the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 May 2021
2D Numerical Simulation Study of Airfoil Performance
Abstract. The aerodynamic characteristics of DTU-LN221 airfoil is studied. ANSYS Fluent is used to simulate the airfoil performance with seven different turbulence models. The simulation results for the airfoil with different turbulence models are compared with the wind tunnel experimental data performed under the same operating conditions. It is found that there is a good agreement between the computational fluid dynamics (CFD) predicted aerodynamic force coefficients with wind tunnel experimental data especially with angle of attack between −5° to 10°. RSM is chosen to investigate the flow field structure and the surface pressure coefficients under different angle of attack between −5° to 10°. Also the effect of changing air temperature, velocity and turbulence intensity on lift and drag coefficients/forces are examined. The results show that it is recommended to operate the wind turbines airfoil at low air temperature and high velocity to enhance the performance of the wind turbines.
- Preprint
(1915 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on wes-2021-45', Ghazaleh Ahmadi, 24 Jun 2021
- I suggest the following structure for Section 3:
3. Results and Discussions (Which has no content now)
3.1 Model validation
Right now this section doesn’t contain any material as result or discussion. Most of the text here can be in section 2.3 (Numerical setup) while talking about numerical simulations’ initial and boundary conditions and solution algorithm. I think it’s a good idea to merge 3.1 and what you have as 3.2 under this subsection. Basically, in this subsection, you are observing Cl and Cd of 7 turbulence models in order to find the best turbulence model in comparison with experimental results.
3.2 Effect of Angle of Attack (AOA)
From this subsection, only the results from the RSM turbulence model are presented. In this subsection which I suggest containing the material from both 3.3 and 3.4 of the current draft, you can present all the results which represent the effect of AOA.
3.3 Effect of air temperature
3.4 Effect of air speed
3.5 Effect of turbulence intensity
- The first sentence talking about Figure 3. You say it is a comparison of 7 turbulence models and theoretical? What do you mean by theoretical? Maybe you mean experimental.
- AOA<-5: You write: "noticeable deviation between experimental results and experiments". I think you need to correct the sentence.
- AOA>10: you say k-w and RSM have the best approximation. From what I see, the orange line (inviscid) is capturing the experimental results beside k-w. In both Figures 3 and 4, the inviscid model shows relatively the best performance; then, why did you choose RSM for the simulation of the flow in the next subsections? I think you need to explain this clearly.
- Effect of AOA: line 321 of the current pdf. You say on the trailing edge the flow stays laminar even on the higher AOA where we should see the separation. Basically, RSM is unable to capture the separation happening on the trailing edge which maybe can explain the difference between the RSM model and experimental results at AOA>10. However, you call it the reason that airfoil aerodynamic performance and experimental results are in agreement and coincide.
- Effect of temperature:
The aerodynamic characteristics of an airfoil are usually represented by Cl, Cd, and their ratio (AKA L/D), and not lift and drag forces which are directly affected by the air density.
According to the definition of lift and drag force, both of them decrease at the same rate when the temperature increases. As they both decrease at the same rate, they can cancel out each other in L/D and it may have no information about the effect of temperature on the aerodynamic performance of the airfoil. As we see no difference between Cl and Cd profiles in different temperatures. The goal of this section is in the first sentence as "investigating the effect of changing air temperature on the airfoil characteristics"; the difference between lift and drag force in different temperatures cannot explain this.
Technical comments
- The definition of lift and drag and their coefficients can be relocated from Introduction (line 26 to 35) to section 3.2 where it is more relevant.
- Section 3 starts with the description of the computational grid; this should be in section 2.3 where you talk about numerical simulations which I suggest calling “numerical setup”.
- I recommend starting section 3.2 directly by discussing/defining lift and drag etc. There is no need to talk about the turbulence models again.
- Section 2.1 title is suggested to be “Experimental setup”.
- Section 2.2 title is suggested to be “Mathematical modeling”.
Section 2.2 is devoted to describing 7 different RANS models; thus, it is more impressive to have 7 subsections, each one dedicated to one of the turbulence models. This is much easier to remember and less confusing for the readers. I suggest moving the content of 2.2.2 to 2.2 . Starting 2.2 with the sentence “In CFD, RANS is …”
- Define the abbreviation form of Angle of Attack at the beginning as "AOA" and use it everywhere in the paper instead of the long-form or other forms like “attacks of angel”.
- Line 383, the section of TI: change in the drag coefficient with the variation of air temperature! Here you are talking about the effect of AOA.
Citation: https://doi.org/10.5194/wes-2021-45-RC1 -
AC1: 'Reply on RC1', Nasser Shelil, 13 Jan 2022
Thank you for your valuable revision which is really enhanced the quality of the paper.
The attached is the revised version of the paper considering the comments with language check.
Note. The separation and stall at higher AOA is the subject of our new work.
-
RC2: 'Comment on wes-2021-45', Anonymous Referee #2, 17 Dec 2021
The manuscript descibes a series of numerical simulation of a single wind turbine airofil. Although the setup tries to match current practice, it seems that it falls short behind its ambition. The descritption reads like it is a study on the impact of several control parameters of the specific numerical solver. It is emphasized more than ones that majorly the selection of the turbulence model is one of the major results. In the text then, parameters like air speed, air temperature and turbulenc intensity are outlined - unfortunately without any reflection of their physical relevance.
Anyhow, there is some doubt on the numerical setup. The provided mesh images don't show a mesh that is reasonably capable to capture viscous boundary layer effects. Judging on turbulnece model by lift and drag coefficient only is much to crude. It neglects the option of compensation of local errors in integaral values. For such an anlysis, at least pressure distributions, even better velocity profiles normal to the wall must be analysed and compared. And - aside that an inviscid Euler simulation is not a turbulence model - the paper fails completely to explain, why all RANS simulations are by 27% far off the experimental data, while an inviscid one is close by.
At the end, in the conclusion mainly standard knowledge on airfoil characterisitics is provided. Such knowledge has been published in summary long before (e.g. Aboot & Doenhoff 1949). It doesn't justify a peer review publication. For such, the study must be more focused on the turbulence models with a proper detailed comparison including pressure distribution, velocity profiles, separation onset. And, it must provide more conclusive results than those which have been cited. The grid convergence study has to be quantified (e.g. Richardson extrpolation, y+-distributions) as there is strong doubt on the suitability of the meshes as plotted.
For detailed comments see the attached PDF. It is strongly required to perform a language check.
-
AC2: 'Reply on RC2', Nasser Shelil, 13 Jan 2022
Thank you for your valuable revision which is really enhanced the quality of the paper.
The attached is the revised version of the paper considering the comments with language check.
Note. The separation and stall at higher AOA is the subject of our new work.
-
AC2: 'Reply on RC2', Nasser Shelil, 13 Jan 2022
Status: closed
-
RC1: 'Comment on wes-2021-45', Ghazaleh Ahmadi, 24 Jun 2021
- I suggest the following structure for Section 3:
3. Results and Discussions (Which has no content now)
3.1 Model validation
Right now this section doesn’t contain any material as result or discussion. Most of the text here can be in section 2.3 (Numerical setup) while talking about numerical simulations’ initial and boundary conditions and solution algorithm. I think it’s a good idea to merge 3.1 and what you have as 3.2 under this subsection. Basically, in this subsection, you are observing Cl and Cd of 7 turbulence models in order to find the best turbulence model in comparison with experimental results.
3.2 Effect of Angle of Attack (AOA)
From this subsection, only the results from the RSM turbulence model are presented. In this subsection which I suggest containing the material from both 3.3 and 3.4 of the current draft, you can present all the results which represent the effect of AOA.
3.3 Effect of air temperature
3.4 Effect of air speed
3.5 Effect of turbulence intensity
- The first sentence talking about Figure 3. You say it is a comparison of 7 turbulence models and theoretical? What do you mean by theoretical? Maybe you mean experimental.
- AOA<-5: You write: "noticeable deviation between experimental results and experiments". I think you need to correct the sentence.
- AOA>10: you say k-w and RSM have the best approximation. From what I see, the orange line (inviscid) is capturing the experimental results beside k-w. In both Figures 3 and 4, the inviscid model shows relatively the best performance; then, why did you choose RSM for the simulation of the flow in the next subsections? I think you need to explain this clearly.
- Effect of AOA: line 321 of the current pdf. You say on the trailing edge the flow stays laminar even on the higher AOA where we should see the separation. Basically, RSM is unable to capture the separation happening on the trailing edge which maybe can explain the difference between the RSM model and experimental results at AOA>10. However, you call it the reason that airfoil aerodynamic performance and experimental results are in agreement and coincide.
- Effect of temperature:
The aerodynamic characteristics of an airfoil are usually represented by Cl, Cd, and their ratio (AKA L/D), and not lift and drag forces which are directly affected by the air density.
According to the definition of lift and drag force, both of them decrease at the same rate when the temperature increases. As they both decrease at the same rate, they can cancel out each other in L/D and it may have no information about the effect of temperature on the aerodynamic performance of the airfoil. As we see no difference between Cl and Cd profiles in different temperatures. The goal of this section is in the first sentence as "investigating the effect of changing air temperature on the airfoil characteristics"; the difference between lift and drag force in different temperatures cannot explain this.
Technical comments
- The definition of lift and drag and their coefficients can be relocated from Introduction (line 26 to 35) to section 3.2 where it is more relevant.
- Section 3 starts with the description of the computational grid; this should be in section 2.3 where you talk about numerical simulations which I suggest calling “numerical setup”.
- I recommend starting section 3.2 directly by discussing/defining lift and drag etc. There is no need to talk about the turbulence models again.
- Section 2.1 title is suggested to be “Experimental setup”.
- Section 2.2 title is suggested to be “Mathematical modeling”.
Section 2.2 is devoted to describing 7 different RANS models; thus, it is more impressive to have 7 subsections, each one dedicated to one of the turbulence models. This is much easier to remember and less confusing for the readers. I suggest moving the content of 2.2.2 to 2.2 . Starting 2.2 with the sentence “In CFD, RANS is …”
- Define the abbreviation form of Angle of Attack at the beginning as "AOA" and use it everywhere in the paper instead of the long-form or other forms like “attacks of angel”.
- Line 383, the section of TI: change in the drag coefficient with the variation of air temperature! Here you are talking about the effect of AOA.
Citation: https://doi.org/10.5194/wes-2021-45-RC1 -
AC1: 'Reply on RC1', Nasser Shelil, 13 Jan 2022
Thank you for your valuable revision which is really enhanced the quality of the paper.
The attached is the revised version of the paper considering the comments with language check.
Note. The separation and stall at higher AOA is the subject of our new work.
-
RC2: 'Comment on wes-2021-45', Anonymous Referee #2, 17 Dec 2021
The manuscript descibes a series of numerical simulation of a single wind turbine airofil. Although the setup tries to match current practice, it seems that it falls short behind its ambition. The descritption reads like it is a study on the impact of several control parameters of the specific numerical solver. It is emphasized more than ones that majorly the selection of the turbulence model is one of the major results. In the text then, parameters like air speed, air temperature and turbulenc intensity are outlined - unfortunately without any reflection of their physical relevance.
Anyhow, there is some doubt on the numerical setup. The provided mesh images don't show a mesh that is reasonably capable to capture viscous boundary layer effects. Judging on turbulnece model by lift and drag coefficient only is much to crude. It neglects the option of compensation of local errors in integaral values. For such an anlysis, at least pressure distributions, even better velocity profiles normal to the wall must be analysed and compared. And - aside that an inviscid Euler simulation is not a turbulence model - the paper fails completely to explain, why all RANS simulations are by 27% far off the experimental data, while an inviscid one is close by.
At the end, in the conclusion mainly standard knowledge on airfoil characterisitics is provided. Such knowledge has been published in summary long before (e.g. Aboot & Doenhoff 1949). It doesn't justify a peer review publication. For such, the study must be more focused on the turbulence models with a proper detailed comparison including pressure distribution, velocity profiles, separation onset. And, it must provide more conclusive results than those which have been cited. The grid convergence study has to be quantified (e.g. Richardson extrpolation, y+-distributions) as there is strong doubt on the suitability of the meshes as plotted.
For detailed comments see the attached PDF. It is strongly required to perform a language check.
-
AC2: 'Reply on RC2', Nasser Shelil, 13 Jan 2022
Thank you for your valuable revision which is really enhanced the quality of the paper.
The attached is the revised version of the paper considering the comments with language check.
Note. The separation and stall at higher AOA is the subject of our new work.
-
AC2: 'Reply on RC2', Nasser Shelil, 13 Jan 2022
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 610 | 1,791 | 28 | 2,429 | 29 | 16 |
- HTML: 610
- PDF: 1,791
- XML: 28
- Total: 2,429
- BibTeX: 29
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
