Global Mechanical Response Sensing of Corrugated Compensators Based on Digital Twins
Round 1
Reviewer 1 Report
In my opinion, visual presentation of valules in Table 2 should be improved. It is really hard for the reader to compare values in present form.
Page 9: "As shown in Fig. 8, the..." - There is no Fig 8. Did you mean Fig. 7? Same paragraph: The authors claim that "...DT model... showed satisfactory consistency with the FEA calculation results, which further illustrates the reliability of the predicted performance...". And the conclusion states "... DT has excellent predictive performance." I'd advise the authors to add a comment about the DT model predictive performance with respect to stress values in the conclusion, so that the readers can grasp a better overview of the obtained results.
Author Response
Response to Reviewer 1
- In my opinion, visual presentation of values in Table 2 should be improved. It is really hard for the reader to compare values in present form.
Response: The data of Table 2 are plotted in Fig. 5, where the red dots are R2, the blue triangles are CVavg.
- Page 9: "As shown in Fig. 8, the..." - There is no Fig 8. Did you mean Fig. 7?
Response: Thank you very much for your warm reminding. We have corrected this error.
- The authors claim that "...DT model... showed satisfactory consistency with the FEA calculation results, which further illustrates the reliability of the predicted performance...". And the conclusion states "... DT has excellent predictive performance." I'd advise the authors to add a comment about the DT model predictive performance with respect to stress values in the conclusion, so that the readers can grasp a better overview of the obtained results.
Response: Thank you very much for your reminding, we have supplemented the conclusion.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The technical content of the paper is good enough. Research is clear and well-balanced but it should be improved before publication:
1. Please remove acronyms from the abstract (DT and FEA) and put them in the main body of the paper. In the abstract be as much general as possible and as much readable as possible.
2. Authors use an element type S4 that is a fully integrated, general-purpose, finite-membrane-strain shell element available in ABAQUS. Excellent choice. May I suggest to add a recent reference to support this selection and help the reader to go deep inside the question: DOI 10.1515/cls-2022-0027
3. The “Fatigue crack” that authors emphasize is a low cycle or a high cycle fatigue crack ?
4. May I suggest to add a reference in the text (page 3) to both MATLAB and ABAQUS. And then to add this reference in the bibliographic section at the end of the article.
5. Is there a strong motivation to assume “where m(x) is often assumed to be 0.” Equal to zero ?
6. I suggest to improve quality of Figure 4.
Author Response
Response to Reviewer 2
- Please remove acronyms from the abstract (DT and FEA) and put them in the main body of the paper. In the abstract be as much general as possible and as much readable as possible.
Response: Thank you very much for your reminding, we have improved the abstract.
- Authors use an element type S4 that is a fully integrated, general-purpose, finite-membrane-strain shell element available in ABAQUS. Excellent choice. May I suggest to add a recent reference to support this selection and help the reader to go deep inside the question: DOI 10.1515/cls-2022-0027
Response: Thank you very much for your reminding, we have added this reference.
- The “Fatigue crack” that authors emphasize is a low cycle or a high cycle fatigue crack?
Response: Most of the fatigue cracks on corrugated compensators are high cycle fatigue cracks.
- May I suggest to add a reference in the text (page 3) to both MATLAB and ABAQUS. And then to add this reference in the bibliographic section at the end of the article.
Response: Thank you for your reminding, we had added references to the relevant MATLAB toolboxes.
- Is there a strong motivation to assume “where m(x) is often assumed to be 0?” Equal to zero?
Response: Yes, setting the mean value to 0 is a general way of Gaussian process regression.
- I suggest to improve quality of Figure 4.
Response: Thank you very much for your reminding, Figure 4 focuses on demonstrating the good space-filling property of the Latin hypercube sampling method.
Author Response File: Author Response.pdf
Reviewer 3 Report
COMMENTS TO THE AUTHOR(S)
This study examined the “Global mechanical response sensing of corrugated compensators based on digital twins”. Given that corrugated compensators are important components in the piping system, absorbing mechanical deformation flexibly, this paper is timely and could offer new insights in the mentioned area. The manuscript is generally well written and easy to understand. I suggest that the authors revise the manuscript incorporating the following comments and suggestions into an updated version.
· On what basis the authors selected the number of test samples is one-quarter of training samples.
· Why the authors targeted U-shape corrugations? Is it the most important one?
· On what basis the authors selected the value of ranges of the three displacements.
· On what basis the authors selected the geometric parameters and material properties of the unreinforced corrugated bellow.
· The recommended threshold CVavg (0.1) value much higher than computed values as obvious from Fig 5. Please clearify
Author Response
Response to Reviewer 3
- On what basis the authors selected the number of test samples is one-quarter of training samples.
Response: The number of training and test sets is usually set at eight to two or seven to three in machine learning.
- Why the authors targeted U-shape corrugations? Is it the most important one?
Response: Thank you very much for your reminding, the bellow with U-shape corrugations is most widely used in practical engineering.
- On what basis the authors selected the value of ranges of the three displacements.
Response: Thank you very much for your warm reminding. The value of ranges of deformation is related to the structural form of the pipe, the working condition of the corrugated compensator and other factors. The values in this paper are mainly based on the authors' engineering experience. The main purpose of this study is to verify the reasonableness and applicability of the technical framework of the digital twin by comparing the predicted values with the theoretical values obtained through finite element analysis.
- On what basis the authors selected the geometric parameters and material properties of the unreinforced corrugated bellow.
Response: The geometric parameters and material properties of the unreinforced corrugated bellow refer to Ref. 27.
- The recommended threshold CVavg (0.1) value much higher than computed values as obvious from Fig 5. Please clarify.
Response: As mentioned in the text, the recommended threshold of CVavg is application-specific but CVavg < 0.1 is typically considered as a criterion to prove the validity of the surrogate model, which is described in detail in Ref. 24.
Author Response File: Author Response.pdf
