the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Oct 2023
Rotary-wing drone-induced flow – comparison of simulations with lidar measurements
Abstract. Ultrasonic anemometers mounted on rotary-wing drones have the potential to provide a cost-efficient alternative to the classical meteorological mast-mounted counterpart for wind energy applications. However, the propeller-induced flow may deteriorate the accuracy of free wind velocity measurements by wind sensors mounted on the drone. Therefore, we performed an experiment using three short-range continuous-wave Doppler lidars (DTU WindScanners) to measure the complex and turbulent three-dimensional wind field around a hovering drone at low ambient wind speeds. The results obtained by lidar measurements and computational fluid dynamics simulations are in good agreement. Both methods conclude that the disturbance zone on a horizontal plane 0.7 meters below the drone, extends about 2 meters upstream from the drone center for the horizontal wind velocity and more than 5 meters for the vertical wind velocity. By comparing wind velocities along horizontal lines in the upstream direction, we find that the velocity difference between the two methods is less than 0.1 ms−1 in most cases. Both plane and line scan results validate the reliability of simulations. Furthermore, simulations of flow patterns in a vertical plane at low ambient speed indicate that it is difficult to accurately measure the vertical wind component with less than 1 % distortion by drone-mounted sonic anemometers.
- Preprint
(10176 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1546', Anonymous Referee #1, 11 Nov 2023
The manuscript discusses the use of ultrasonic anemometers mounted on rotary-wing drones as a potentially cost-effective alternative to traditional meteorological mast-mounted anemometers for wind energy applications. However, concerns are raised about the accuracy of wind velocity measurements due to propeller-induced flow disturbances. The study presents an experiment using three short-range continuous-wave Doppler lidars (DTU WindScanners) to measure the complex and turbulent three-dimensional wind field around a hovering drone at low ambient wind speeds. The results from lidar measurements are compared to computational fluid dynamics (CFD) simulations to validate the accuracy of drone-mounted wind sensors. While the data and measurements are promising, the manuscript has several weaknesses. The explanations and discussions need improvement, and the reviewer provides specific suggestions:
- Clarify the novelty of this research in comparison to existing studies using drones with anemometers.
- Specify whether the proposed method complements or extends current techniques in wind field measurements.
- Enhance the logical relationships between references, discussing the limitations of previous research and their relevance to the current study.
- Address the variations in flow disturbances and spatial flow fields induced by different types of rotary-wing drones.
- Clearly state the contributions of this study regarding the development of advantages for rotary-wing drones in wind field measurements.
- Consider using more conventional symbols to represent horizontal wind speed components to improve readability and understanding.
- Provide clearer explanations for critical aspects such as sampling frequency, sampling time, and numerical simulation parameters.
- Address the potential reliability issues of averaging radar wind measurements in terms of error analysis and experimental design.
- Explain the rationale behind choosing three radar wind devices and their arrangement, considering potential sources of error.
- Discuss the impact of drone-mounted wind sensors on the measurement of turbulence characteristics, in addition to average wind speed.
- Improve English language expression, particularly regarding sentence structure and readability.
Citation: https://doi.org/10.5194/egusphere-2023-1546-RC1 -
AC1: 'Reply on RC1', Liqin Jin, 24 Jan 2024
Dear Reviewer,
We are glad that you find our research valuable and the results are promising. After reading through your comments, we implemented your comments to our manuscript. We sincerely appreciate your time and effort in reviewing our manuscript.
-
RC2: 'Comment on egusphere-2023-1546', Anonymous Referee #2, 26 Nov 2023
Jin et al. present a method to determine the flow around a rotary-wing UAS in the field. Such experiments are extremely valuable, and the solution to use a triple-Doppler short-range lidar setup is innovative and unique. It is very difficult to get such data in other ways. Despite the very good concept, I think the study misses a lot of opportunities for a more profound analysis. From the manuscript I cannot conclude if CFD simulations are good enough to study the flow around a multicopter well enough for sensor placement in all relevant conditions. It also remains unclear if the results can be transferred to other conditions and other platforms. I can only recommend the manuscript for publication in AMT after major revision. I give some general and specific comments below:
General comments:
- The authors motivate the work with wind energy research. The experiment is carried out at wind speeds below 4.1 m/s. Values of turbulence intensity are not even given. These are not conditions that are relevant for wind energy. I understand that these are conditions that are critical for sensor placement, but if the application is at higher wind speeds, it would be important to know if the sensor placement can be changed if low wind speeds are disregarded.
- To my understanding all the analysis is based on two short measurement periods. This seems very little in order to draw general conclusions.
- One main difference between the field experiment and the CFD which it is compared to is in my opinion the turbulence. The authors say that the k-epsilon model is used in RANS, but do not give details about the parameters that are set in the model setup (i.e. boundary conditions for k and epsilon) and how they compare to the measurements (I did not find it in Ghirardelli 2023 either). How do the results depend on the turbulence settings?
- I think an opportunity is missed to measure the downdraft with the sonic anemometers and the short-range lidars at the same time with a lower hover altitude above the sonics. This would give more confidence in the measurements and the comparison to the simulations.
- Is it really necessary to describe how wind is retrieved from Doppler lidar spectra in this study?
Section 2.2 I personally think that the Windscanner system is well described in literature and since it is only used for validation measurements here, a specification of the uncertainty and reference to the literature would be sufficient. The data processing description could be omitted in my opinion.
- Is a 5-m distance to the UAS a realistic position for a sensor in flight? What does that mean for the weight of the system, the flight time, the stability?Specific comments:
p.1, l.6: maybe introduce the CFD simulations first, before comparing to measurements.
p.2, l.41: what is a reference for the extensive CFD studies?
p.2, l.45f: I think it would be better to give a distance relative to the rotor diameter, like in the text before.
p.6, l.130f: It would be nice to know upfront what is the purpose of plane and line scan.
p.7, ll.134ff: I do not quite understand the reasoning for that pattern. If the lidars had just scanned a simple square with 20 lines (corresponding to the 20 orange dot lines) successively in y-direction, the scan would only last 20 s in my understanding. Would that not be better?
p.7, l.142: Was RTK GPS used in that case as mentioned in the drone description? I would not expect the drift in that case.
p.10, l.176 and Fig. 7: what is the "certain threshold" and how is it determined. The numbers in Fig.7c are much higher than in 7d. While it seems obvious for 7c that hard target hits are possible, why are the areas in 7d so broad?
p.10, l.188: Can you also give information about turbulence intensity?
p.12, l.206: can the differences in the comparison between CFD and measurement also be due to effects of unmodelled atmospheric turbulence?
p.15, l.227: Please also provide the distance relative to the drone size.
p.15, l.235: Maybe repeat what is the criteria for the disturbance zone.Citation: https://doi.org/10.5194/egusphere-2023-1546-RC2 -
AC2: 'Reply on RC2', Liqin Jin, 24 Jan 2024
Dear Reviewer,
We sincerely appreciate your time and effort in reviewing our manuscript and we are glad that you find our research valuable and the results are promising. After reading through your comments, we implemented your comments to our manuscript and replied in the attached file.
-
AC2: 'Reply on RC2', Liqin Jin, 24 Jan 2024
Status: closed
-
RC1: 'Comment on egusphere-2023-1546', Anonymous Referee #1, 11 Nov 2023
The manuscript discusses the use of ultrasonic anemometers mounted on rotary-wing drones as a potentially cost-effective alternative to traditional meteorological mast-mounted anemometers for wind energy applications. However, concerns are raised about the accuracy of wind velocity measurements due to propeller-induced flow disturbances. The study presents an experiment using three short-range continuous-wave Doppler lidars (DTU WindScanners) to measure the complex and turbulent three-dimensional wind field around a hovering drone at low ambient wind speeds. The results from lidar measurements are compared to computational fluid dynamics (CFD) simulations to validate the accuracy of drone-mounted wind sensors. While the data and measurements are promising, the manuscript has several weaknesses. The explanations and discussions need improvement, and the reviewer provides specific suggestions:
- Clarify the novelty of this research in comparison to existing studies using drones with anemometers.
- Specify whether the proposed method complements or extends current techniques in wind field measurements.
- Enhance the logical relationships between references, discussing the limitations of previous research and their relevance to the current study.
- Address the variations in flow disturbances and spatial flow fields induced by different types of rotary-wing drones.
- Clearly state the contributions of this study regarding the development of advantages for rotary-wing drones in wind field measurements.
- Consider using more conventional symbols to represent horizontal wind speed components to improve readability and understanding.
- Provide clearer explanations for critical aspects such as sampling frequency, sampling time, and numerical simulation parameters.
- Address the potential reliability issues of averaging radar wind measurements in terms of error analysis and experimental design.
- Explain the rationale behind choosing three radar wind devices and their arrangement, considering potential sources of error.
- Discuss the impact of drone-mounted wind sensors on the measurement of turbulence characteristics, in addition to average wind speed.
- Improve English language expression, particularly regarding sentence structure and readability.
Citation: https://doi.org/10.5194/egusphere-2023-1546-RC1 -
AC1: 'Reply on RC1', Liqin Jin, 24 Jan 2024
Dear Reviewer,
We are glad that you find our research valuable and the results are promising. After reading through your comments, we implemented your comments to our manuscript. We sincerely appreciate your time and effort in reviewing our manuscript.
-
RC2: 'Comment on egusphere-2023-1546', Anonymous Referee #2, 26 Nov 2023
Jin et al. present a method to determine the flow around a rotary-wing UAS in the field. Such experiments are extremely valuable, and the solution to use a triple-Doppler short-range lidar setup is innovative and unique. It is very difficult to get such data in other ways. Despite the very good concept, I think the study misses a lot of opportunities for a more profound analysis. From the manuscript I cannot conclude if CFD simulations are good enough to study the flow around a multicopter well enough for sensor placement in all relevant conditions. It also remains unclear if the results can be transferred to other conditions and other platforms. I can only recommend the manuscript for publication in AMT after major revision. I give some general and specific comments below:
General comments:
- The authors motivate the work with wind energy research. The experiment is carried out at wind speeds below 4.1 m/s. Values of turbulence intensity are not even given. These are not conditions that are relevant for wind energy. I understand that these are conditions that are critical for sensor placement, but if the application is at higher wind speeds, it would be important to know if the sensor placement can be changed if low wind speeds are disregarded.
- To my understanding all the analysis is based on two short measurement periods. This seems very little in order to draw general conclusions.
- One main difference between the field experiment and the CFD which it is compared to is in my opinion the turbulence. The authors say that the k-epsilon model is used in RANS, but do not give details about the parameters that are set in the model setup (i.e. boundary conditions for k and epsilon) and how they compare to the measurements (I did not find it in Ghirardelli 2023 either). How do the results depend on the turbulence settings?
- I think an opportunity is missed to measure the downdraft with the sonic anemometers and the short-range lidars at the same time with a lower hover altitude above the sonics. This would give more confidence in the measurements and the comparison to the simulations.
- Is it really necessary to describe how wind is retrieved from Doppler lidar spectra in this study?
Section 2.2 I personally think that the Windscanner system is well described in literature and since it is only used for validation measurements here, a specification of the uncertainty and reference to the literature would be sufficient. The data processing description could be omitted in my opinion.
- Is a 5-m distance to the UAS a realistic position for a sensor in flight? What does that mean for the weight of the system, the flight time, the stability?Specific comments:
p.1, l.6: maybe introduce the CFD simulations first, before comparing to measurements.
p.2, l.41: what is a reference for the extensive CFD studies?
p.2, l.45f: I think it would be better to give a distance relative to the rotor diameter, like in the text before.
p.6, l.130f: It would be nice to know upfront what is the purpose of plane and line scan.
p.7, ll.134ff: I do not quite understand the reasoning for that pattern. If the lidars had just scanned a simple square with 20 lines (corresponding to the 20 orange dot lines) successively in y-direction, the scan would only last 20 s in my understanding. Would that not be better?
p.7, l.142: Was RTK GPS used in that case as mentioned in the drone description? I would not expect the drift in that case.
p.10, l.176 and Fig. 7: what is the "certain threshold" and how is it determined. The numbers in Fig.7c are much higher than in 7d. While it seems obvious for 7c that hard target hits are possible, why are the areas in 7d so broad?
p.10, l.188: Can you also give information about turbulence intensity?
p.12, l.206: can the differences in the comparison between CFD and measurement also be due to effects of unmodelled atmospheric turbulence?
p.15, l.227: Please also provide the distance relative to the drone size.
p.15, l.235: Maybe repeat what is the criteria for the disturbance zone.Citation: https://doi.org/10.5194/egusphere-2023-1546-RC2 -
AC2: 'Reply on RC2', Liqin Jin, 24 Jan 2024
Dear Reviewer,
We sincerely appreciate your time and effort in reviewing our manuscript and we are glad that you find our research valuable and the results are promising. After reading through your comments, we implemented your comments to our manuscript and replied in the attached file.
-
AC2: 'Reply on RC2', Liqin Jin, 24 Jan 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 262 | 115 | 18 | 395 | 12 | 13 |
- HTML: 262
- PDF: 115
- XML: 18
- Total: 395
- BibTeX: 12
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
