Gradient-Based Aero-Stealth Optimization of a Simplified Aircraft
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe submitted article introduces a methodology for gradient-based aero-stealth optimization problems. It couples a RANS-based aerodynamic solver and a PO-based stealth solver. A wing platform problem for a simplified aircraft is provided to demonstrate the proposed method with 14 variables. The improvements are apparent and prove that the proposed method could successfully find better designs.
However, there are still several parts that might require further explanation or discussion to clarify the topic for readers, which I listed below:
[1] Introduction: From the literature review point of view, the discussion in this part might be insufficient. What are the challenges and limitations of current approaches? What are the motivations/reasons and advantages of the proposed method? Merely claiming to introduce a method without elucidating its benefits may fall short of meeting the criteria for a comprehensive article.
[2] Page 2: I fully agree that “In an industrial design process, gradient-free algorithms are typically used for preliminary design, with low-fidelity models, while gradient-based methods are more adapted to large-scale aircraft shape design with mid to high-fidelity models.”
However, in the current work, the proposed method, which claims to be an adjoint-based optimization method, is only tested with a problem with no more than 20 variables. This might be insufficient to demonstrate the value of the proposed method. Please consider large-scale aircraft shape design problems if possible.
[3] In Paragraph 1st of Introduction 1.2, “…the coupling of the NSGA-II genetic algorithm and the multi-objective efficient global optimization approached…” and “…used hierarchical kriging models and surrogate models for aero-stealth optimization …”. They are repeated statements. NSGA-II is a multi-objective algorithm. Kriging model is one of the surrogate models.
Besides, this paragraph should consist of several sentences with the end of “.”, not “,”.
[4] Chapter 3 - Validation on Page 8 has only one section. It might not need one extra subtitle, “Validation of the COPOLA code”. Besides, in this part, one should mention that sensitivity validation is included in the latter part.
[5] Chapter 3 - Validation on Page 8: The validation of the aerodynamic solver might be necessary.
[6] Section 4.1 on Page 9: “… too small a step size can lead to a bad gradient, so we have to find an optimal step size for each design variable.” Would you mind illustrating this step size more? What is the final value used in this work? And how is it from?
[7] Section 4.2 on Page 10: The basic geometry parameters, such as reference area and reference chord, used by aerodynamic analysis should be included to help the understanding of the aerodynamic coefficients.
[8] Section 4.4 on Page 13: For comprehensive understanding, the total number of design variables could be listed.
[9] Figure 12 on Page 15: This figure has given for various values of τ, including 0.5, 1, 5, and 10. However, the chosen one τ = 0.1 has not been included. It is better to add this one to the figure, either.
[10] Line 376 on Page 15: what is “the DF gradients”? “FD”? The same issue also appears in Figure 14.
[11] Figure 14 on Page 17: it isn't easy to understand this figure. Which one is “DF 1”, “forward”? Which one is “DF 2”, “centered”? Besides, consider using “the index of variables” to replace “the number of variables”. The latter one is usually used to illustrate the total number of variables.
[12] Page 19: It is unsuitable to skip Figure 18-22 and directly illustrate Figure 23-24. The sequence of figures should be in the same order as the statements.
[13] Section 7.1 on Page 17: It is desirable to demonstrate the comparison with aerodynamic coefficients, like Cd, Cl and Cm, which is the most direct way to show the improvement. Besides, the optimization history, like the convergence history of the design objective, is also desirable for work related to optimizations. It is a good evidence to demonstrate your method. What’s more, the number of iterations, the number of function evaluations and the number of gradient evaluations, are both important indicators of optimizations.
[14] Section 7.2 on Page 19: It has the same issue as Section 7.1. Please have a check of the former comment.
[15] Section 7.3 on Page 21: Table 1 should not be put on Page 17. It should be at the bottom of the first paragraph of Section 7.3. Besides, it has the same issue as Section 7.1. Please have a check of the former comment.
[16] The positions of Table 2 and Table 3 are pretty strange. Are there any statements about these tables? I haven’t found them.
[17] Conclusion on Page 27: “…For the sake of keeping computation time low, we showcased our method on a simplified shape with fifteen parameters…” From Section 4.4, we get 13 global variables and 1 AoA variable. What is the 15th variable?
The English language should be revised to improve the quality of the paper within the journal environment. There were phrases that, as a non-native English speaker, I found unsuitable. Below are a few instances that should be modified.
[1] Line 64 on Page 2: “…the multi-objective efficient global optimization approached…” “approached”?
[2] Line 76 on Page 2: “… trade-off of teh aero and stealth criteria” What is “teh”?
[3] Line 105 on Page4: “…Physical Optics (PO)… ” The authors give the explanation of PO here, but not in the Introduction. This should be explained at the first position of the whole manuscript. Please do have a check for the other symbols.
[4] Line 408 on Page 19: “aero optimal shape”…a more suitable word should be “optimized aerodynamic shape”
[5] Line 469 on Page 27: “After developing and validating the OP solver…” What is “the OP solver”? “PO”?
Comments on the Quality of English LanguageThe English language should be revised to improve the quality of the paper within the journal environment. There were phrases that, as a non-native English speaker, I found unsuitable. Below are a few instances that should be modified.
[1] Line 64 on Page 2: “…the multi-objective efficient global optimization approached…” “approached”?
[2] Line 76 on Page 2: “… trade-off of teh aero and stealth criteria” What is “teh”?
[3] Line 105 on Page4: “…Physical Optics (PO)… ” The authors give the explanation of PO here, but not in the Introduction. This should be explained at the first position of the whole manuscript. Please do have a check for the other symbols.
[4] Line 408 on Page 19: “aero optimal shape”…a more suitable word should be “optimized aerodynamic shape”
[5] Line 469 on Page 27: “After developing and validating the OP solver…” What is “the OP solver”? “PO”?
Author Response
Please see attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsPlease find review attached.
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for addressing the comments. I recommend accepting the manuscript.
Comments on the Quality of English LanguageNo comments on quality of English language.
