Location Detection Method of Detector in Pipeline Using VMD Algorithm and Machine Learning Classifier
Round 1
Reviewer 1 Report
Justifications as to why the Variational Mode Decomposition (VMD) algorithm is used needs to be explained. The scope of experimentation in this study is too few to be designated as a scientific peer reviewed manuscript. The overall language contains too many jargon and grammatical errors. Overall, I felt like this study has some major methodological flaws and requires additional details regarding the approach made, as well as the purpose of this study.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The min problem is the quality of English text: there are many incorrectly formed sentences, many of them incomprehensible.
The analysis of vibration signals seems to be new: a combination of VMD and some classification algorithms. The superiority of VMD is shown.
Due to imprecise formulation and description i is difficult to understand what methods are used, which signals were compared at a given chapter of the paper. Weak point is also low number of comparisons to other methods.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper under consideration proposes to use the ground marker based on the method of detecting vibration signals along with Variational Mode Decomposition algorithm in internal detector so as to inspect pipelines. It is proposed instead of the magnetic flux leakage detection which has some disadvantages as compared to vibration signal detection.
On the whole, the paper looks scientifical:
The theme of article is in the scope of Electronics,
Used English is in good level,
There are all essential sections,
Abstract is adequate to article content but the unproven advantages of vibration signal detection
Keywords are correctly proposed
Data are presented clear,
Content is coherent and cohesive.
Quality of figures and tables is sufficient
Conclusion give the main findings
However, there are concerns to the manuscript:
- The literature review contains of 20 sources which look up-to-date and are in considerable journals but only 3 out 5 of positions are not from Chinese authors. 2 of 3 are devoted to Machine Learning Classifier. The question is why the problem of inspecting pipelines and reducing noise introduced in the introduction are not addressed by other authors in your review? Is it not relevant to them? I doubt it. Please, expand the review in accordance with this concern.
- Authors state that the main innovations of the paper is the use of the vibration signal detection method which makes up for the problem of magnetic flux leakage detection due to the low cost and fast processing speed of vibration signal detection. But there are no values comparisons between these 2 kinds of detections. I doubt this assumption unless there are discussion with numbers.
- Authors state that magnetic flux leakage detection is the most widely used among many detection methods. But they list only this kind of detection and further proposed by them vibration signal one. Please, devote more sources to the detection’s methods. Just short description.
- Authors state that magnetic flux leakage detection has the following shortcomings: high power consumption, high cost, low data transmission efficiency and complicated operation process whereas vibration signal detection has advantages in low cost and high data transmission efficiency. What about other 2 shortcomings? Which detection method is better?
- Please, clarify why do you cite the 18th source (2021 year) for decision tree which was the base of an ensemble learning algorithm (random forest) proposed by Leo Breiman and Adele Cutler in 1995 year?
- The flowchart of random forest algorithm will not work because it is depicted not correctly. According to ISO 5807 there are no terminal, Input\output symbols. The used symbols are in doubt. Please, redraw flowchart and describe it step by step (more clearly) in the manuscript.
- Please, check the meaning of sentence “Among many detection methods, magnetic flux leakage detection is the most widely used and widely used method at present.” I guess that the ending of the sentence after the word “and” should be deleted
- Where is the call out of Figure 1?
- Authors should confirm that the source 20 is correctly used because the article 20 is devoted to the application of MLP to analyze a patient's cardiac activity through ECG (electrocardiogram). Is this paper is devoted to classification and prediction in image? If not better to add “for example” after other fields. But anyway, it demands more sources, in my opinion.
- According to journal requirements “All figures and tables should be cited in the main text as Figure 1, Table 1, etc.”. Do not use FIG. Just correct it.
- Please, correct the caption to Figure 4 which must be not in italics
- Authors have cropped the “e” in words “amplitude” in figure 5 and “y” in word “layer” in figure 7
- Why there is no discussion to Figure 7?
- In the last paragraph of 5.3.2., Change “By analyzing the two images” with “By analyzing the two figures”. Thank you.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors made major improvements towards research clarification statement. While the methodology and results are better related now, the text itself requires proofreading. There are too many grammatical error so I would recommend that part corrected.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
There is still a lot of problems with the clarity of text, however, the authors did a lot of efforts and answered all my concerns. The paper in present form is almost ready for publication. I strongly recommend proofreading by a native speaker.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Thank authors that they have corrected the manuscript in accordance with my concerns. I think the paper could be published in that form after the last concern. It is the comment number 7. Please, use Input\output symbol correctly. For authors' case is "Input the original training set". It means - change a rectangle with a parallelogram.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
While all the concerns have been addressed and methodology now adheres to the scientific domain, the conclusive comments and the results are still not very clear. Minor editing needed in Results and conclusion.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
