Actuator Weak Fault Diagnosis in Autonomous Underwater Vehicle Based on Tri-Stable Stochastic Resonance

Round 1

Reviewer 1 Report

General Comments:

The paper is well written, although some minor proofreading may be necessary. The introduction could be expanded by adding details on how the work referenced is relevant to the article. The rest of the paper covers the tri-stable stochastic resonance theory, influencing factors and application. The conclusion could also be expanded, as it is very brief and does not capture everything that was covered in the paper.

General Revisions:

1. A lot of vague terminology (i.e. huge losses, great significance, good research, easy to operate, high requirements, etc.) is used in section 1 and section 2.3. Given that the paper is highly mathematical, the use of vague terms should minimized and replaced with measurements/statistics when possible.
2. Since this research is devoted to actuator weak fault diagnosis, it would be beneficial to explicitly define “weak faults” and how they are different from “strong faults”. This would also help to clarify the novel contribution, which is only partially introduced at the end of the introduction section: “Therefore, the method in this paper can adaptively…”
3. There is no transition between the introduction/literature review and the stochastic resonance theory section. It would be beneficial to include a paragraph at the end of the introduction that describes how the paper is organized and what is covered in each section.
4. The plots in figure 3 are very hard to read, thus I suggest making the image larger. In addition, the text in figure 3 a and c is cut out.
5. (minor) Figure 13 is shown before it’s referenced/described.
6. In the conclusion, the value range of a is not stated.
7. In the conclusion, the author claims that the proposed algorithm for weak fault detection is “better” than bi-stable system. The word “better” should really be quantified, especially in the conclusion.

Author Response

Firstly thank you editors and reviewers!

1.A lot of vague terminology (i.e. huge losses, great significance, good research, easy to operate, high requirements, etc.) is used in section 1 and section 2.3. Given that the paper is highly mathematical, the use of vague terms should minimized and replaced with measurements/statistics when possible.

I have changed it as recommended.

2.Since this research is devoted to actuator weak fault diagnosis, it would be beneficial to explicitly define “weak faults” and how they are different from “strong faults”. This would also help to clarify the novel contribution, which is only partially introduced at the end of the introduction section: “Therefore, the method in this paper can adaptively…”

The weak faults of actuators are usually early faults. If these early weak faults of actuators can be detected as early as possible, the strong faults of actuators that will eventually stop working will be avoided. The AUV can make corresponding judgment and treatment in time to avoid the occurrence of greater accidents.

3.There is no transition between the introduction/literature review and the stochastic resonance theory section. It would be beneficial to include a paragraph at the end of the introduction that describes how the paper is organized and what is covered in each section.

The second section is the related theory of Tri-stable stochastic resonance fault diagnosis, which is introduced from (1) AUV actuator dynamics model. (2) Principle of stochastic resonance. (3) Parameter compensation stochastic resonance. (4) Ant colony optimization algorithm principle. four aspects respectively. The third section is Influencing factors of multi-stationary stochastic resonance system. The fourth section is Practical engineering application. The fifth section are Conclusions.

4.The plots in figure 3 are very hard to read, thus I suggest making the image larger. In addition, the text in figure 3 a and c is cut out.

I have changed it as recommended.

5.(minor) Figure 13 is shown before it’s referenced/described.

I'm really sorry, picture 13 is Latex automatic typesetting.

6.In the conclusion, the value range of a is not stated.

I have changed it as recommended.

7.In the conclusion, the author claims that the proposed algorithm for weak fault detection is “better” than bi-stable system. The word “better” should really be quantified, especially in the conclusion.

I have changed it as recommended.

Reviewer 2 Report

The article represents very interesting scientific aspects regarding the problem of fault diagnosis of AUV actuators "Autonomous Underwater Vehicle" using Tri-stable Stochastic Resonance. Overall, the contents of the manuscript are clearly and precisely represented. We found a problem with the formatting of the bibliography in the introduction, i.e. the name of the author and the reference number are given. The conclusions should be better articulated by highlighting the advantages and limitations of Tri-stable Stochastic Resonance in ActuatorWeak Fault Diagnosis in AUV systems.

Author Response

Firstly thank you editors and reviewers!

I have changed it as recommended.

Reviewer 3 Report

The tri-stable stochastic resonance model is analyzed, and the ant colony tri-stable stochastic resonance model is used to diagnose the weak fault in the research. The overall article is well written, however the quality of the publication can be improved if following changes are made:

• The authors use alternatively small and capital letters in the abbreviations, consistency is required. E.g. abbreviations should be used. E.g.  autonomous underwater vehicle (AUV) is used in the abstract, while Recurrent Neural Networks (RNN) is used in the introduction section.
• Captions of Figure 3 (a) and (c) needs corrections.
• The heading Section 3.1 is followed by Section 3, while there is no other sub-heading e.g. Section 3.2 in the article. Kindly check.
• The quality of Figure 9 blur.
• It feels that the results are repeated in the conclusion, the authors should explain that what they have achieved from the research. Kindly revise the text.

Author Response

Firstly thank you editors and reviewers!

• The authors use alternatively small and capital letters in the abbreviations, consistency is required. E.g. abbreviations should be used. E.g.  autonomous underwater vehicle (AUV) is used in the abstract, while Recurrent Neural Networks (RNN) is used in the introduction section. 
• I have changed it as recommended.

• Captions of Figure 3 (a) and (c) needs corrections.
• I have changed it as recommended.
• The heading Section 3.1 is followed by Section 3, while there is no other sub-heading e.g. Section 3.2 in the article. Kindly check.
• I have changed it as recommended.

• The quality of Figure 9 blur.
• I have changed it as recommended.

• It feels that the results are repeated in the conclusion, the authors should explain that what they have achieved from the research. Kindly revise the text.
• I have changed it as recommended.