the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Mar 2024
Drop-size-dependent effects in leading-edge rain erosion and their impact for erosion-safe mode operation
Abstract. Leading-edge rain erosion poses a significant challenge for the wind turbine industry due to its detrimental effects on structural integrity and annual energy production. Developing effective mitigation strategies requires understanding the precipitation conditions driving erosion. The influence of the rain droplet diameter on both the formation of erosion damage and on erosion mitigation strategies has yet to be sufficiently understood. This study proposes an enhanced damage model based on the impingement metric as used in the state-of-the-art, but improved by including important and so far neglected physical mechanisms such as the recently described droplet slowdown and deformation effect. Several drop-size-dependent effects are identified within the damage model. Subsequently, their significance for leading-edge erosion is established by deliberately including and excluding them for comparison. Thereafter, the influence of the drop-size effects on the viability of the erosion-safe mode (ESM) is investigated. The outcome is that drop-size effects strongly impact the erosion process and should not be neglected during modeling. Large droplets are considerably more damaging than small droplets, even when normalized for water volume. This directly influences the parameter space of erosion, such as the relevant droplet diameter range that should be studied. The drop-size effects shift damage production to higher rain intensities. Roughly half of the erosion damage is produced by only 10 % of rain events. When drop-size effects are excluded, this value shifts to more than 20 %. Regarding the ESM, it is found that it can be utilized up to twice as efficiently when drop-size effects are adequately modeled. The findings highlight the criticality of drop-size effects in rain erosion modeling for wind turbine blades, impacting lifetime predictions, ESM viability, and the parameter space of leading-edge erosion. This paper also provides a formal derivation of impingement and describes a method for finding optimal ESM strategies.
- Preprint
(3066 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on wes-2024-33', Anonymous Referee #1, 25 Apr 2024
Use of DTU test data – 0.76 data is based on water spraying not pure droplet impacts and if the DNV RP 0171 calculation has been used to generate the VN curves then high speed video has shown the calculation doesn’t give accurate values of number of impacts. The data is therefore not directly comparable with real wind turbine data.
High speed camera and sensor measurements also show that there is a mixture of direct impacts and glancing blows which have a dramatic difference in terms of the measured strains. You have assumed in your models that all droplets impinge head on at 90° which is at least not true in the test rig and likely not true on the turbine. Please include this assumption in the paper.
Impingement on the blade is affected by blade geometry and needs to account for impact location. Many droplets of different sizes under different wind and rotational velocities don’t impact the leading edge according to two independent CFD model studies (OREC and Fraunhofer).
90 & 95 Averaged law and drop-size dependent law – erosion performance is heavily related to material properties so will be unique for each LEP and the test parameters. This should be made clear in the paper that you are only considering one case. We expect and have observed that some LEPs will be drop-size independent whilst others are dependent – as observed by Verma and Amirzadeh. The droplet size damage model is based on the response of a limited set of materials and discludes the individual stress strain response of those LEPs.
215 Vimpact – LEP materials can be tuned to erode differently at different velocities although in general Vimpact is the key driver.
220 according to CFD the effect of Vimpact leading to more impingement is correct but it isn’t linear as it changes impact location which can then be away from the LE. Please remove the statement that it is linear.
335 the normalized damage model only works for purely elastic LEPs
Citation: https://doi.org/10.5194/wes-2024-33-RC1 -
AC1: 'Reply on RC1', Nils Barfknecht, 23 Jul 2024
Dear Reviewer,
We would like to thank you for your time and effort in reviewing our manuscript. We believe that your comments have improved the quality of our work.
Sincerely,
Nils Barfknecht and Dominic von Terzi
Citation: https://doi.org/10.5194/wes-2024-33-AC1
-
AC1: 'Reply on RC1', Nils Barfknecht, 23 Jul 2024
-
RC2: 'Comment on wes-2024-33', Anonymous Referee #2, 01 May 2024
The paper investigates the effects of drop-size on leading edge erosion by developing a simplified model for computing the damage per rotor blade section. It heavily relies on assumptions from another paper that is still under review.Â
Overall the approach is interesting and research questions are of great interest, however the execution needs improvement.Â
The paper is written as if a general answer is provided to the research questions posed, instead it is a case study of a particular turbine in a particular location. This should be made clear throughout the paper. The authors would need to model several turbines and locations if they would like to arrive at some more general conclusions, this would also raise the significance/impact of the paper.Â
The model assumptions need to be discussed and their sensitivity needs to be assessed. A sensitivity analyses needs to be performed. Why is all damage assumed to happen around the LE? From an aerodynamic point of view this is not realistic. Generally the entire derivation seems convoluted and excessively detailed, whilst important limitations are left out. By using such a simplified model would it not make sense to reduce its complexity even further? After all some big assumptions are underlying its derivation so one could drop details without impacting its accuracy. Why is the model not verified with some high-fidelity simulations? This should be added.Â
The authors should consider splitting the paper, as currently it really is made-up of two parts, a model and discussion of drop size and the ESM. Each one could stand on its own, however currently the ESM section is extremely short and hard to follow. If the authors add to this section, which is necessary for the reader to follow the author's arguments, the paper will get even longer, especially if they address the comments made regarding the first section.
More detailed comments are given in the attached pdf using the comment feature of a pdf viewer. The authors can directly answer the comments in the pdf or address them in a sperate document.Â-
AC2: 'Reply on RC2', Nils Barfknecht, 23 Jul 2024
Dear Reviewer,
We would like to thank you for your time and effort in reviewing our manuscript. We believe that your comments have improved the quality of our work.
Sincerely,
Nils Barfknecht and Dominic von Terzi
Citation: https://doi.org/10.5194/wes-2024-33-AC2
-
AC2: 'Reply on RC2', Nils Barfknecht, 23 Jul 2024
- AC3: 'Comment on wes-2024-33', Nils Barfknecht, 23 Jul 2024
Status: closed
-
RC1: 'Comment on wes-2024-33', Anonymous Referee #1, 25 Apr 2024
Use of DTU test data – 0.76 data is based on water spraying not pure droplet impacts and if the DNV RP 0171 calculation has been used to generate the VN curves then high speed video has shown the calculation doesn’t give accurate values of number of impacts. The data is therefore not directly comparable with real wind turbine data.
High speed camera and sensor measurements also show that there is a mixture of direct impacts and glancing blows which have a dramatic difference in terms of the measured strains. You have assumed in your models that all droplets impinge head on at 90° which is at least not true in the test rig and likely not true on the turbine. Please include this assumption in the paper.
Impingement on the blade is affected by blade geometry and needs to account for impact location. Many droplets of different sizes under different wind and rotational velocities don’t impact the leading edge according to two independent CFD model studies (OREC and Fraunhofer).
90 & 95 Averaged law and drop-size dependent law – erosion performance is heavily related to material properties so will be unique for each LEP and the test parameters. This should be made clear in the paper that you are only considering one case. We expect and have observed that some LEPs will be drop-size independent whilst others are dependent – as observed by Verma and Amirzadeh. The droplet size damage model is based on the response of a limited set of materials and discludes the individual stress strain response of those LEPs.
215 Vimpact – LEP materials can be tuned to erode differently at different velocities although in general Vimpact is the key driver.
220 according to CFD the effect of Vimpact leading to more impingement is correct but it isn’t linear as it changes impact location which can then be away from the LE. Please remove the statement that it is linear.
335 the normalized damage model only works for purely elastic LEPs
Citation: https://doi.org/10.5194/wes-2024-33-RC1 -
AC1: 'Reply on RC1', Nils Barfknecht, 23 Jul 2024
Dear Reviewer,
We would like to thank you for your time and effort in reviewing our manuscript. We believe that your comments have improved the quality of our work.
Sincerely,
Nils Barfknecht and Dominic von Terzi
Citation: https://doi.org/10.5194/wes-2024-33-AC1
-
AC1: 'Reply on RC1', Nils Barfknecht, 23 Jul 2024
-
RC2: 'Comment on wes-2024-33', Anonymous Referee #2, 01 May 2024
The paper investigates the effects of drop-size on leading edge erosion by developing a simplified model for computing the damage per rotor blade section. It heavily relies on assumptions from another paper that is still under review.Â
Overall the approach is interesting and research questions are of great interest, however the execution needs improvement.Â
The paper is written as if a general answer is provided to the research questions posed, instead it is a case study of a particular turbine in a particular location. This should be made clear throughout the paper. The authors would need to model several turbines and locations if they would like to arrive at some more general conclusions, this would also raise the significance/impact of the paper.Â
The model assumptions need to be discussed and their sensitivity needs to be assessed. A sensitivity analyses needs to be performed. Why is all damage assumed to happen around the LE? From an aerodynamic point of view this is not realistic. Generally the entire derivation seems convoluted and excessively detailed, whilst important limitations are left out. By using such a simplified model would it not make sense to reduce its complexity even further? After all some big assumptions are underlying its derivation so one could drop details without impacting its accuracy. Why is the model not verified with some high-fidelity simulations? This should be added.Â
The authors should consider splitting the paper, as currently it really is made-up of two parts, a model and discussion of drop size and the ESM. Each one could stand on its own, however currently the ESM section is extremely short and hard to follow. If the authors add to this section, which is necessary for the reader to follow the author's arguments, the paper will get even longer, especially if they address the comments made regarding the first section.
More detailed comments are given in the attached pdf using the comment feature of a pdf viewer. The authors can directly answer the comments in the pdf or address them in a sperate document.Â-
AC2: 'Reply on RC2', Nils Barfknecht, 23 Jul 2024
Dear Reviewer,
We would like to thank you for your time and effort in reviewing our manuscript. We believe that your comments have improved the quality of our work.
Sincerely,
Nils Barfknecht and Dominic von Terzi
Citation: https://doi.org/10.5194/wes-2024-33-AC2
-
AC2: 'Reply on RC2', Nils Barfknecht, 23 Jul 2024
- AC3: 'Comment on wes-2024-33', Nils Barfknecht, 23 Jul 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 420 | 75 | 28 | 523 | 16 | 18 |
- HTML: 420
- PDF: 75
- XML: 28
- Total: 523
- BibTeX: 16
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1