the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Offshore wind farm optimisation: a comparison of performance between regular and irregular wind turbine layouts
Maaike Sickler
Bart Ummels
Michiel Zaaijer
Katherine Dykes
Abstract. Layout optimisation is essential for improving the overall performance of offshore wind farms. During the past 15 years, the use of yield optimisation algorithms has resulted in a transition from regular to more irregular farm layouts. However, since the layout affects many factors, yield optimisation alone may not maximise the overall performance. In this paper, a comparative case study is presented to quantify the effect of the wind farm layout on the overall performance of offshore wind farms. The case study was performed to investigate two performance indicators: power performance, using yield calculations with WindPRO and wake-induced tower fatigue, using the Frandsen model. It is observed that irregular wind farm layouts have a higher annual energy production compared to regular layouts. Their power production is also more persistent (less sensitive) to wind direction, improving predictability and thus market value of power output. However, one turbine location in the irregular layout has a 24 % higher effective turbulence level, leading to additional tower fatigue. As a result, fatigue-driven tower designs would require increased wall thicknesses, which would result in higher capital costs for all turbine locations. It is demonstrated in this study that layout optimisation using a minimum inter-turbine spacing effectively resolves the induced wake issue while maintaining high-yield performance.
Maaike Sickler et al.
Status: final response (author comments only)
-
RC1: 'Comment on wes-2022-97', Andrew P.J. Stanley, 05 Dec 2022
Offshore wind farm optimisation: a comparison of performance between regular and irregular wind turbine layouts
Maaike Sickler , Bart Ummels, Michiel Zaaijer, Roland Schmehl, and Katherine Dykes
The paper compares several performance metrics of a regular and irregular wind farm layout. The authors conclude that irregular layouts are better in terms of energy production and power variability as a function of wind direction, but regular layouts are better in terms of fatigue on the turbine towers.
My Main Conclusion
While the paper has some interesting information, I don’t think any general conclusions about regular versus irregular layouts can be draw from what has been presented.
First, because most of the paper introduces only two turbine layouts for a single location, Section 5 should be the most important section that presents and defends why any of the observations are globally applicable. However Section 5 section does not convince me. Part of this could be that I don’t fully understand what was done, I had to read through the section several times and am still unsure if I understand correctly. How many additional cases were considered? What are the light and dark gray bars in Figure 8? I think for anyone to agree that the observations previously made between regular and irregular layouts, there need to be several demonstrations for a wide variety of wind plants (size, boundary shape, several wind roses, objective…).
Second, the majority of the paper compares one regular and one irregular layout, which were taken from another paper and I assume were optimized for some objective. With this information alone, I wouldn’t conclude that the observations made are general to all regular and irregular layouts, even for this specific location. What if I optimized a grid for a different objective, or included some additional constraints? Same with the irregular layout? What if I used a different regular layout that wasn’t a grid? The observations of this paper may be general, but I don’t think that should be concluded without more evidence.
A Few Other Things
- Line 45 – “Third, an optimisation of the farm layout inherently leads to an irregular pattern, as shown by most optimisation studies in the literature.” This is not necessarily true. It depends how you setup the problem. Plenty of wind farms are optimized with a regular grid.
- Figure 5 and the related discussion – Is this the right metric? This gives no indication of the probability of a high power fluctuation. I agree that a gridded layout will have some wind directions that are associated with very high losses, but if the grid is optimized those directions will be associated with very low probability. I’m not convinced that “The difference between the imbalance cost of the regular and irregular wind farm would then become visible within this 1.6 %.” Small note, I would put the two subfigures on the same plot, it would be easier to see the differences and conserve some space!
- Line 109 - “Interestingly, while monopile foundation designs are optimised for individual locations within an offshore wind farm, typically only a single tower design is applied based on the turbine location with the highest turbulence intensity.” This seems like an opportunity to optimize tower designs for single locations, rather than save enormous amounts of money by reducing the turbulence of the worst case tower location. It seems silly to hugely overdesign every tower because of one very poor instance.
- Already mentioned above, but I have very little idea what Figure 8 means.
- Line 168 – “Irregular wind farm layouts outperform regular layouts regarding energy production…” this information is still worth including, but it is important to remember that this is true by definition. Less constrained problems either perform just as well or better than more constrained problems.
- Line 169 – “A notable finding was that the irregular layout also increases the persistence to wind direction…” again already mentioned above, but is this still notable when the directional probabilities are considered?
- Figure 4 and 7 are really interesting!
I recognize that I may not have understood everything as was intended, and I may not have communicated my thoughts as clearly as I would have liked. If anything is unclear or you disagree with anything I have said, please reach out to me and we can continue this discussion.
PJ Stanley
Citation: https://doi.org/10.5194/wes-2022-97-RC1 -
RC2: 'Comment on wes-2022-97', Christopher Bay, 16 Mar 2023
This manuscript presents some analyses of performance of regular and irregular wind farms, proposing that irregular wind farms outperform regular wind farms in several performance indicators. The overall topic is of interest to the wind energy community, but I believe the manuscript needs to be improved significantly to be considered for publication. The analyses, while interesting, need to be broadened in scope to better draw general conclusions. I have provided specific comments below.
- I agree with the authors that the irregular wind farm studied in Section 3 outperforms the regular wind farm for energy yield and predictability, but I do not think the authors included sufficient discussion to justify the irregular farm outperforms in value to the electricity market. More discussion should be included to illustrate to the reader how this is true.
- Section 4 would be strengthened by including discussion around the changes in performance of the irregular farm after the turbines have been shifted by the 4 RD. Most likely, the change in energy production is minimal, but would be good to show.
- Section 4 would also be strengthened by including an approximation in the difference in tower cost between the 3 layouts (regular, irregular, and repositioned irregular). Discussion on the change of the repositioned irregular layout compared to the regular layout would be beneficial as well.
- Figure 8 needs a legend to indicate what the light gray and darker gray bars indicate. I also wonder if this information would not be more readable in a table. With the different y-axis scales between the plots, it is difficult to compare relative impacts across parameters. Also, some changes are so small they do no appear on the plots.
- Section 5 would be significantly improved with more details and/or a better illustration of the different cases being examined. As written, it is difficult to follow. Suggestions include: 1) including a figure that shows the two wind roses used in the analysis so the reader can understand the difference between the uniform and unidirectional wind roses, 2) a diagram or flow chart depicting the changes made between the cases; this is currently only in the text and is difficult to follow, and 3) a more clear description of the analysis occurring with the MDAO- and WindPRO-optimized layouts; the authors state that the farm layout pairs were compared, but I am unsure of what the pairs are and how they vary across Base, MDAO, and WindPRO.
- The conclusions drawn hold for the two cases examined in the manuscript, but are not proven for all cases, and should be noted as such. An interesting and worthwhile addition to this study would be an analysis of a larger sampling of irregular wind farm layouts which would help prove or disprove the conclusions in a more general sense. Additional analysis across other wind roses would also be helpful.
- A discussion of other potential effects of an irregular wind farm layout should be included as well, or at least acknowledged that an irregular layout can have other effects such as changes in cabling costs and impacts on navigability of ships through a wind farm. These two topics are not exhaustive of all the effects of irregular layouts compared to regular layouts.
Other comments/corrections:
- Line 58: Are there four levels in Figure 2? Or three levels? Are there two levels within the PI group? If Electricity Price is on a different level than Annual Energy Production, it is difficult to discern that through the flow of the arrows.
- Line 61: Consider rewording; as it reads, the first item does not fit grammatically.
- Line 62: Are there five sub-performance indicators? I see seven in Figure 2.
- Line 100: Typo in “indicator”.
- Vectorized images are preferred for readability.
Citation: https://doi.org/10.5194/wes-2022-97-RC2
Maaike Sickler et al.
Maaike Sickler et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 273 | 85 | 11 | 369 | 5 | 3 |
- HTML: 273
- PDF: 85
- XML: 11
- Total: 369
- BibTeX: 5
- EndNote: 3
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1