A Hierarchical Decision Fusion Diagnosis Method for Rolling Bearings
Round 1
Reviewer 1 Report
The paper is sufficiently clear, but I have some comments that should be addressed before the paper publication:
- Please indicate where the vibration sensor was installed. What kind of sensor was it, acceleration, velocity, or displacement?
- What type of rolling bearing has been tested?
- Figure 8,9,12-a,12-b,12-c,13. - Please include in the figures captions which parameters correspond to the appropriate numbers.
- What exactly was the failure of individual parts of the bearing. Pictures of these lesions would be useful.
- Line 455 - What kind of load was given, sinusoidal, random…?
Author Response
Dear review
Thank you for your useful comments and suggestions on the manuscript. We have modified the manuscript accordingly and the corrections have been labeled in red color font. The detailed responses to your comments are listed below point by point:
1) Question: Please indicate where the vibration sensor was installed. What kind of sensor was it, acceleration, velocity, or displacement?
Response: We have explained the type and installation position of the accelerometer in the first paragraph of Section 3.1：Vibration data was collected by using accelerometers, which were installed at the 12 o’clock position at fan end of the motor housing.
2) Question: What type of rolling bearing has been tested?
Response: The type of rolling bearing is SKF6205.
3) Question: Figure 8,9,12-a,12-b,12-c,13. - Please include in the figures captions which parameters correspond to the appropriate numbers.
Response: The parameters correspond to the appropriate numbers in Figure 8,9,12-a,12-b,12-c,13 have been include in the figures captions in the revised manuscript.
4) Question: What exactly was the failure of individual parts of the bearing. Pictures of these lesions would be useful.
Response: The single point faults in the inner raceway, the outer raceway and ball were introduced to the test bearings using electro-discharge machining with fault diameters of 0.007 inch, 0.014 inch, 0.021 inch. We add a bearing structure diagram to help readers know the location of fault in the revised manuscript.
5) Question: Line 455 - What kind of load was given, sinusoidal, random…?
Response: Three constant loads HP1, HP2 and HP3, which are 1 horsepower, 2 horsepower, and 3 horsepower respectively, were given.
Special thanks to you for your good comments. and the responses to your comment, please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
I do not have any remarks.
The article is interesting. The authors of the paper proposed their own hierarchical method of diagnostics of rolling bearings. An interesting solution is the use of an artificial neural network with back propagation including a fault detection layer, a fault isolation layer and a fault identification layer. The methodology of hierarchical diagnostics of bearing damage allows for an accurate assessment of the working condition of the bearing, determining the degree of place and type of its damage, and the cost of the calculations is low. It is worth emphasizing that the research results were verified on the basis of data of bearings existing in the West Reserve University databases.
Author Response
Dear review
Thank you for your positive comments and encouragements.
Reviewer 3 Report
This is a very good paper in general, that presents an interesting approach of fault diagnosis. I find interesting the 3rd stage of the diagnosis architecture, i.e. the fault degree estimation layer, which is uncommon in a diagnosis scheme.
After a short introduction, the authors go straight into presenting their method. I think authors should add 1-2 short paragraphs (in Introduction if you do not want a separate section) about previous works on fault diagnosis of rolling bearings and associated methods used in this paper.
I do not make much sense of first paragraph in Section 2.2 and Figure 2. I think they are not presented well and it needs attention. Also check all equations in Section 2.2.
Not clear why Section 3 (Hierarchical fault diagnosis ..) is not Section 2.4, as it is part of the Method, which us Section 2.
Language should be checked in places, and typos can be found in the paper. They need to be fixed.
With these minor changes the paper can be accepted and may be published as it presents a reasonable approach and quite extensive results.
Author Response
Dear review,
Thank you for your useful comments and suggestions on the manuscript. We have modified the manuscript accordingly and the corrections have been labeled in red color font. The detailed responses to your comments are listed below point by point:
Point 1: After a short introduction, the authors go straight into presenting their method. I think authors should add 1-2 short paragraphs (in Introduction if you do not want a separate section) about previous works on fault diagnosis of rolling bearings and associated methods used in this paper.
Response 1: According to the reviewer’s suggestions, we have modified the first paragraph and the third paragraph in introduction to analyze the previous works on fault diagnosis of rolling bearings and associated methods used in this paper.
Point 2: I do not make much sense of first paragraph in Section 2.2 and Figure 2. I think they are not presented well and it needs attention. Also check all equations in Section 2.2.
Response 2: According to the reviewer’s suggestions, we have revised the first paragraph in Section 2.2 and Figure2. I am very sorry that there are some errors about the BPNNs in the manuscript. These mistakes have been corrected in the revised manuscript.
Point 3: Not clear why Section 3 (Hierarchical fault diagnosis ..) is not Section 2.4, as it is part of the Method, which us Section 2.
Response 3: According to the reviewer’s suggestions, We've revised the structure of the manuscript and adjusted the section 3 to section 2.4.
Point 4: Language should be checked in places, and typos can be found in the paper. They need to be fixed.
Response 4：I am very sorry that there are some text errors and grammar errors in the manuscript. These mistakes have been corrected in the revised manuscript.
Special thanks to you for your good comments, and the responses to your comment, please see the attachment.
Author Response File: Author Response.docx
