A New Research Scheme for Full-Scale/Model Test Comparisons to Validate the Traditional Wind Tunnel Pressure Measurement Technique
Round 1
Reviewer 1 Report
First, I congratulate the authors for the interesting work with the novelty. However, some corrections and unclear information are found.
The abstract is clear and concise but too qualitative; quantitative information should be provided.
The introduction is OK but too short. Citing other work and mentioning like the following...[1-3], ...[4-7] etc. without further details of those cited papers look like descriptions of conference papers. This is a journal paper, thus details (although short) of those papers must be given.
Figure 6 requires further explanation of the not similar results of Iu=9.83% to others between 60 and 90 degrees. Furthermore, it is not clear with different turbulence intensities in Figure 6. If (say) 3 different turbulence intensity is checked, 3 curves should be provided. In fact, only 1 curve is shown.
The way to write decimal digits (between lines 238 and 247) should follow common sense. It is normally 3 digits. For instance, 0.49328e-03 should be typed as 0.493e-02. Further, the way to write them should be fixed and the following 0.493x10-2 is the correct one.
In lines 250 to 257, the authors said that the difference or error is within 10% and the authors said that those are accurate. I do not agree with the authors. The authors should make it into a maximum of 5% as most errors are accepted.
Figure 11 showed the model test without Re effects, in which the result is far from the others. However, it is not clear how big the Reynolds number is and why the difference occurs. A physical explanation is needed. A similar phenomenon is shown in Figure 12. But now, the model test without Re effects showed higher results. A physical explanation is required.
Citations within Conclusions should be avoided.
The appendix should be shown after the References.
Author Response
First, I congratulate the authors for the interesting work with the novelty. However, some corrections and unclear information are found.
The abstract is clear and concise but too qualitative; quantitative information should be provided.
Response: Thank you for the comment. Yes, the original abstract is too qualitative, and the result part (subsection 3.5) also lacks quantitative information. Therefore, we have improved these parts by including quantitative data (see the part highlighted in red) which can support our conclusions drawn in the abstract and the conclusion part.
The introduction is OK but too short. Citing other work and mentioning like the following...[1-3], ...[4-7] etc. without further details of those cited papers look like descriptions of conference papers. This is a journal paper, thus details (although short) of those papers must be given.
Response: Thank you for the suggestion. We have added the details of those citations in introduction (see the part highlighted in blue) and extended the length of introduction as much as we can.
Figure 6 requires further explanation of the not similar results of Iu=9.83% to others between 60 and 90 degrees. Furthermore, it is not clear with different turbulence intensities in Figure 6. If (say) 3 different turbulence intensity is checked, 3 curves should be provided. In fact, only 1 curve is shown.
Response: The region between 60 and 90 degrees is called the vortex separation zone, because vortexes are formed and separated near that position. The physical phenomenon of vortex shedding has been defined as being alternating vortex shedding from the cylinder and a clear “vortex trail” formed downstream. According to Niemann and Holscher (1990), the free-stream turbulence strongly influence the vortex shedding phenomenon with the flow past a circular cylinder with regard to the Strouhal number, the span-wise correlation of vortices and the mean wind pressure coefficients in the vortex separation zone. Since the free-stream turbulences are different between the case Iu=9.83% and other cases shown in Fig. 5, the mean wind pressure coefficients of the case Iu=9.83% are different from those of other cases in between 60 and 90 degrees. Theses are explained in the revised manuscript (see the part highlighted in green). Besides, the curve in Fig. 6 is the average values of the six curves presented in Fig. 5 for different turbulence intensities, and the error bars in Fig. 6 are variation ranges of the six curves presented in Fig. 5. Therefore, only one curve is shown in Fig. 6. I don’t know if I explain this clearly. Anyway, thank you for your good comment.
Reference:
Niemann HJ and Holscher N (1990) A review of recent experiments on the flow past circular cylinders. Journal of Wind Engineering & Industrial Aerodynamics 33(1): 197–209.
The way to write decimal digits (between lines 238 and 247) should follow common sense. It is normally 3 digits. For instance, 0.49328e-03 should be typed as 0.493e-02. Further, the way to write them should be fixed and the following 0.493x10-2 is the correct one.
Response: Thank you for your good suggestion. The data presented in between lines 238 and 247 have been changed to the correct form.
In lines 250 to 257, the authors said that the difference or error is within 10% and the authors said that those are accurate. I do not agree with the authors. The authors should make it into a maximum of 5% as most errors are accepted.
Response: The relative errors between our FE model and the prototype are all within ±5%. However, most relative errors between the aero-elastic model manufactured and the prototype are in between ±5% and ±10%. These discrepancies should be attributed to the model manufacturing error, the modal experiment error, and the environmental factors in physical tests, which can hardly be eliminated. We have tried to optimize the natural frequencies of the aero-elastic model by tuning the copper lead blocks arranged on it, but it is hard to reduce the relative errors to within ±5%. I am sorry the physical model test is finished and cannot be redone. All we can do now is to add the above explanations for readers to evaluate the accuracy of our model test (see the part highlighted in orange). We thank the reviewer for suggesting that the relative errors listed in Table 1 are not acceptable.
Figure 11 showed the model test without Re effects, in which the result is far from the others. However, it is not clear how big the Reynolds number is and why the difference occurs. A physical explanation is needed. A similar phenomenon is shown in Figure 12. But now, the model test without Re effects showed higher results. A physical explanation is required.
Response: Re for the full-scale scenario is generally greater than 1×107; however, that for the wind tunnel scenario is around 5×105. Basically, the flow separation location for flow around a bluff body without corners is sensitive to Re. Since the Re for model tests is usually two orders of magnitude smaller than the Re for the prototype flow event, significant distortions might be caused for the model test data. These are explained in the revised manuscript (see the part highlighted in bold).
Citations within Conclusions should be avoided.
Response: We have removed the citations from the conclusion part.
The appendix should be shown after the References.
Response: The Appendix is put after the References.
Reviewer 2 Report
This paper is acceptable but minor revision is required.
2-clarify the aim of paper in introduction[d1] .
3-check the data units in all tables.
4- describe about flow characteristic effects
5- What’s the effect of Reynolds number effects on pressure measurement model tests
6- describe about model surface roughness and placing the actively controlled devices
Author Response
This paper is acceptable but minor revision is required.
2-clarify the aim of paper in introduction[d1] .
Response: Thank you for the comment. The aim of this paper is to propose and validate a new research scheme for future full-scale/model test comparisons to verify the wind tunnel pressure measurement technique by the wind engineering community, which can reasonably separate the flow characteristic effects, the Re effects and the aero-elastic effects. The ultimate objective is that the wind engineering community can reasonably find the most significant adverse influence to the reliability of the traditional ABL wind tunnel pressure measurement technique after extensive comparisons using the proposed research scheme. These are emphasized in the revised manuscript (see the part highlighted in purple).
3-check the data units in all tables.
Response: Thank you for the suggestion. Data units in all tables have been checked. Data in the second column in Table 1 are with no unit in the original manuscript. We have added the unit for them. Other data units are correct. Thank you for helping us.
4- describe about flow characteristic effects
Response: Flow characteristic effects mean that relying on passive simulation devices, complete turbulent flow characteristics of the realistic ABL flow field can hardly be truthfully simulated in a traditional passive wind tunnel. Wind tunnel model tests treat empirical ABL flow characteristics presented in Codes of Practice and monographs as simulation targets. However, it was found that the turbulent flow characteristics simulated in the wind tunnel deviated from their simulation targets in many cases, causing significant negative effects. According to Pang (2006), the realistic mean wind speed profiles could be correctly simulated using spires and roughness elements in the wind tunnel, and the simulated turbulence intensities and power spectral densities were accurate below a certain height. However, when the height was increased, the simulated turbulence intensities were sometimes lower than their simulation targets, and the simulated power spectral densities became unrealistic. The simulated turbulence integral scale decreased with the increase in height, which was against the physical truth. These are explained in the introduction part in the revised manuscript (see the part highlighted in yellow).
Reference:
Pang JB (2006) Observation and wind tunnel simulation of strong wind characteristics in coastal and mountainous areas. Doctoral Dissertation, Tongji University, Shanghai (in Chinese).
5- What’s the effect of Reynolds number effects on pressure measurement model tests
Response: Basically, the flow separation location for flow around a bluff body without corners is sensitive to Re. Since the Re for model tests is usually two orders of magnitude smaller than the Re for the prototype flow event, significant distortions might be caused for the model test data. These are explained in the introduction part in the revised manuscript (see the part highlighted in navy blue).
6- describe about model surface roughness and placing the actively controlled devices
Response: Model surface roughness means the method of sticking paper belts on the model and adjusting the flow velocity in the wind tunnel, which has been extensively used in the wind-engineering community to reduce Re effects. So it is called high Re effects simulation. Through high Re effects simulation, model test data obtained with a low Re can approach the full-scale measurement results with a high Re. Placing the actively controlled devices means to place actively controlled devices in the beginning of the wind tunnel’s working section to increase the turbulence intensity and the turbulence integral scale of the simulated flow. The employed actively controlled devices included vibrating grids and active vanes. These are explained in the revised manuscript (see the part highlighted in italic).
Round 2
Reviewer 1 Report
All of my queries are appropriately answered by the authors. I thank you very much for this.
The word "simulaiton" in Figures 11 and 12 is wrong and should be changed to "simulation".
Author Response
All of my queries are appropriately answered by the authors. I thank you very much for this.
Response: Thank you so much for your time and efforts invested in reviewing our manuscript. We hope the quality of the revised manuscript can reaches the standard of a publication in the present journal.
The word "simulaiton" in Figures 11 and 12 is wrong and should be changed to "simulation".
Response: Thank you for pointing out our typo. We have revised it accordingly.
