends abruptly after the results are presented. Furthermore, only one case study is presented, and the conclusions are drawn based only on this example. The authors should add a brief discussion, including the limitations of this methodology and how the proposed method is an improvement from the techniques discussed in the literature review.
Author Response
Point 1: The Introduction is good but could be enriched by emphasizing the importance of this study. For example, explaining the consequences if the gas kick is not controlled properly and expanding on the difficulty of establishing the formation pore pressure during drilling for carbonated formations with well-developed fractures and caves.
Response 1: We have enriched the Introduction by emphasizing the importance of this study in the first paragraph. Lines 28-31 and lines 37-39.
Point 2: Lines 61-78. This paragraph starts by stating the importance of AI technology, followed by the paper's organization and then explaining the AI algorithm. This reviewer suggests separating the paragraph in two, with the last paragraph describing how the paper is organized.
Response 3: We have separated the paragraph in two, with the last paragraph describing how the paper is organized.
Point 3: The paper ends abruptly after the results are presented. Furthermore, only one case study is presented, and the conclusions are drawn based only on this example. The authors should add a brief discussion, including the limitations of this methodology and how the proposed method is an improvement from the techniques discussed in the literature review.
Response 3: In the Restults and Discussion section, we have added calculations at different times and compared them with drilling examples in references. The applicability of the method proposed in this paper is discussed, and the advantages of the proposed method over existing methods are discussed.
Special thanks to you for your good comments.
Author Response File: Author Response.docx
Reviewer 6 Report
Dear authors,
I read carefully your manuscript, titled ‘Combining Knowledge and Data Driven Method for Identifying Gas Kick Type in Fractured Formation', and I expected more than presented. In my opinion your conclusion is too general, and I don’t see any scientific contribution in this paper. Maybe it would be better that you present this article on a conference or send to professional journal.
Best regards,
Reviewer
Author Response
Point 1: I read carefully your manuscript, titled ‘Combining Knowledge and Data Driven Method for Identifying Gas Kick Type in Fractured Formation', and I expected more than presented. In my opinion your conclusion is too general, and I don’t see any scientific contribution in this paper. Maybe it would be better that you present this article on a conference or send to professional journal.
Response 1: First of all, thank you very much for reviewing our paper. Please allow us to introduce the work of this article to you again.
The paper presents the investigation of gas kick type in the fractured formation for differentiating between kick form either due to the underbalanced condition and gravity displacement. A model presenting the wellbore and formation to simulate well kick is developed and the artificial intelligence (AI) technique dynamic time wrapping was applied for distinguishing the kick form. Our paper presents important aspect of the drilling industry and is required especially drilling fractured formation. The model formulation is appropriate although there are several simplifications are adopted for representing the system. As the study is focused on the application of the fluid mechanics and AI equation for the evaluation of the kick form the work is weak in terms of scientific contribution. However, the applied part is important for addressing the important part of the drilling industry and adding value to the application. Compared with existing method, the method proposed in this paper does not need to wait for a lag time, and there is no need to adjust wellhead backpressure, also the fine outlet flow measurement also not required, which is not only suitable for managed pressure drilling but also for conventional drilling.
Author Response File: Author Response.docx
