Fault Diagnosis of Bearings Using Wavelet Packet Energy Spectrum and SSA-DBN

Round 1

Reviewer 1 Report

The article entitled Multi-fault diagnosis of bearing using wavelet packet energy spectrum and SSA-DBN is exciting and well-written. However, there are a few notes that may improve the quality of the article

Basic notes

1. The tests show that single failures were tested, not multi-damage cases as written in the title. The title should be corrected. I suggest eliminating the word "Multi" from the title

2. Experimental dataset: Vibration measurements - describe the sensor and apparatus used for the data collection shown in Fig.6

Other remarks

1. Please write in the summary about the SSA-DBN method (expand the abbreviation)

2. Line 16: "4 and 10 types.." - sounds vague, please rephrase

3. Line 101: expand the DBN abbreviation, the expansion is later on line 167

4. Line 322: "Bearings with EDM single point damage" - what is EDM?

5. Is the data normalized before entering the network?

6. Line 340: Please explain where the number 400 in the feature vector comes from.

7. Fig.10: Why are input data 280x256 feature vector?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors,

 

I found your article very interesting, but I suggest to introduce following remarks, which have to be discussed and corrected before publishing the paper:

 

1.       Firstly, please revise the font style in the whole manuscript, that it is inconsistent in many places. The same concerns typos, misusing of commas and so on.

2.       In the Abstract, please formulate the novelty and results obtained of your research. There is no place for the data treatment in the Abstract.

3.       In the Introduction I don’t see any references regarding the Hilbert Transform application in the diagnostics of rotating systems. I suggest to add 2 following papers to your list of references:

·       Extracting periodically repeating shocks in a gearbox from simultaneously occurring random vibration, MFPT 2018, 2018, pp. 456-464.

·       Improving the accuracy of fault frequency by means of local mean decomposition and ratio correction method for rolling bearing failure, Applied Sciences, 9(9), 1888.

4.       Figure 3, I presume, there should be Labeled Data.

5.       Please add the reference to the CWRU.

6.       Figure 8, what is the sampling frequency of the signal. In the present form it is very plain presenting no characteristic frequency.

7.       Please specify the characteristic frequencies for the bearing faults.

8.       Table 3, what the 0 refers to in Standard Deviation?

9.       In conclusions please refer to the results obtained. There is only a general description in it.

After improving above described issues in the paper I’d like to give my positive opinion on signing my review report.

 

 

The language requires some revision, however it is on quite good level. In the manuscript there is an inconsistence in the font style and multiple typos.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

After getting familiar with the response, I accept the paper in the present form.