Rotate Vector Reducer Fault Diagnosis Model Based on EEMD-MPA-KELM
Khalaf Salloum Gaeid
Round 1
Reviewer 1 Report
the figures fonts should be same as the main text fonts
Author Response
Expert Opinion:
Point 1: the figures fonts should be same as the main text fonts.
Response 1: Thanks to the experts for this question. All images in the article have been redrawn and all graphic fonts have been changed to Palatino Linotype, consistent with the text fonts. See the revised graphics section for details. Using Figure 1 and Figure 2 as an example, modify as follows.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript "RV Reducer fault diagnosis model based on EEMD-MPA-KELM" focused on faults in Rotary-Vibrator Reducers using machine learning in vibration analysis.
The Introduction section provides an analysis of the machine learning field for fault identification and reflects current trends in this area, and can be useful to a wide range of readers.
The second section of the paper is devoted to the mathematical description of the processes in RV reducer. The authors should clarify coefficient B in formula (1). Formulas (7) and (8) contain stylistic problems.
The third section is dedicated to the description of machine learning algorithms - KELM and MPA. The section contains correct references to the founders of the methods.
The fourth section focused on the EMD method mathematical description. The authors should either indicate the sources of the mathematical description or clarify their personal contribution. Also, the section provides RV reducer diagnostics structural scheme. The authors clarify that bearing fault data was used for modeling. However, it is not clear from the structural scheme what exactly is being studied in the laboratory. The results of the simulation modeling show that the developed diagnostic algorithm has good convergence with practice.
The fifth section describes the test bench. It is not clear why a table with partial data is provided. Table 3 does not match Figure 9. The efficiency in Table 3 reaches 90%, but there are no such data on the graphs. The authors should clarify what they wanted to show in Table 3. In Figure 9, the average efficiency value is 86% in the absence of faults and 87% in the presence of faults. The authors should explain this. The authors should also explain how the experiment was conducted, what exactly failed, and what faults were present.
The simulation modeling presented in section 5.2 also requires clarification regarding the description of faults. The graphs in Figures 10 and 11 should be better presented on axes that correspond to each other (in the same range of axes). The simulation modeling shows good convergence of the proposed method on the test set.
In the conclusions, the authors should clarify the scientific contribution and provide a more detailed description of the results obtained for prediction.
The article requires improvement in terms of the experiment.
Author Response
Thesis revision instructions
Manuscript ID: applsci-2267917
Title: RV Reducer fault diagnosis model based on EEMD-MPA-KELM
Applied Sciences
Dear Reviewer,
Thank you very much for your thoughtful service work on our thesis.
I am very grateful to the reviewers for their comments on the paper, which are very enlightening to our research work. We very much agree with the opinions of the reviewers and have made changes accordingly. Relevant revisions are indicated in the revised manuscript.
Kind regards,
Xiaoyan Wu
2023.03
In response to the comments of the reviewers, the amendments are as follows:
Modify the description
----------------------------------------------------------------------------------------------------------------------
Response to Reviewer 2 Comments
Expert Opinion: The manuscript "RV Reducer fault diagnosis model based on EEMD-MPA-KELM" focused on faults in Rotary-Vibrator Reducers using machine learning in vibration analysis. The Introduction section provides an analysis of the machine learning field for fault identification and reflects current trends in this area, and can be useful to a wide range of readers. The third section is dedicated to the description of machine learning algorithms - KELM and MPA. The section contains correct references to the founders of the methods. In the conclusions, the authors should clarify the scientific contribution and provide a more detailed description of the results obtained for prediction. The article requires improvement in terms of the experiment. In summary, the reviewer recommends that the authors can clarify the following issues: (Refer to the red font part of the revised manuscript for the specific revisions)
Point 1: The second section of the paper is devoted to the mathematical description of the processes in RV reducer. The authors should clarify coefficient B in formula (1). Formulas (7) and (8) contain stylistic problems.
Response 1: Thanks to the experts for this question. The coefficient B in formula (1) has been reinterpreted. The stylistic problem of formulas (7) and (8) is caused by the lack of corresponding text format in the input formula. The revised formulas (7) and (8) are as follows. In addition, all formulas in this paper have been reedited by the formula editor and are now in a consistent format. See red in the revised version for details.
Point 2: The fourth section focused on the EMD method mathematical description. The authors should either indicate the sources of the mathematical description or clarify their personal contribution. Also, the section provides RV reducer diagnostics structural scheme. The authors clarify that bearing fault data was used for modeling. However, it is not clear from the structural scheme what exactly is being studied in the laboratory. The results of the simulation modeling show that the developed diagnostic algorithm has good convergence with practice.
Response 2: Thanks to the experts for this question. The fourth section focused on the EEMD method mathematical description. Now the source of the mathematical description has been indicated, see the red part in the revision. The research plan of this paper is to use the spectrum diagram to diagnose the fault from the RV reducer measured data. The data measured by the RV reducer test platform is decomposed by EEMD to obtain several intrinsic mode components (IMF). After the overall average checking and optimization of each IMF, multiple groups of characteristic values are obtained. The eigenvalues were input into the nuclear Extreme Learning Machine (KELM) optimized by the Marine Predator algorithm (MPA), and the fault diagnosis model was established. Finally, compared with other models, the prediction results showed that the proposed model could judge the RV reducer working state more effectively. The purpose of this paper is to accurately evaluate the working state of the RV reducer, so as to propose a fault diagnosis model based on EEMD-MPA-KELM.
Point 3: (1)The fifth section describes the test bench. It is not clear why a table with partial data is provided. Table 3 does not match Figure 9. The efficiency in Table 3 reaches 90%, but there are no such data on the graphs. The authors should clarify what they wanted to show in Table 3. (2)In Figure 9, the average efficiency value is 86% in the absence of faults and 87% in the presence of faults. The authors should explain this.
Response 3: Thanks to the experts for this question. (1) Table 3 with partial data is provided in this paper for the purpose of verifying the authenticity of the test data. Due to the large amount of experimental data and the randomness of the selected data, the 30 groups of data in Table 3 are not the data before and after the RV reducer failure. There is no direct connection between Table 3 and Figure 9, so there is a mismatch between Table 3 and Figure 9. In order to avoid interference, the content of Table 3 has been deleted.
(2) As shown in Figure 9, the average efficiency of RV reducer in the presence of faults is higher than that in the absence of faults, because the wear of the RV reducer crank after failure leads to weak sliding of the planetary wheel along the axis, reduced preloading force, reduced axial Angle of crankshaft force and crankshaft force, reduced radial component of crankshaft force and decreased eccentricity. The gear meshing off-load state and lubrication state are improved, the friction at the meshing point is reduced, so the transmission efficiency is increased after the failure. See red in the revised version for details.
Author Response File: Author Response.docx
Reviewer 3 Report
This work discusses the fault diagnosis of the rotate vector reducer (RV reducer), which is newly used in the field of robotics. The paper is well written and represents good work. However, some points should be taken into consideration before acceptance:
· The abbreviations like “VR, EEMD, IMF, MPA ….” should be defined when it is used for the first time, even though it is well known.
· The introduction section is well represented, and the gab in the literature has been defined by the authors. However, the contribution of the paper is not well stated. The paragraph “In this paper, a MPA-KELM fault diagnosis and classification model for RV reducer of small sample industrial robots based on EEMD is proposed” should be extended to explain in more detail how this paper will deal with this problem.
· I think that Figure 1 does not belong to the authors, and therefore, a reference should be added.
· The authors should be sure that all the symbols are defined in the paper since there is no nomenclature.
· The subscripts in Eqs. (7) and (8) should be checked.
· The subfigures in figures 10, 11 and 12 should be defined.
· The references are up to date.
Author Response
Thesis revision instructions
Manuscript ID: applsci-2267917
Title: RV Reducer fault diagnosis model based on EEMD-MPA-KELM
Applied Sciences
Dear Reviewer,
Thank you very much for your thoughtful service work on our thesis.
I am very grateful to the reviewers for their comments on the paper, which are very enlightening to our research work. We very much agree with the opinions of the reviewers and have made changes accordingly. Relevant revisions are indicated in the revised manuscript.
Kind regards,
Xiaoyan Wu
2023.03
In response to the comments of the reviewers, the amendments are as follows:
Modify the description
----------------------------------------------------------------------------------------------------------------------
Response to Reviewer 3 Comments
Expert Opinion: This work discusses the fault diagnosis of the rotate vector reducer (RV reducer), which is newly used in the field of robotics. The paper is well written and represents good work. However, some points should be taken into consideration before acceptance: (Refer to the red font part of the revised manuscript for the specific revisions)
Point 1: The abbreviations like “VR, EEMD, IMF, MPA ….” should be defined when it is used for the first time, even though it is well known.
Response 1: Thanks to the experts for this question. The abbreviations like “VR, EEMD, IMF, MPA ….” have been defined when it is used for the first time. See the red part of the revised manuscript for details.
Point 2: The introduction section is well represented, and the gab in the literature has been defined by the authors. However, the contribution of the paper is not well stated. The paragraph “In this paper, a MPA-KELM fault diagnosis and classification model for RV reducer of small sample industrial robots based on EEMD is proposed” should be extended to explain in more detail how this paper will deal with this problem.
Response 2: Thanks to the experts for this question. To illustrate in more detail how this paper addresses the target issue, Now extended the content of the paragraph "In this paper, a MPA-KELM fault diagnosis and classification model for RV reducer of small sample industrial robots based on EEMD is proposed ". See the red part of the revised manuscript for details.
Point 3: I think that Figure 1 does not belong to the authors, and therefore, a reference should be added.
Response 3: Thanks to the experts for this question. Figure 1 refers to the document "Recognition of Fault State of RV Reducer Based on self-organizing feature map Neural Network", which has been supplemented as reference 19. See the red section of the revised draft for details.
Point 4: The authors should be sure that all the symbols are defined in the paper since there is no nomenclature.
Response 4: Thanks to the experts for this question. After checking and revising, I confirm that all the symbols in the paper have been defined. See the red section of the revised draft for details.
Point 5: The subscripts in Eqs. (7) and (8) should be checked.
Response 5: Thanks to the experts for this question. The stylistic problem of formulas (7) and (8) is caused by the lack of corresponding text format in the input formula. The revised formulas (7) and (8) are as follows. In addition, all formulas in this paper have been reedited by the formula editor and are now in a consistent format. See red in the revised version for details.
(7)
(8)
Point 6: The subfigures in figures 10, 11 and 12 should be defined.
Response 6: Thanks to the experts for this question. The subfigures in figure 10, 11 and 12 have been defined. see the red section of the revised manuscript.
Point 7: The references are up to date.
Response 7: Thanks to the experts for this question. Some references have been updated , see the red section of the revised manuscript.
Author Response File: Author Response.docx
Reviewer 4 Report
The proposed manuscript titled RV Reducer fault diagnosis model based on EEMD-MPA-KELM dealt with the problem related to : RV reducer; EEMD; MPA; KELM; Fault diagnosis.
The authors presented topics: Industrial robot RV reducer, Theoretical Basis, Establishment of RV reducer fault diagnosis model, Test and data analysis, and cConclusion were supported by the data .
The paper is relevant and interesting and Schematic of controller was done with no errors.
It seems to be original , according to other results from current literature mention on the proposed manuscripit .
I t deserves publication after revisions:
The Numerical simulations must show the efficiency of the control methods, as well as the sensitivity of each control strategy to parametric errors.
The state of the art is no complete ( see paper published on Mechanical Systems and Signal Processing journal, MDPI, and others)
Author Response
Thesis revision instructions
Manuscript ID: applsci-2267917
Title: RV Reducer fault diagnosis model based on EEMD-MPA-KELM
Applied Sciences
Dear Reviewer,
Thank you very much for your thoughtful service work on our thesis.
I am very grateful to the reviewers for their comments on the paper, which are very enlightening to our research work. We very much agree with the opinions of the reviewers and have made changes accordingly. Relevant revisions are indicated in the revised manuscript.
Kind regards,
Xiaoyan Wu
2023.03
In response to the comments of the reviewers, the amendments are as follows:
Modify the description
----------------------------------------------------------------------------------------------------------------------
Response to Reviewer 4 Comments
Expert Opinion: The proposed manuscript titled RV Reducer fault diagnosis model based on EEMD-MPA-KELM dealt with the problem related to : RV reducer; EEMD; MPA; KELM; Fault diagnosis.The authors presented topics: Industrial robot RV reducer, Theoretical Basis, Establishment of RV reducer fault diagnosis model, Test and data analysis, and Conclusion were supported by the data . The paper is relevant and interesting and Schematic of controller was done with no errors.It seems to be original , according to other results from current literature mention on the proposed manuscripit . It deserves publication after revisions: (Refer to the red font part of the revised manuscript for the specific revisions)
Point 1: The Numerical simulations must show the efficiency of the control methods, as well as the sensitivity of each control strategy to parametric errors.
Response 1: Thanks to the experts for this question. I quite agree with experts. For the effectiveness of control methods, besides the common model evaluation indexes, I also added the diagnostic accuracy and calculation speed data of each control method for comparison, so as to more comprehensively prove the performance superiority of the RV reducer fault diagnosis model proposed in this paper based on EEMD-MPA-KELM. See the red part of the revised manuscript for details.
Point 2: The state of the art is no complete (see paper published on Mechanical Systems and Signal Processing journal, MDPI, and others)
Response 2: Thanks to the experts for this question. Based on expert advice, I have read the papers published on Mechanical Systems and Signal Processing journal and MDPI, and I know that the state of the art is no complete. As for the problem that the RV reducer fault diagnosis model based on EEMD-MPA-KELM proposed in this paper is not fully applied, at present, our laboratory has started to carry out tests for different types of retarders to expand the universality of the model proposed in this paper. However, as the test takes a long time, it is too late to supplement this time, and this part will be supplemented at the end of the later test.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
The manuscript could be published.
Author Response
Thanks to the experts for this question.As for the areas to be improved in the paper, my team and I will continue to study and try to improve the paper as soon as possible. Finally, thank you again for your recognition.
Reviewer 3 Report
The authors have addressed all the comments.
Author Response
Thanks to the experts for this question.As for the areas to be improved in the paper, my team and I will continue to study and try to improve the paper as soon as possible. Finally, thank you again for your recognition.
Reviewer 4 Report
The Background of the problem was done with success.
The proposed problem is clearly presented
The obtained results and Discussions were written with no mistakes.
The concluding remarks were supported by the data.
The work is easily read
The paper is acceptable as is and recommend publication because the authors satisfactorily answered the questions posed
Author Response
Thank the experts for their recognition and love of this manuscript. As for the areas to be improved in the paper, my team and I will continue to conduct in-depth research and try to improve the paper as soon as possible. Finally, thank you again for your recognition.
