Current Status of Research on Fault Diagnosis Using Machine Learning for Gear Transmission Systems

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Manuscript title

Current Status of Research on Fault Diagnosis of Gear Transmission System

 

Authors

Xuezhong Fu, Yuanxin Fang, Yingqiang Xu, Haijun Xu, Guo Ma and Nanjiang Peng

 

The purpose of the presented manuscript is to analyze the current state of fault detection of gears used in power transmissions. The authors summarize the development and current state of the diagnosis of mechanical gear transmission faults. Then, the methodology of fault diagnosis of gear transmission system is discussed in three scenarios, time domain, frequency domain and frequency time domain. Then, the relevant research progress in shallow learning and deep learning in the field of fault diagnosis is outlined. Future research directions in the field of gear transmission fault diagnosis are outlined in the paper, in terms of signal dispersion, separation of adjacent characteristic components, weak signal extraction, composite fault identification, multifactor combinations in fault diagnosis, and multi-source data fusion technology.

The authors described suggestively the content of the work in the graphical abstract placed right at the beginning of the manuscript.

 

I appreciate that the paper is well structured, and the bibliographic study is well organized, so that it classifies and briefly presents most of the problems related to the detection of defects in gear transmissions.

 

The main topics addressed are as follows: development of fault diagnosis of gear transmission system, fault diagnosis method based on mathematical signal processing, time domain analysis method, frequency domain analysis method, time-frequency domain analysis method, fault diagnosis method based on intelligent diagnosis, diagnostic methods based on shallow learning, diagnostic methods based on deep learning.

Although the work does not contain a paragraph dedicated to discussions, the authors introduced relevant and pertinent comments during the bibliographic analysis, relevant to the advantages and limitations of the analyzed methods.

 

I would suggest introducing a comparative discussion under the aspect of practical implications regarding surface learning and deep learning in the field of fault diagnosis.

I appreciate the manuscript provides relevant data useful for the design of gear transmissions  with high reliability and improved operation.

I would recommend a careful check of the text to correct existing typographical errors.

 

Finally, I appreciate that the manuscript can be considered for publication, after a minor revision.

 

Comments for author File: Comments.pdf

Author Response

We thank you for the valuable suggestions on our manuscript. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have seriously considered the questions and made a detailed point-by-point response to the comments. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I have read the manuscript several times, with the desire to find its meaning and purpose, as well as benefit for the readers. The work is almost encyclopedic with poor overview content, without critical reviews, comparative analyses and case studies - concrete examples, related to gear transmissions. First, the given manuscript is not an article but a review paper of a very small volume. It does not bring any scientific contributions or original scientific solutions. The authors only summarize the various fault diagnosis methods in general, including their development, current research trends, and future directions. No new experimental data or analytical findings are presented, which is characteristic of research articles. Instead, they extensively review existing methods, including mathematical signal processing and machine learning techniques, claiming that this is in the context of gear fault diagnosis. The scope is quite broad and may be relevant to other rotating machinery components, such as rolling bearings, for which the same diagnostic techniques are used. Many of the methods discussed (eg wavelet transform, empirical mode decomposition) are applicable to various types of rotating machinery, not just gear systems. There is no content in the paper that refers specifically to gear pairs, to connect the described diagnostic methods with the specifics of the gear transmission. An analysis of the connection between the results of the applied diagnostic methods and the specific parameters of the gear transmission (rotational speed, pitch, manufacturing accuracy, degree of meshing, stiffness of the teeth...) is missing. In the entire text, instead of "gear transmissions" you can write "rolling bearing" for example and everything will be the same and correct. To improve the manuscript, the authors could highlight the unique original aspects of the diagnosis of gear pairs, which distinguish them from other rotating machinery components. To provide examples of case studies or specific applications that demonstrate the use of these diagnostic techniques specifically for gears, with appropriate critical view and analysis. This would provide the necessary balance between general principles and specific applications, making the content more targeted.

Author Response

We are very grateful to the reviewers for their detailed review of our manuscript and their valuable comments. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have seriously considered the questions and made a detailed point-by-point response to the comments. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

 

 

 

 

 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

-

Author Response

We thank you for the valuable suggestions on our manuscript. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have seriously considered the questions and made a detailed point-by-point response to the comments. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript should be declared as a review paper, not an article!