Effects of Static Stability Margin on Aerodynamic Design Optimization of Truss-Braced Wing Aircraft
Round 1
Reviewer 1 Report
The authors employed multi-disciplinary optimization (MDO) to numerically investigate the aerodynamic benefits of the truss-braced-wing (TBW) configuration. In particular, the authors applied relaxed static stability (RSS) to increase the useful lift, thus reducing the associated drag and overall structural weight of the aircraft. The merit of the paper lies in the thorough analysis of results, especially regarding the trends for variations in CG location and off-design results.
In order to make the paper more complete, the authors should address the following points. First, it should be noted that candidate configurations for future airliners, such as the TBW, will surely rely on MDO considering the propulsive characteristics. However, there is no mention of a propulsive system in the paper. While it is understandable that this aspect was not investigated in this paper for simplification, the authors should probably acknowledge that it is very relevant to this kind of configuration.
Second, it should be noted that RSS is already utilized on most of the new designs of fly-by-wire airliners. As a result, these aircraft require an active control system to exhibit nominal flying qualities, which deteriorate to alternate or direct modes in case of serious failures. Hence, although the open-loop stability is relaxed, it is still positive. Perhaps the authors could add this discussion in the introduction.
As a suggestion for future work, the authors could investigate another relevant aspect for future configurations: the active load alleviation capability, which reduces the design weight loads that ultimately size the structure.
Comments and suggestions to improve the paper are given below.
Line 253:
"The more front aerodynamic center..."
more forward aerodynamic center
Line 255:
"...friction drag like B-52 CCV, or with a canard applied in the fighter aircraft."
Can the authors provide references for these cases?
In academic writing, it is advisable to avoid starting sentences with coordinating conjunctions like "and" and "so". The authors tend to start sentences with "And". This happened in lines 28, 29, 46, 84, 106, 109, 163, 292, 300, 311, 384, 400, 479, 486, 496, 506, 550, 598. Most of these instances can simply be suppressed.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The study presents a study that aims to evaluate the importance of taking into account the stability margin during aerodynamic shape optimization for cruise drag reduction of an innovative configuration. The study itself is carefully setup and conclusive. Nevertheless, there are three major points, to be addressed for an iproved manuscript for publication:
(i) Although it is the aim of the study, in the conclusion it fails to clearly answer whether it is important to use the relaxed stability margin (RSS) during optimization. What is shown that there is an impact of RSS on drag (even before optimitzaion) and that optimization finds a better solution. But it is not evaluated that taking RSS into account during optimization is crucial (although it is clear that the data to analyse this is available). It is left unclear, what is the effect of taking RSS into account for each optimization or if the improvements based on RSS are independent on the shape optimization. This should be worked out more clearly. It would be most easily done by comparing the geometric results of the optimiztaion - how large is the difference between the different optima - or more indirectly, what would be the improvement of the geometry optimized at 5% CG but simulated at 35% CG - how does this differ from the optimizaion result directly at 35% CG. Only this analysis would clarify the importance for RSS during optimization.
(ii) It is a shortcoming that only for the last 35% CG optimized configuration the analysis is extended to L/D and M x L/D on the fine (L1) mesh. On the one hand, the results reported on L2 show much higher CD values due to the low grid resolution. But for an aircarft designer, such high values read that this configuraiton is out of any discussion (compared to 250-300dcts for a classical tube & wing and here side effects like engine integration and fairingas are not yet taken into account). It is therefore strongly advised to highlight this more clearly and to complete the evaluation of aerodynamic efficiency for the other configruations on the fine L1 mesh, too - at least for the baseline and for the baseline at 35% CG to get a feeling for the improvement.
(iii) The images (Figs 6-9, 11-13) are very busy but explanation is only few. It is acknowledged that the author may directly see the differences of the various configurations, but for the reader it is hard to detect any differences depending on the CG position without getting more explanation where to look at in detail. In addition, the large amount of sub diagrams make the texts very small up to unreadable. It is therefore advised to split up the images: one for the top views, one for the spanwise distributions and one for the pressure distributions. The front view can be omitted. It only provides a significant information when comparing the baseline and the optimized. This could be done separately in one figure. The sectional pressure contours are similarily only of minor importance and best to compare a baseline to an optimized case to pin-point on the achievements of the optimiztaion. Or limit on most forward to most backward CG to highlight in larger images on the tendencies (as the intermediates are pretty linearly inbetween). In the pressure graphs of the optimized configuration, there seems to be only one geometry plotted - or are they all exactly on top of each other? This would be an important information, as otherwise it may be assumed that the optimization result (in shape) is independent of the CG position and thus on taking into account RSS (see comment (i) ). On the other hand the tables describing the cases are repeating a lot of information and can be shortened (e.g. the combination of M and Re must not be repeated for every case). Streamlinig the tables would give space for improved figures without extending the length of the manuscript.
Further specific remarks:
page 1: reference [1] points to Chinese source for market forecast only, but the text mentions estimates from Boeng and Airbus - please provide references here
page 2 first paragraph: There might be better references than references [2-4] for the statements. At least, if searching for such information, one would not have looked for the given references as an information source.
page 2, line 54-56: Statement on NASA N+4 configuration. Provide reference directly in first sentence instead of at the end of the paragraph. Otherwise EIS 2025-2035 and N+4 don't really match if it is not clear from the beginning that the NASA study is a decade old.
page 5, line 200: The chosen references [40-44,46] suggests FFD to be a new development. In contrast, the technique can be traced back to 1986 (Sederberg & Parry) (later used as ref 47 - but should be put prominentely here) with first applications to aerodynamic shape design by Ronzheimer (2002) and Samareh (2004):
Ronzheimer, A., “Post-Parameterization of Complex CAD-Based Aircraft-Shapes Using Freeform Deformation,” 8th International Conference on Numerical Grid Generation in Computational Field Simulations, June 2–6, 2002, Honolulu, Hawaii, USA
Samareh, J.A., "Aerodynamic Shape Optimization Based on Free-form Deformation", 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, 30 August - 1 September 2004, Albany, New York
page 6: Notations of e.q. (1) and (2) must be more specific/exact. In eq. (1) it should be the neutral point x_n where dCm/dCl=0, not the aerodynamic center where Cm=0 . The latter needs to be at xCG for a trimmed aircraft in steady flight. Further, the notation of eq. (1) can be misleading if the condition in this sentence is overlooked that the coordinates must be normalized by the MAC (which is even not done for the image in Fig. 2). It would be recommended to reformulate eq. (1) as ususal in flight mechanics textbooks to
K_n = (x_n - x_CG) / MAC
It doesn't change the meaning but is more universal.
For the evaluation of C_m_alpha it is important to specify the reference center for the pitching moment - which should be the CG in this case.
It is further assumed that the K_n is evaluated by evaluating eq.2 but this needs either the derivative calculation (adjoint?) or a finfite difference. eq.1 woul need something similar. Please specifically report what is used here. The statement found in section 4.1 (page 11) should be placed here.
page 6, line 253: Typo: The aircraft's name is "Concorde" (with an 'e' at the end) - it's a French name.
page 7, line 271: Why did you use O-H-type grids and why not a C-H-type grid? in block structured, the effort is similar and the accuracy in viscous flow genereally better due to better capturing the wakre flow (especially when using trimmed configurations with tails).
page 7: description of Study 1 lacks description of method to get the trimmed state. In study 2, the trim angle of the tail is a design parameter and trimming is obtained by variation during optimization. But for the baseline, there is a statement missing.
page 7, line 300: Grammar: Remove "And " at beginning of sentence
page 8, line 319: Typo: Lower case 'b' in "both Fedaral Aviation ...."
page 13, second paragraph: In conclusion, the explanation for drag reduction is twofold:
a) due to rear CG , the lift is shiffted to the tail, and by this, the load reduction on the wing/truss enables a reduction of shock streght.
b) shifting load to the tail reduces the trim drag by reduing the induced drag of the tail due to less downforce.
This is not quite unexpected and a key reason for reducing the stability margin. Probabaly worth to streamline the discussion here.
page 17, lines 473+474: incomplete sentence
page 21, line 541: Grammar: "in Table" instead of "as Table"
page 21, last paragraph: in this discussion, it is left out that - with p x M^2 x CL proportional to the aircraft weight - these conditions are either a funciton of the aircraft mass or the flight altitude. So what is the rationale for these selected conditions in terms of aircraft design?
page 24, line 635: I the text, the Mach number is noted as "M" (not "Ma"). Both san be seen in literature. Although, in English sources M is preferred.
page 24: references format: Please verify with publishers preferences. If Journals have numbers in addition to volumes, they should be noted. If conference papers have paper-ID's (as e.g. for AIAA) they should be noted. Publishing years printed in bold or not.
page 24, line 666: Provide publisher for CleanSky Book.
The written language is understandable - at least for non-native English speakers. Although, throughout the text, some improvement can be achieved regarding grammar and incomplete sentences. It is adviced to perform proof-reading by an experienced or native English speaker.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Thank you for thoroughly implementing the review suggestions. The paper has significantly imporved in these aspects.
Please only check your cross references - there are some "Error! Reference source not found"
And, in nomenclature, there is still x_AC instead of x_n for eq. 1
Aside this, for me o.k. to be accepted
although it is clear that the text is written by a non-natvie speaker, the text is well understandable. If there is some effort in re-reading anyhow, it would be worth to look out for grammar glitches - but not critical.
Author Response
Thanks a lot for the reviewer’s comment.
Firstly, we have checked and corrected all the cross references.
Secondly, we have updated the expression of xn and xCG in nomenclature.
All these changes are marked with red color in the new uploaded version manuscript.
