Failure Threshold Determination of Rolling Element Bearings Using Vibration Fluctuation and Failure Modes
Round 1
Reviewer 1 Report
A method was investigated and proposed to optimize the FT of REBs from the measured vibration data, and the performance and accuracy of the proposed method were estimated by an accelerated life test dataset and an industrial dataset. The structure of the manuscript is well organized logically, and the description is clear. It can be published by improving the quality of the manuscript according to the following points:
- The quality of the figures in the manuscript need to be improved, they need to be modified more clearly and readable. Especially for Figure 5, 7, 9.
- For the industrial data, Model 1 was used to estimate the FT, why? And what is the differences between the three models? And what is the limitation of industrial data that may affect the performance of the models?
Author Response
Dear Reviewer 1
Thank you very much for reviewing our manuscript and your valuable comments. We have added some more description in the new revision of the manuscript regarding the comments. Please find our explanation to your points below:
1) The quality of the figures in the manuscript need to be improved, they need to be modified more clearly and readable. Especially for Figure 5, 7, 9.
Response:
The quality of figures in our original submission were high enough. However, the automatic converted file by the submission system had reduced the quality of the figures. You could check the docx files for original figure please. In addition, in the new submission, we have sent our own made pdf file too.
2) For the industrial data, Model 1 was used to estimate the FT, why? And what is the differences between the three models? And what is the limitation of industrial data that may affect the performance of the models?
Response:
About the reason for employing Model 1 to estimate the FT of industrial data, we have added more explanation (highlighted in yellow) in the text (Please check: Section 4.2, Paragraph 1).
About the difference between three models: It was described in the first paragraph of section 4.1. In order to clarify the difference between these models, more explanation was added in the second paragraph of section 4.1 (highlighted in yellow).
About the limitation of industrial cases, there was a short description in the last paragraph of section 5.2. another limitation was added in the new revision of the manuscript (last paragraph of section 5.2).
Author Response File: 
Author Response.docx
Reviewer 2 Report
The paper presents a study 15 features with PCA algorithm in relation with REBs. In most studies the FT is presented as a constant value and the paper research demonstrate the FT influence in bearing condition monitoring. Authors propose a FT based on a probability distribution, not constant. The results support the proposed FT.
I encourage authors to include in Conclusions section next expectations about their research: new application fields, next steps…
Minor points for improving the paper:
- Line 42, text “FT may be defined by three different approaches”. A reference is required.
- Figure 1, improve the quality of the image.
- Line 156-157, text: “some standard codes (e.g., ISO 10816-3 [3])”. I suggest, “some standard codes [3] ”
- Figures from 3 to 13, improve the resolution of the images.
Author Response
Dear Reviewer 2
Thank you for reviewing our manuscript and your comments. We have updated manuscript regarding the comments. Regarding your valuable advice, we have added an application and our future study plan to the last paragraph of conclusion (Highlighted in yellow). Please find the point by point explanation of us for the other comments, below:
1) Line 42, text “FT may be defined by three different approaches”. A reference is required.
Response:
The presented classification is ours. So, we have not reference for this sentence. However, to prevent misunderstanding, we have changed your mentioned sentence to show that this classification is not from another reference.
2) Figure 1, improve the quality of the image.
Response:
The quality of figures in our original submission were high enough. However, the automatic converted file by the submission system had reduced the quality of the figures. You could check the docx files for original figure please. In addition, in the new submission, we have sent our own made pdf file too.
3) Line 156-157, text: “some standard codes (e.g., ISO 10816-3 [3])”. I suggest, “some standard codes [3]”.
Response:
It is done. Thanks.
4) Figures from 3 to 13, improve the resolution of the images.
Response:
The reason was described in the response of comment 2.
Author Response File: Author Response.docx
