the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Experimental Validation of Parked Loads for a Floating Vertical Axis Wind Turbine: Wind-Wave Basin Tests
Abstract. Parked loads are a major design load case for vertical axis wind turbines (VAWTs) because of persistent high loads on the rotor when in standstill conditions. This paper examines the aerodynamic parked loads of model-scale floating troposkein VAWTs tested in a wind-wave basin to support development and validation of a parked loads model for floating VAWTs. We analyze the effects of wind speed, and turbine solidity (varying number of blades), and rotor azimuth on parked loads and investigate the impact of different platform conditions (comparing locked (fixed tower base) versus floating cases with and without waves). The experimental results indicate that parked loads (for both locked and floating platform conditions) and amplitude of turbine tilting increases with the wind speed, which is expected. The parked loads also increase with the increase of solidity, however the variation in loads in a revolution decreases for 3 blades versus 2 blades. If only aerodynamic parked loads are considered, the turbine with a floating platform exhibits lower parked loads compared to turbine with a locked platform (fixed base) due to the effect of the tilted condition of the floating platform. Moreover, comparison between floating with wind and wave, and floating with wind only cases show that although both exhibits park loads of similar magnitude, the former exhibits more high frequency variation due to coupled effects of wind, wave, and floating platform dynamics. Additionally, we present a semi-numerical tool for estimating parked loads of VAWTs that we improved and validated to predict the floating parked loads. The analytical model accurately predicted the parked load behavior of VAWTs for the range of effects noted above. The datasets from this experimental work can serve as benchmarks for validating other computational parked load estimating tools.
- Preprint
(2428 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 18 Dec 2024)
-
RC1: 'Comment on wes-2024-156', Anonymous Referee #1, 02 Dec 2024
reply
The study evaluates the aerodynamic parked loads of a model-scale floating troposkein VAWT in a wind-wave basin under different conditions: a fixed tower base, floating without waves, and floating with waves. The influence of wind speed, solidity (via blade count variation), and rotor azimuth on the parked load is analyzed. In general, the paper presents a good approach and addresses an important aspect of floating VAWTs, an area that remains underexplored in the literature. However, several comments are provided below to enhance the quality of the paper, particularly in terms of the findings and discussions:
Methods:
1. The authors highlight that using static airfoil polars is reasonable in the context of parked loads. However, this assumption holds only for the fixed-base system. For the other two cases (with the floating system and with floating system plus waves), the variation in the angle of attack could occur at a frequency high enough to impose unsteady load conditions. This aspect should be further discussed.
2. There is a typo in equations (1) and (2); both equations are identical. Please correct this.
3. Why not nondimensionalize all the parked load forces or even normalize them with respect to the rated forces? This would make it easier for readers to compare and interpret the results.
Results and Discussion:
4. Why not compare the experimental measurements directly against the UTD semi-numerical model, instead of presenting them in two separate graphs? How can the reader assess the accuracy of the numerical model if the results are not directly compared?
5. I believe it is crucial to show the influence of tower tilting on the angle of attack variation, either in the case of floating alone or floating with waves. These changes are likely to impact the estimation of the aerodynamic loads and should be addressed in the analysis (at least reporting the values might give a reader a sense on how unsteady are the loads).
6. The general division of sections could be improved. In both the abstract and the introduction, the study compares the fixed tower system with the floating system and the floating system with waves, highlighting how results vary as complexity increases. However, throughout the results section, these comparisons are not directly made. It would be beneficial to address these differences directly within the results section.
7. In case of floating tower with wind and wave: “the numerical model neglects the dynamic nature due to floating wind and wave effect” please elaborate with more details, why the model neglects these effects, what are the implications on the model predictions and how can it improved?
Conclusions:
8. The statement in the conclusion that the numerical model is validated and optimized is unclear, as the numerical model was not directly compared with experimental measurements throughout the paper. Additionally, other data from the numerical model, such as the variation in the blade angle of attack, should be presented. This is important for assessing the assumption that unsteady load corrections were not needed for the aerofoil polars.
9. The paragraphs starting at lines 359 and 364 are identical and appear to be replicated by mistake.
10. In the wave+floating parked load scenario, the numerical model has some drawbacks as reported in the results section. This issue is not highlighted in the conclusions.
11. The sentence “This study has advanced our understanding of the experimental parked loads on VAWTs and their impact on turbine performance” seems to describe the main objective of the paper. The numerical model, while useful, should be considered a tool to gain insight into unmeasurable quantities after validating the load measurements. It might be better to frame this as the central goal of the study.
Citation: https://doi.org/10.5194/wes-2024-156-RC1 -
RC2: 'Comment on wes-2024-156', Anonymous Referee #2, 05 Dec 2024
reply
The manuscript presents results on the thrust and lateral forces of a laboratory-scale model of a Vertical Axis Wind Turbine (VAWT) tested in a wind-wave basin under parked conditions. Experimental data were compared with a semi-numerical code, which retrieves gravitational loads from the experimental measurements, in several cases: i) fixed-bottom – wind; ii) floating – wind – no waves; iii) floating – wind – waves. The goal of the work was to assess parked loads in relation to design parameters, such as the number of blades and free-stream velocity.
Scientific Significance:
The manuscript presents an original investigation using a laboratory-scale model to measure parked loads. Assessing forces in parked conditions is crucial for estimating the lifespan of a wind turbine in wind scenarios involving velocities higher than the cut-out wind speed. The authors also propose a semi-numerical code to predict the loads, which could be useful for designing full-scale turbines according to industry standards.Scientific quality:
Questions are addressed below in regards to scientific quality:
- The authors propose a sensitivity analysis on the number of blades and wind speed. The effects of the number of blades on rotor loads have already been demonstrated in Sakib and Griffith (2022). The question that arises is: why should the number of blades have a greater impact on the design of a VAWT in parked loads than on the design of the turbine in operating conditions? Furthermore, why does the analysis focus solely on the number of blades, while the variation of solidity due to chord is disregarded? Please provide justification for these choices.
- Is there any comparison in operating conditions to validate the numerical code against experimental data? The numerical model depends on data obtained from experiments. Is it not possible to formulate the model in such a way that it does not rely on experimental measurements? A model that does not depend on experiments would be useful for estimating loads without the need for additional testing. Furthermore, comparing numerical results with experimental data would serve as proper validation.
- Line 121: Please clarify the method used to obtain ‘full-scale’ wind speeds and the parameters the authors used to derive the range of 14.7 – 36.4 m/s according to Froude’s law. It may be helpful to present the data in a table, showing both the scaled model and full-scale values.
- Line 137: Are the values for wave period and height representative of full-scale scenarios?
- Please clarify the sentences in line 143: “No wake effects have been included…” and line 147: “Blade wake […] were considered…”. Two different wake effects are mentioned, one of which is included while the other is not. Please provide further explanation. Induction due to the wake should be accounted for anyway: also VAWT in steady conditions produces blockage caused by the wake despite the rotor is still.
- Dynamic stall is not considered, nor are unsteady effects in the case of wind and wave excitation. Have the reduced frequencies been evaluated to demonstrate that unsteady aerodynamics does not occur?
- Equations 1 and 2 are identical.
- How have the polars been adjusted to extend their applicability to VAWT operations, where angles of attack can reach very high values (especially in parked conditions, where peripheral speed is nearly zero)?
- Inertial effects are defined as the projection of gravitational loads on the structure. Shouldn't proper inertial effects be considered when the turbine is subjected to both wave and wind forces? The entire structure experiences motion, with associated velocities and accelerations that affect the aerodynamics. Consider using a different term instead of “inertial effects,” as this might be interpreted as including inertia forces in the model.
- Figure 8: The use of ‘uncertainty’ bars is unclear. According to the authors, the variability of thrust and lateral forces is attributed to azimuthal positions. Are the uncertainty bars representing the amplitudes over a period?
- A general question regarding experimental data: Have the authors assessed the uncertainty in the measurements?
- Figures 10, 11, and 12: The comparisons at different wind speeds in separate plots make it harder to understand the results. Consider merging them into a single plot or showing the effect of wind speed in one plot and the effect of the floating motion in another (e.g., in Figures 10 and 11). Additionally, consider discretizing the x-axis in increments of 90 degrees.
- Figures 14, 15, 17, and 19: If the purpose is to compare measurements with numerical data, these comparisons should be shown in the same plot for clarity, for each wind speed.
- Can the authors clarify how the data over one rotation have been considered (e.g., in Figure 19)? Is the force over the θ angle an average of several rotations or the last period? It could be useful to present the phase-averaged data.
Presentation quality
Please find a list of typos or request of clarification in the manuscript:
- Line 39: the authors refer to HVAWTs, it is suggested to call H-shape/H-shaped since HVAWT might be confusing (between HAWT and VAWT).
- Line 61: a instead of an
- Line 79: in instead of on
- Line 98 and Line 117: refer to the table 2 in the previous sentence, so it is clear that is connected to the concept explained.
- Line 128: is the operating condition
- Line 194: typo in “ for are”.
- Line 200: weight
- Figure 9. The figure has no legend. Please provide for better understanding
- Line 236-239: repetition
- Line 263: a dot is missing
- Line 317: delete a ‘.’
Citation: https://doi.org/10.5194/wes-2024-156-RC2
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
87 | 43 | 5 | 135 | 4 | 1 |
- HTML: 87
- PDF: 43
- XML: 5
- Total: 135
- BibTeX: 4
- EndNote: 1
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1