the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Feb 2025
Assessing the impact of wind profiles at offshore wind farm sites for field data-enabled design
Abstract. As wind turbines grow and wind farms become denser, more insight into real metocean conditions is essential for operational efficiency and load assessment. Light Detection And Ranging LiDAR) technology, which can substitute the use of meteorological masts, has garnered significant attention in the literature. However, it indirectly measures wind parameters, relying on assumptions and built-in algorithms. Wind field reconstruction (WFR) methods offer users greater control over LiDAR measurements, enabling customised flow assumptions and parameter estimation. These measurements were taken during a measurement campaign on a wind turbine in the Belgian offshore zone. The WFR method has detected weather events, such as high shear, during the measurement campaign. These events are also linked to on-site weather conditions by using open-source metocean data. The findings align with the current literature on the correlation between events and weather conditions and the clear difference between wind profiling and a power law wind profile for loads design as proposed by the International Electrotechnical Commission (IEC) standard. The results emphasise the importance of real measurements in understanding wind field characteristics, offering improved accuracy compared to standard assumptions, such as the IEC power law profile used for load design. This work underscores the value of real-life wind profiling for designing and operating wind farms in offshore environments.
- Preprint
(3991 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2025-9', Anonymous Referee #1, 21 Feb 2025
Dear authors and editors,
This article focuses on analyzing the wind conditions located in the Belgian Sea, by combining nacelle LiDAR data, wind turbine SCADA data, and site buoy marine meteorological data. On the level of research significance, I think this article is worth publishing. Because these wind conditions can be references for the design and control of offshore wind turbines, and reveal the limitations of the current turbine design standards. Overall, the authors have conducted a higher quality data analysis and data presentation. However, from the perspective of an academic paper, I think the writing logic of the article can be further improved. For example, certain content that should appear in the results section is presented in the methods section. For a data analysis type of article, I think an early introduction in the second chapter on how the data was collected, what data was used for the analysis, what key parameters the authors wanted to present, and so on, would have helped to improve the logic and readability of the article.
In addition, on the technical side, I think there is one more important point the authors did not mention:
1. The probe volume effect has not been mentioned in this paper. For the continuous wave lidar analyzed in this work, the probe length can not be ignored when For the continuous wave lidar analyzed in this work, the probe length can not be ignored when focusing at a far distance. The measured LOS speed is not point like but contaminated by LOS speeds over different heights and longitudinal distances. some of the distances in the prove volume near the rotor, it might be affected by the induction zone and making the homogeneous flow assumption used by the authors invalidated. This should be clarified or analyzed.Minor comments are listed below:
1. Abstract: I suggest pointing out what lidar measures (spatially averaged LOS speed) to give readers who do not know about lidar a basic idea of why lidar does not directly measure wind parameters.
2. Abstract: The abstract should be reconstructed. It should be read as a mini paper indicating the motivation, the data collected, the method applied to analyze the data, and finally the conclusions (preferably with key indicators, e.g., how many samples the measured wind profile is significantly different from the log profile).
3. Line 25, Can you add some references on how the wind profile can affect the turbine loads?
4. Line 34, How the averaged wind characteristics can improve wind farm operation should be further explained.
5. Line 35 to 45, There is an IEC-61400-13 standard which requires turbine manufactures to validate the turbine loads. This standard also recommends that nacelle lidar can be used for load validations of offshore turbines if the lidar system can provide a TI equivalent to anemometer measurement. The difference in TI by lidar and anemometer is still an ongoing research topic. Successful applying nacelle lidar for load validation has been demonstrated by https://doi.org/10.1002/we.2385. I suggest the authors to look into IEC-61400-13 and https://doi.org/10.1002/we.2385 as they are highly linked to the current study.
6. The citation in Line 40 (Commission and other, 2019) is a bit strange. I recommend to check all citations and preferably add hyper links to them.
7. Line 47 a methodology for resource assessment -> for site condition assessment.
8. Line 56. Normally the conclusions are given at the end. Maybe use “:an overview of points observed “
9. Line 63, direct wind measurements ->line-of-sight speed measurements.
10. Most of the sentences between Section 2 and Section 2.1 are background knowledge about lidar and wind field reconstruction and are not related to the method proposed by the authors. The authors should consider putting this content in the introduction or a subsection.
11. The caption of Figure 2 does not read clear.
12. Line 110, is it possible to provide a map of the measurement plane indicating the dominate wind direction and turbines? Also, the turbine rotor size should be mentioned.
13. Line 125, the authors say that the beams are nearly horizontal. This contradicts with Volume N in Figure 3.
14. Line 131-132, Firstly, the wind turbine yaw misalignment can be 10 degrees for typical MW level turbine. With the homogeneous flow assumption, one gets the wind direction relative the nacelle (lidar) orientation. To obtain the wind direction in the inertial coordinate system, one still needs the nacelle direction. Please clarify.
15. Line 161: theta -> theta_z.
16. Line 160 to Line 164 repeats the contents in Line 156-159.
17. Equation 4, not all variables are explained.
18. Line 171: The space in the beginning should be avoided.
19. Line 173 to Line 175 how the TI is being calculated? By the nacelle anemometer or LOS speeds, please clarify.
20. Figure 4 should appear at the “Results” section.
21. “Meetnet Vlaamse Banken data” is explained in Section 3 but it firstly appear at Line 185.
22. The caption of Figure 7 does not read clear, please provide more details.
23. Line 232, can you explain why 2.56D is chosen? To avoid rotor induction effect?
24. Figure 8: not all lines are provided with legends.
25. Line 316, please specify the figure number
26. Section 4.6, it would be great to show the non-power law MAE histogram as well
27. Figure4, the most left chart looks like a wake profile. The authors should provide further information on these Events, e.g. wind direction, atmosphere stability, TI, wind wake misalignment on so on.
28. I suggest that the authors put the overview of the measurement campaign (basically Section 3) after introduction (Section 1) to improve the logic of this manuscript.
Citation: https://doi.org/10.5194/wes-2025-9-RC1 -
RC2: 'Comment on wes-2025-9', Anonymous Referee #2, 05 Mar 2025
Dear authors,
I have strong difficulties understanding what the scientific objective of your study is, the research question(s), and what exactly you want to address with the analysis you present in your study. Right now, the manuscript reads as a pure description (a technical report) of a measurement campaign performed with a nacelle lidar on a turbine at the Belgian Sea. There should be a question to address, a method to pursue, an idea/hypothesis to test in order to write a scientific paper. None of these things are attempted in this work. Therefore, I am unfortunately recommending the rejection of the paper. However, since it seems that the main author is a PhD student, I am giving here recommendations on the actual manuscript to improve the work and perhaps focus the analysis to answer a research question. If your intention is to attract the attention of the community with an interesting dataset, then a research paper is not the right choice but instead you can use a journal that receives datasets contributions (I think wind energy science has now an option to write a dataset-type of manuscript). However, even if you decide to go that way, you need to show that your dataset is indeed attractive for this and that reason, and not simply by plotting the measurements you have acquired during your project (also you probably need to archive the dataset for community use).
General comments:
- Introduction: you provide a very long context on the importance of offshore wind energy. Of course, the introduction must have this context, but a paragraph is more than sufficient. Also, you are providing too broad contexts and backgrounds to the subjects (for example when you talked about lidar). Even the introduction needs to be concise, precise and have only the elements a reader needs to understand your work.
- Figures should really be explanatory or provide results. Figure 1, as an example, does not add to the manuscript. Figure 3 could be combined with Fig. 2 (although I think you do not actually use a nacelle lidar configured as in Fig. 2b but only acquiring radial velocities at one range, so probably Fig. 2b needs to be corrected).
- In large parts of the other sections (after introduction), you again introduce and provide long context to subjects. An example (not the only one) is the first three paragraphs of Sect. 2, which are more appropriate for an introduction (In a much shorter fashion).
- About the power law: I really hope you are wrong and the power law is not the current industry standard for wind profiles. The power law can only be used in a climatological fashion or to extrapolate winds very locally (small vertical differences). I do not think it makes sense to give the power law the importance that you do as it is well-known by the community of the limitations of its usage.
- Repetitions: some places of your manuscript you repeat things, and it gives the impression that the version of the manuscript is not well revised by the co-authors. An example of this is lines 111 and 112 where you write the same twice.
Also lines 325 vs 300
- Section 2.2.2 this is over descriptive and superfluous. You can just introduce the metric (Eq. 4) you use to evaluate veer and that is all
- Section 4.4 about the shear. If you estimated shear then you need to tell information about the vertical levels you actually use to derive it and methods. For example, if you use the power law and all the vertical levels you seem to be analyzing (20—180 m), then the computation is wrong as the shear exponent varies with height. Please do not say things like “The wind profile is usually extrapolated using the power law”
Specific comments (I will provide some specific remarks until page 17 because I realized that the rest of the manuscript was not going to address a research question anyway so I kind of gave up, sorry to be this honest):
- Line 21 replace “gaining popularity” by “attractive”
- Line 30 and similar: LiDAR is for many years a word in the AMS glossary, just like radar and so you do not need to explain the acronym and you do not need to capitalize, so it is just lidar.
- Line 33 “This allows … to be time averaged”: No, the fact that you can velocity projections with a lidar does not allow for velocities to be time-averaged
- Your references to the IEC standard, such as in line 40 need to be changed. I mean the style you use to reference those (it should not be “Commission and other, 2019” or similar)
- I guess you need to add around line 56 that you also have a section on Conclusions?
- Line 90 the cyclops dilemma does not create ambiguities in the los measurements
- Line 94 you do not need at least 2 los, you need 3; you only need 2 if you assume something about the third los (for example that w=0)
- Some of your style to refer to work is wrong. For example, line 99 should read “… such as those in Schlipf et al. (2012) and Wagner et al., (2014),…”
- Captions on figures are weird. For example in Fig. 2 should be “Figure 2. (a) Nacelle lidar schematics or a 2-beam reconstruction method implementation. (b) A representation…”
Similar happens in Fig. 6, for example.
- Line 105 units and quantities should be separated by a small space
- Lines around line 110: how long time does it take this ZX lidar to scan the full circle? 1 s? Please rephrase so that the reader understand that the instrument provides radial velocities at 64 positions around the scanning circle.
- Lines around line 120: The 2-beam method assumes w=0 to infer the two horizontal velocity components. Nothing more. The lines between 120 and 133 are also very superfluous.
- Lines 137-138 where did these lines come from?
- Line 147 please no this 1/7value
- 7 you use different sign for u and V
- Figure 4: where is this data from? You have not introduced the dataset yet
- Lines 210 and 211 now you change units to be in italics… they should be always normal text
- Line 224 a lidar or anything measures the data
- In Sect. 31. You should introduce the lidar, type, specification
- Caption of Fig. 5 I guess it is “wind farm” and not “wind turbine”
- First 5-6 lines in Sect. 4.1 again superfluous.
- Line 174 if you provide a wind speed value then you need to say at what height it is
- Section 4.2 if you use SCADA measurements, then you need to say from which sensor specifically. If there is an estimation you do not know the origin (e.g. the so-called fit-derived method) then you cannot use it for your analysis as only god knows how this is derived.
- Line 308 it is Obukhov length, not Monin-Obukhov length
Citation: https://doi.org/10.5194/wes-2025-9-RC2 -
RC3: 'Comment on wes-2025-9', Anonymous Referee #3, 07 Mar 2025
In this paper, the authors have collected LiDAR-measurements from an offshore wind turbine in Belgian waters. A method called wind field reconstruction is applied to obtain useful data from the measurements, and wind speed, wind shear profile and wind veer gradients are extracted. Additionally, SCADA-data and measurements from piles and buoys (Meetnet Vlaamse Banken) are used to validate and supplement the LiDAR results, in addition to LiDAR results directly from the manufacturer’s post-processing system. The data are collected over at total of 12 months, with an interruption in the summer months. The study further investigates wind directions, wind shear profile, wind veer gradient, atmospheric stability. High shear and veer events, and stability conditions are categorized based on months, and correlations between such events and atmospheric stability, turbulence intensity (from SCADA-measurements), wind-wave misalignments and differences between air temperature and sea temperature. Several events are seen where the power law profile does not fit well with the measured wind profile, for instance during storms. Moreover, veer gradients above 0.2 deg/m frequently occurs.
Collecting and post-processing such data is important for assessing power production and loads in wind turbines, and this paper provides valuable understandings in the applicability of LiDAR-measurements, seasonal variations in environmental conditions, and discrepancies between measured wind profiles and standardized methods. However, the paper needs considerable revision to improve readability and quality, and some issues must be clarified.
Generally, a better categorization of content in terms of introduction, methodology and results should be implemented. As of now, much of the methodology section should be moved to introduction because it describes motivation and state-of-the-art rather than the methodology. The author is invited to think through the purpose of each section before revising.
See more text-specific comments below. Please note that I cannot comment upon the technical content related to LiDAR-measurements, as it is outside my aera of expertise.
Title: The title is misleading, as the actual impacts of the wind profiles on turbine performance and response is not assessed in this work.
Abstract
In general, the abstract should be more specific about the methods and results of the paper. It now consists of many diffuse sentences such as “The findings align with the current literature on the correlation between events and weather conditions and the clear difference between wind profiling and a power law wind profile for loads design as proposed by IEC”.
- Introduction
Line 43: “Such data reflects the actual atmospheric conditions, including localised effects of atmospheric stability, turbulence, and wind direction variability, which are essential for accurately assessing turbine performance and loading” Can you add some references discussing turbine performance and loading under these effects?
Line 44-45: Are real-life conditions not usual?
Line 47: “and the factors for turbine load design are analysed”. What does this mean? There is no evaluation on influence on turbine loads in this work.
Line 49: “validate key insights on the long-term steady state environmental conditions…” Do we have long-term steady state conditions in the wind?
Line 54: Be consistent with “Sect”. vs “Section”
Figure 1: It is not clear what this figure is intended to represent. Does it include data measurements/field data? The text with the reference (line 43-45) and the caption gives opposite impressions. And what is meant by field-enabled design validation?
- Methodology
Lines 58 to 61: Would be better off in the introduction. Same goes for lines 70-83. Wind veer should be defined early, as it is not as known as wind shear and turbulence intensity (not a typical design requirement).
Line 87: the abbreviation WFR has already been stated.
Figure 2: The LOS-abbreviations are not defined in the text.
Line 112: “in the first row of the wind farm” is repeated.
Chapter 2.2: It is unfortunate that alpha is used both to describe the LiDAR setup and the shear exponent
Line 120: It is not clear how Vx and Vy are obtained.
Lines 138-139: It seems this sentence is misplaced as it has no direct connection to the previous text.
Lines 143-144: “it has been proven that it cannot be as correct since shear is affected by numerous factors, such as turbulence intensity, atmospheric stability and surface roughness.” I think it would be more meaningful to say that TI and shear are measures of the atmospheric stability rather than that shear depend on TI. I would also recommend looking at the work by Olsen et al.: “Evaluation of Marine Wind Profiles in the North Sea and Norwegian Sea Based on Measurements and Satellite-Derived Wind Products” (2022).
Line 147: At first, it was a bit unclear that it is the power law exponent that is set to 0.14 in most (but not all?) of the attempts to fit the LiDAR wind profiles to the power law.
Line 152: Gao refers directly to another source when stating this – it would be better to refer to the initial/base source.
Line 154: Shu et al does not investigate loads and performance – why are they referred here?
Equation (4): The definition of the bulk veer gives an absolute value – how do you differ between veer and backing?
Line 160-164 is a replica of line 155-157.
Line 175-175: sigma_u or sigma_V?
Line 181-182: This sentence is not complete and should be revised. (“being only in near neutral conditions that this profile usually occurs”)
Line 184: “Length” is misspelled.
Line 185: There is a dot between the sentence and the reference (Holtslag).
Eq (8): A temperature parameter (273.15 + T_a) is missing in the denominator (ref Albornoz et al. who is referring to Bahamonde and Litran). Please clarify whether this is a typo, or if this is the equation applied in the analyses.
Eq (8): Why not use the same symbols for T_air and T_sea as in equation 6? And where is z taken? At the hub-height?
Equation (9) and this section: How is the Obhukov length used for categorization? That is, what are the criteria for being categorized as “stable”, “neutral”, etc?
Line 189: “the air and … temperatures” “sea” is missing.
Line 200: It would be more clear if it was explicitly stated that alpha is taken as 0.14 (if that is the case).
Line 203-204: What do you mean by “non-correlated” here, and what is a “weather event”?
Lines 215-220: It would have been useful to read something like this overview of the dataset earlier, perhaps in the introduction (in the paragraph starting with “In this study, a met…”
Line 206: Is the extreme direction change a spatial or temporal change?
Line 210-211: Why are units in italics?
3 Dataset
In general, it would be better to place this section before the Methodology-section, or to implement it in the beginning of the Methodology-section.Line 216: It is stated that three data sources are discussed, but only the LiDAR and Meetnet Vlaamse Banken data are discussed in the following sections.
Line 219: Do you have a reference to the Meetnet Vlaamse Banken data?
Line 226: What is meant by “heights for hub height”?
Line 240: Refers to Section 3, but this is also placed in Section 3.
Line 244: What does this mean: “However, the absence of wake data presents a limitation for wind …”?
Line 256: “These results are compared with the distribution of events” – what is meant by “distribution of events”?
4 Results
Line 262: “time series compared with SCADA and the manufacturer’s reconstruction.” What manufacturer?
Figure 7: The caption refers to a “first row (a) and a second row (b)”. Is this correct?
Line 294: Be consistent with using “2-beam” or “2-Beam”.
Line 306-309: This should be in the introduction.
Figure 9: For this figure to be useful in terms of seasonal variation, it should show the name of the months.
Line 310: “…the figure indicate that neutral conditions dominate…”. To me, based on Figure 9, it looks like very unstable conditions dominate in all months except month 8 and 10. Please clarify.
Lines 315-320: According to the authors, stable and very stable conditions are seen more frequently in the colder months. This explained by colder air moving over warmer water. However, winter, and this temperature difference is typically associated with unstable conditions, and summer, with warmer air over colder water is associated with stable conditions (see. e.g. Maarten Holtslag’s phd thesis “Far offshore wind conditions in scope of wind energy”). Please clarify. You discuss this in lines 351-355 too, with the opposite conclusion than in lines 315-320. I believe the discussion in lines 351-355 is more correct.
Lines 321-323: For the offshore location OWEZ, Sathe found that neutral conditions dominated in high wind speeds, but this was not the case below say 10-11 m/s. Similar trends were also found for the other three locations. I think this generalization of their results is erroneous. Do you have other resources supporting the findings that neutral conditions dominate offshore? Referring also to the statement in lines 311-312.
Line 332-333: “…if the power law is assumed, the IEC standard will be more conservative in the majority of cases.” Is it conservative to assume high shear? Conservative for the power production? For the turbine loads?
Line 335: this phrase “which implies a non-power law profile the large absolute value it is” should be rephrased.
Line 339-340: “… it is possible to reconstruct the wind speed at several heights, so it is also possible to calculate the value of the power exponent using E1q. 3, in which the assumption of the power law profile is taken.” Wasn’t this also done when calculating the average shear coefficient in line 330? Why is it mentioned here?
Figure 11: Again, it would be useful to label the months (or at least categorize them into “summer”, “spring”, “winter”…
Line 346-347: This sentence is not complete (“While wind-wave..”)
Line 361: Can veer be both positive and negative? Or does backing have negative and veer positive gradients?
Line 371: What is the unit of the veer value of 0.2? What categories do you refer to?
Line 371-375: Be careful with assuming a causal relationship between veer and these “weather parameters”. While there can be a correlation between turbulence intensity and high veer, it doesn’t mean that turbulence intensity is influencing veer.
Lines 374-375: I am not sure if you are discussing the results in Figure 16c (e.g. the correlation between high veer and turbulence intensity) or discussing the uncertainties in the measurements.
Lines 380-385: I would recommend referring to Julie Lundquists chapter 27 “Wind shear and wind effects on wind turbines” in “Handbook of Wind Energy Aerodynamics”.
Lines 385-390: I think this would fit better in the introductory chapter, at is considers motivation for assessing veer.
Lines 398-400: You show an example of a storm – should this not be put under section “4.6.1 Storms”?
Line 405: “it is evident that the IEC standard assumption holds for most of the day, as seen in figure 10”. How does figure 10 and 19 support this?
Line 414: “appear unrelated” could be replaced by “is not possible to describe by the power law”
Line 415: This is not a complete sentence.
Line 418: “has very little similarity to the power law profile” would be better.
Line 426: Why is wind field reconstruction unavailable during the EDCs?
Conclusions:
As for the abstract: be more precise in what the methodologies and results of this work are.
Check these references:
- Line 36: NREL “.n.d.” – what does this mean?
- Generally, the references to the IEC-standards seem off (they refer to someone named “Comission et al.”
Generally:
- Use \noindent to prevent a new paragraph after an equation (if the following text belongs to the same paragraph as the equation).
- Avoid lonely subsections (e.g. 2.3.1, 2.4.1, 3.1.1, 4.1.1, 4.4.1).
- For citations referring directly to the author, the author name should not be in parentheses. E.g. line 235: “As described in (Marini et al., 2024)” should be “As described by Marini et al. (2024)”. In latex, using \cite{} instead of \citep{} resolves this issue.
Citation: https://doi.org/10.5194/wes-2025-9-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 168 | 34 | 7 | 209 | 6 | 6 |
- HTML: 168
- PDF: 34
- XML: 7
- Total: 209
- BibTeX: 6
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1