Incipient Fault Feature Extraction of Rolling Bearing Based on Signal Reconstruction
Round 1
Reviewer 1 Report
This study focuses on signal noise reduction, which is interesting and fits the scope of the journal. The author processed the fault signals by use of several methods, and achieved better results. However, this article still needs some revisions before its acceptance and the corresponding comments are given as below:
1. Many abbreviations and mathematical symbols have been used in the manuscript. For better understanding to the readers, a nomenclature presented in tabular form is suggested.
2. The content of this paper is inconsistent with the abstract. The article missed the parts of sparse decomposition and multi-source vibration separation. More work needs to be added.
3. In Figure 3, the flowchart of ICEEMDAN should show what the iteration conditions are.
4. This paper also suffers from grammar-related issues, which needs thorough proofread.
This paper also suffers from grammar-related issues, which needs thorough proofread.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Your article is generally clear and well-structured. However, there are a few points that require attention and revision before it can be accepted for publication.
The units indicated on the graphs must correspond to the SI system. Please check the spelling of the word “amplitude”. According to the figures, it is measured in volts, through it is not mentioned in the text. It is better to enclose the measurement value in brackets or separate it with a comma to avoid misinterpretation as division.
It would be better to add a grid to the figures to enhance the readability of the graphs.
Line 151: it would be beneficial to specify precisely how the amplitude value increases and what it signifies.
Eq. (1): There is no need to explain the format of expressing derivatives with respect to time, as this form is widely accepted. However, it would make sense to complete it with an explanation of what the function x(t) represents.
Eq. (5): not all components are explained. Eq.(11): consider replacing “log” with “lg”, “ln” or define it.
Fig (9): please indicate the primary components of the experimental device.
The analysis of fig.10 and fig.11 contains similar text. It would be helpful to either merge them or rephrase.
Conclusion N1 has a sentence that is excessively, it is advisable to break it down into more concise sentences for improved readability.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors proposed a novel incipient fault feature extraction method for rolling bearing. They first used Teager energy operator (TEO) to obtain the envelope of the impulse components in the vibration signal. Then, they used a PSO based Savitzky-Golay (S-Golay) to suppress the noise. Finally, they reconstructed the fault signal by multiplication of the filtered envelope signal and original one. Theoretical analysis and experimental results are presented in this paper. My concerns are listed in the following.
- In Line 89, Line 98, Line 102 and so on, the authors use Chinese comma.
- In Eq. 5, v(t+1) and x(t+1) should be in separate lines.
- Sometimes you use Fig., sometimes you use Figure. Please unify the representation.
- Please be careful about the English spelling, like “ `through” in Line 314 and 339.
- In Table 2, why the Outside diameter is listed as the first line and bold? You may need to name another header.
- You have compared your algorithm with others in correlation analysis. Why don’t you compare your algorithm with other algorithms in 5. Experimentation analysis?
Please double-check the English writing!
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
The paper is interesting because it concerns the study of bearing fault diagnosis methods that can be applied to rolling bearings. However some minor corrections are necessary.
1. In the introduction it should be clean explained what types of bearing defects can be detected and analysed using the proposed method.
2. There are many abbreviations in the article. The nomenclature list is recommended.
3. Experimentation analysis:
Please add more information about bearing defects that have been produced. Have the inner and outer rings been damaged?
4. Why SKF6006 bearings have been analysed?
5. Why only 600r/min rotary speed has been used in an examination? The limiting speed of SKF 6006 is 17 000 r/min.
6. Conclusion
The conclusion is too short. Please describe the limitations of the proposed method. Furthermore please add direction for future research.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
