A Reliability-Based Robust Design Optimization Method for Rolling Bearing Fatigue under Cyclic Load Spectrum

Round 1

Reviewer 1 Report

Respected sir

The paper bearing Ms. Ref. No.:  mathematics-2446376 “A Reliability-Based Robust Design Optimization Method for Rolling Bearings Fatigue under Cyclic Load Spectrum” is well organized and represents a reliability based robust design method that overcomes the above problems for rolling bearing fatigue failure. A fatigue time-varying reliability model based on the fourth-order moment method has been developed and the reliability-based robust design index was proposed. Finally, the fatigue reliability-based robust design of rolling bearing is achieved by multi-objective optimization. In addition, the paper is well written with proper explanation. But some modifications are required before acceptance. The authors should address the following concerns:

 

1.      Motivations and novelties are not spelled out properly. The contributions and objectives are to be justified properly.

2.      Abstract is to be rewritten along with the major contribution.

3.      The gap analysis is to be provided clearly.

4.      How the β value is decided in the Kriging model.

5.      The fatigue failure of bearings is supported mostly by Weibull distribution then why a Gaussian distribution is adopted in Kriging model.

6.      Is Kwofie et al. [11] theory of modified Miner rule is valid for loading condition of bearing bearings. Miner rule is applied to bearings for fatigue life predictions.

7.      How the constraints of the optimization are decided. How the limiting constraints ranges are decided and how these constraints are handled. Plz. Provide constraint handling methods.

8.      Why first fourth-order moments is adopted in this formulation for variables?

9.      Provide the implementation steps for the proposed design. Explain why it is different from others?

1How the sample size in MCS method is decided?

1Why the reliability is decreasing after 1000-1100 hrs of operations.

1 There are some typos and grammar related issues, so try to correct it. Provide some more recent references.

1 There is marginal difference in values of optimized design parameters of rolling bearing before and after optimization, which does not justify the used of computing tools like GA etc. Explain why there is marginal difference. How the bound values of input design parameters are selected.

It needs minor corrections

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The subject of the reliability of bearings is important and up-to-date, because damage to these elements can result in large losses in the industry. Therefore, it is good that the authors undertook research on this issue.
However, some issues require clarification. 

1. In the introduction, the authors write about 6 sections of the article (lines 90 - 93) - and only 4 sections are present.

2. The authors in section 1 write about the curve obtained from the test showing the relationship between the load factor and the fatigue life under the present material (lines 37 - 42). It would be worth presenting this curve in the article.

3. Figure 7 is unnecessary - the data presented in it are in Table 4. Moreover, when one value is many times smaller than the other three, it is not visible in the figure - which is unprofessional.

4. Figure 6 shows a comparison between the method proposed by the authors and the MCS method. So, what is the advantage of the developed method over the MCS method if the error is so small? Does this comparison serve only to confirm the effectiveness of the method?

5. It is worth improving the "Discussion" section - so as to clearly present the advantages of the developed method in relation to other methods.

There are several punctuation mistakes in the article that should be corrected.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf