Bearing Fault Diagnosis Based on Stochastic Resonance and Improved Whale Optimization Algorithm

Round 1

Reviewer 1 Report

Title: Bearing Fault Diagnosis Based on Stochastic Resonance and Improved Whale Optimization Algorithm

In this manuscript by Huang et al., a parameter adaptive stochastic resonance strategy based on an improved whale optimization algorithm (WOA) is proposed to effectively detect the bearing fault signal of rotating machinery. Overall, the manuscript is interesting however, various issues need to be addressed, which are as follows:

1.      The language of the article needs some revision and should be reviewed by a native language expert.

2.      The abstract needs to include some background of the problem as well. Also, results need to be mentioned quantitatively in the abstract.

3.      There is no description of what kind of data was used for whole experiments. Authors just refer to the data as signals, whereas it is not explicitly mentioned how these signals were generated.

4.      The problem statement is not well formulated in the introduction section.

5.      The organization of the manuscript is ambiguous. The authors should have explained their dataset collection method, dataset features and all relevant details related to their model, however, in the current state, the manuscript has nothing related to this.

 

6.      The results need to be explained more. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors of the paper studied bearing fault diagnosis problem and propose an interesting method based on modified whale optimization algorithm. There are several issues and questions that need to be fully addressed in a revision.

1. The abstract is suggested to major revise. There are many existing bearing fault diagnosis methods. Why proposing a new one? The research gap and motivation of this work are unclear.

2.In the model in section 2, gravity is ignored. Will that cause a problem to precision and other key performances?

3. U(x) is modeled by formula (2). Why in such a form? This part is a bit confusing. Have you carried out prior tests to see whether it aligns well with real circumstance?

4. Figure 2 is unsuitable and seems unprofessional to be included in such a scientific paper. It is confusing how it can interpret "humpback whales swim spirally to their prey and shrink the enclosure at the same time". 

5. In the modified method, how did you prove "accelerates the convergence speed of the algorithm"? How did you prove "improves the stability"? More rigorous quantitative analysis is expected.

6. More alternative solutions to fault diagnosis can be discussed, as in Optimized design of parity relation based residual generator for fault detection data-driven approaches.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed all the comments.

Reviewer 2 Report

The paper has been improved based on the reviewers comments.