Hybrid-Lambda: A low specific rating rotor concept for offshore wind turbines

Ribnitzky, Daniel; Berger, Frederik; Petrovic, Vlaho; Kühn, Martin

doi:https://doi.org/10.5194/wes-2023-72

Preprints

https://doi.org/10.5194/wes-2023-72

Preprints

17 Jul 2023

| 17 Jul 2023

Status: this preprint is currently under review for the journal WES.

Hybrid-Lambda: A low specific rating rotor concept for offshore wind turbines

Daniel Ribnitzky, Frederik Berger, Vlaho Petrovic, and Martin Kühn

Abstract. We introduce an aerodynamic rotor concept for an offshore wind turbine which is tailored for an increased power feed-in at low wind speeds by a substantial increase of the rotor diameter while maintaining the rated power. The main objective of the conceptual design is to limit the quasi-steady loads (blade flapwise root bending moment (RBM) and thrust) to the maximum values of a reference turbine. The outer part of the blade (i.e. outer 30 % span) is designed for a higher design tip speed ratio (TSR) and a lower axial induction than the inner part. By operating at the high TSR in light winds, the slender outer part fully contributes to the increased power capture. In stronger winds the TSR is reduced and the torque generation is shifted to the inner section of the rotor. Moreover, the blade design efficiently reduces the power losses when the flapwise RBM is limited through peak shaving, below rated wind speed. This is of high importance, given the wind speed distribution at offshore sites. The characteristics of the rotor are first investigated with stationary blade element momentum simulations and further analyzed with aeroelastic simulations, considering the flexibility of blades and tower to show that a structural design is feasible even for a blade of this size and complexity. The economic revenue and the cost of valued energy of the turbine is estimated and compared to the IEA 15 MW offshore reference turbine, considering a fictitious wind speed-dependent feed-in price. Our results for the turbine concept with an increase of rotor diameter by 36 % show that the revenue can be increased by 30 % and the cost of valued energy can be reduced by 16 % compared to the reference turbine.

Received: 05 Jul 2023 – Discussion started: 17 Jul 2023

Download & links

Daniel Ribnitzky et al.

Status: final response (author comments only)

RC1:
'Comment on wes-2023-72', Anonymous Referee #1, 01 Sep 2023
General comments
The manuscript addresses the philosophy and methodology for rotor re-design to achieve a turbine that is better suited for electricity markets with high wind-energy penetration. Subsequently, the methodology is applied, and the resulting example design is evaluated on main performance indicators. The research is well motivated and introduced, with a clear description of the objectives. The main design philosophy is clearly argued and described. However, the methodology has a few complicated aspects that are challenging to understand. Particularly the aspect of pitching and how that influences the design of the inner blade section requires very much attention from the reader to grasp and only emerges gradually throughout the story. Likewise, which variables are optimised and how is not described in one place. In my opinion, the manuscript would benefit from restructuring this, for which I have some suggestions below, under ‘Specific comments’. The results are interpreted fairly, with sufficient criticism, properly supporting the final conclusions.
On the principal criteria for WES publications, I would evaluate this manuscript with:
Scientific significance: Excellent

Scientific quality: Mostly excellent to good, and fair for the treatment of transition/pitch/optimisation

Presentation quality: Mostly good, for a challenging topic to explain, and again fair for the treatment of transition/pitch/ optimisation
Although a rather extensive section with specific comments follows, I would like to stress that I find the research very valuable and very well executed. I just want to share my ideas with the authors to stimulate them to see the work from a slightly different perspective. I’m happy with whichever way they use this information.
Specific comments
Design philosophy and methodology (of aerodynamic design and control)

I apologise up front for the lengthy discussion of this aspect. However, the authors know how many variables interact in the performance of a rotor, let alone in its design, and they have ample experience in trying to convey that to others. My struggle to provide clarity here will probably resemble theirs, so I hope this gives me some leniency.
There are a few aspects of the descriptions of the design that I found difficult to follow. For instance, several design choices are explained and motivated during the execution of the design activities, while they are already touched upon earlier in the description of the methodology. There turns out to be a strong relation between the final control philosophy and the aerodynamic design for low-TSR / strong winds. However, this control philosophy only becomes clear in section 3.2, while several references to its consequences are already used in the descriptions and clarifications in chapter 2 (e.g. lines 122-132, 152-160) and section 3.1 (e.g lines 221, 250-252).
Currently, the rotor design methodology starts with the principle of having three regions: a light wind / high TSR region, a strong wind / low TSR region and a peak shaving region (which is introduced on line 119, without explicitly describing how peak shaving is done). At this point, the reader perceives these regions as being fully separated. During the transition from light winds to strong winds the TSR and the induction drop instantaneously, so the RBM drops instantaneously as well. Therefore, pitching would only need to be applied at even stronger winds, when the (separate) peak-shaving region starts. This would allow for a straightforward design of the inner blade for zero pitch and at the optimal AoA, for low TSR. This is also how it is described in figure 1, right (apart from the dual goal for the induction factor).
However, the final control philosophy introduces a longer transition, due to the choice of keeping rotational speed constant in a transition region (rather than reducing it instantaneously). Consequently, the blade needs to be pitched in the transition region, and the pitch angle in the strong-wind region is no longer zero. Inherently, the transition region is extended by this up to wind speeds where peak shaving is needed. Therefore, there is no ‘clean’ strong-wind / low-TSR region, but this region is immediately combined with peak shaving. It also seems that the term peak shaving is often used loosely, to imply both control regions 2.2 and 2.3. As seen in figure 6, in the strong-wind / low-TSR region the blade is not at constant pitch, so there are no ‘unique’ design conditions for the aerodynamic design for this region (i.e. with constant TSR and constant pitch). These differences with the primary philosophy explained in relation to figure 1 where initially very confusing to me.
It stands to reason that an extended transition region is beneficial. Without it, the drop in TSR will be accompanied with a drop in BRM, but also a drop in power. Most likely, power can be maximised in a transition region, if the BRM remains at its constraint. This can be achieved with 1. constant speed and pitching (as chosen), 2. constant (zero) pitch and a gradual reduction in rotational speed, or 3. a combination of speed and pitch changes. Choosing for one of the first two (simpler) options is reasonable. Unfortunately, this kind of logic using the level of system parameters is not provided. Instead, the more complicated, implicit evaluation of the effect of speed and pitch on load distribution is given (later), leaving it up to the reader to judge if this achieves the desired global behaviour.
For better understanding of this approach, I recommend moving at least the top-left of figure 6, to section 2.1. This speed control is so straight-lined, that it seems to be more like a pre-meditated aspect of the design methodology than a consequence of the execution of a design iteration. This graph will help understanding of many aspects of section 2.1 that are currently unclear. Understandably, the authors did learn from their early design experiments for the tuning of this graph (such as the onset of the speed reduction at 15 m/s), but the same applies to the a-priory choice of TSR 9 and 11. I think it is also necessary to already explain the consequence of this speed control for the extent of the transition region, for pitch control, for the non-zero pitch of the low-TSR design and for the non-constant pitch in strong winds (during low-TSR operation in region 2.3). A schematised version of the bottom-left of figure 6 could be used for that as a qualitative pre-analysis. The shapes of the curve can easily be described with qualitative arguments for all regions. As a follow up of this description, it can then be clarified that the inner blade section is designed for a different/non-zero pitch angle and how that pitch angle will be determined during the design process. It would help to add this change in design pitch-angle to figure 1.
Although the previous description is reverse engineered from the manuscript, I’m fairly sure this captures (part of) the rationale of the authors. The concurrent change of TSR (for constant-speed operation) and pitch angle, will therefore naturally lead to the effects described in figures 4 and 5. As such, those graphs could support the choice for constant speed plus pitch increase, instead of constant pitch with speed reduction. However, the bottom-up approach that starts with the graphs in figures 4 and 5 and ends with exactly constant speed operation is neither convincing nor clear. If the authors agree with (some of) this analysis, I suggest that they restructure the story along similar lines of reasoning.
Up to this point, I agree with the overall philosophy for design and operation. On top of this, the authors introduce two aspects, which I’d like them to either reconsider, or support more clearly. These aspects are the twist offset towards stall for the inner blade and the dual goal for its induction factor (0.33 in high-TSR operation and 0.21 in low-TSR operation). I will start with the dual goal for induction, as this is easier to address. The principle of the design is to provide power by the outer blade section in light winds, to reduce loads on this section in strong winds and to let the inner section take over power production in strong winds. Power production in strong winds is considered to be important for offshore wind turbines, since these have a high probability of occurrence. Hence the interest of the authors in good peak-shaving performance. All these intentions, given by the authors, are counteracted by prioritising the induction factor optimisation of the inner blade for power production in light winds. As above, I would agree if the authors used an analysis of what happens to the induction factor as argument for the choice of a constant speed-increasing pitch transition: if the inner blade section is designed for an induction factor of 0.21 with a positive pitch angle, then it will have a higher induction factor at zero pitch and high TSR, which is a welcome advantage. This advantage is again a natural consequence of the pitch and TSR actions. The need to fine-tune this with a dedicated design for an induction factor of 0.33 in off-design operation is insufficiently argued.
Then, the twist offset. If I’m correct, this offset is relative to its optimal AoA in low-TSR operation, although that conflicts with the information in figure 1. Also here, in terms of design philosophy this doesn’t make sense in a first-order rationale: The primary goal for the inner blade is power production in strong winds, so a compromise of the design for low-TSR operation should be very strongly supported. Furthermore, I’d like to go into the description of the effects of the twist offset, that are used to argue its need. In much of the operational region of the aerofoils, the lift coefficient depends nearly linearly on the AoA. Thus, pitching is almost equally effective with and without the twist offset. Likewise, the effect of changing TSR on the change in lift coefficient over the blade span (relating to figure 4) is hardly dependent on a twist offset, since it is an effect on inflow angle: the change in AoA is not affected by an offset. Furthermore, for the system-level phenomena that are discussed, the optimality of the AoA hardly matters, so the offset of the twist is effectively relative to an arbitrary AoA. This also makes the discussion in lines 294-299 confusing or even misleading: Optimum AoA only tells something about the lift over drag ratio. For this special design and for the many off-design conditions (dual TSRs, transition pitching and peak-shaving pitching) it doesn't tell anything about the bigger picture for induction, power coefficient or thrust coefficient. Because of the operation at high TSR (with zero pitch) and at low TSR (with positive pitch), both the inner blade section and outer blade section have fundamentally two operating AoAs. I would be more concerned about how these two points are situated in the region between maximum lift coefficient/stall and minimum/negative lift. If the margins to those are good, then I would prioritise the optimum AoA under the principal design conditions (inner: low TSR – outer: high TSR) and not the off-design conditions (inner: high TSR).
I concur that the authors might prove to be correct in their arguments for deviation from the first design principles, to fine-tune the performance. However, there is so much going on, that I don’t think it helps understanding the fundamentals. Obviously, the design principles of the Hybrid-lambda rotor can be combined with other philosophies, such as induction reduction towards the tip. However, a separation of effect would be beneficial for obtaining better insights. Induction reduction is here primarily achieved by the Hybrid-lambda design, and secondarily by the inner section design adaptations. Possibly, the authors already have experimented with the straightforward design approach and have found it to lead to unacceptable behaviour. In that case, it would be helpful to describe that more explicitly.
Optimisation procedure

The design methodology (chapter 2) describes how the blade is designed for a particular rotor diameter and doesn’t describe if and how rotor diameter is optimised. Rotor diameter also doesn’t appear as design variable in the optimisation methodology (of section 2.3), where these variables are declared on lines 175-176. Perhaps what is described there is a nested optimisation (inner level), but that is not described. As it is, the value of 326 m for rotor diameter on line 216 comes out of the blue. Similarly, it isn’t clarified in the methodology how the spanwise transition point will be determined. The effect of both variables is discussed later (lines 254-280), which implies that they are also design variables (according to line 213). It would be helpful to know in advance how these design variables are incorporated in the methodology. Along similar lines, lines 78-79 describe that the objective function (implied: for rotor optimisation) is COVE. However, the optimisation of the tower is only described later. It is not clear whether this tower optimisation is included in a global exploration or nested optimisation in this optimisation of COVE.
On line 178, a stall margin is introduced as a constraint for the optimisation. It is not clear how this is implemented, since the aerodynamic design methodology doesn’t (explicitly) accommodate that. Line 180 states that this optimisation is done for a wind speed of 6.9 m/s, but it is not clear to the reader how this can be known. The wind speed at which the light-wind mode ends even seems to be a consequence of the optimisation itself, considering its dependence on rotor diameter.
All in all, I was somewhat confused about which aspects were optimised in a numerical optimisation, which aspects were determined in an analytic design approach and which aspects were designed with the authors in the loop. Relating to that, it wasn’t always clear in which order the various design variables were fixed. It would be helpful to clarify that in the beginning, perhaps with a flow chart of the entire process. In addition, it would help to categorise the variables in table 2 (fixed/chosen, design variables, properties, …). In the results, I propose to start with the discussion of rotor diameter and spanwise transition (lines 254-280), since these are two high-level system parameters.
Results

The design is assessed on AEP, revenue and COVE. Although the design is intended to advance from LCOE optimisation, it would be interesting to add how well the new design and reference perform on that metric. This would help understand to which extent the new design is a conventional improvement on LCOE, and which part can be attributed to the adaptation to the market conditions. This is similar to the comparison between AEP and revenue, which is already made. In addition, it might be useful to show and discuss some cost results separately, and not only hidden inside COVE.
Discussion

There are good messages in the discussion. I would recommend discussing only aspects that are closely related to the proposed concept and the results of this study. Adding other concepts/technologies (such as actuators and bend-twist coupling) is not specific to this concept (or at least it isn’t argued why a combination would be of more interest than for conventional designs). There are numerous other concepts that could otherwise be named as well.
In my opinion, the generalisation of the method to continuous variable-TSR operation (with variable spanwise induction optimisation) is the most interesting part of the discussion. It could be considered to dive a little deeper into this.
Conclusions

On line 710-713 you state that peak shaving is integrated into the design process. As you have seen in my earlier comments, I found this part somewhat confusing. I struggled with the use of the term peak shaving for both the transition region and for the conventional peak-shaving region. Furthermore, the bottom-up argumentation for the chosen control was difficult to follow. It didn’t give a reproducible procedure to merit the name ‘integration in the design process’. To claim this integration, I would like to see at least a stricter process for this particular part of the design approach, such as could be given with a flow chart, a formal optimisation problem description or graphs with dependencies on relevant design variables. As outlined above, in my opinion you provide arguments for a sensible choice of operation in the transition region, but that wouldn’t go as far as a design process. As it is, you only show one design point, with only circumstantial evidence that it provides superior performance thanks to the claimed mechanisms. Perhaps a similar combination of speed and pitch control can achieve similar performance for peak shaving with a conventional rotor design.
Having said that, the conclusions provide a concise overview of the relevant insights that have been achieved with this research.
Smaller comments about the content (in order of appearance)

On line 53 a similar design philosophy from Wobben is mentioned. This is later discussed on line 656, where it becomes clearer in which sense that philosophy differs. It could be useful to clarify this already in the introduction.

Line 64 (and many other places): The authors use ‘zero pitch’ for the operation of the blade at design conditions. This is implicitly defined on line 64. However, many blade designers and control designers define the structural twist with respect to zero twist at the tip and then use something like ‘fine pitch’ to get the design twist at the tip. Thus, this offsets the definition of pitch from the one used in this manuscript. It seems that even the authors confused themselves about this, since figure 6, bottom-left, shows a negative pitch angle for high-TSR operation. The chosen definition could be made more explicit (and used consistently).

Line 257-258: The sentence ‘If … reached’ is not so clear.

The authors claim on lines 319-320 that the reduced thrust coefficient leads to much lower wake losses. This cannot be known, since the effect of increased rotor diameter cannot be ignored. The increase in rotor diameter will extend the wake over longer distances and over a wider area. The next sentence implies that actually more momentum is taken from the wind.

Lines 353-362: Does Wisdem take the special care that is meant here? For instance, this region would experience stress concentration. Is that accounted for? Otherwise, the reduction in spar-cap thickness could be more related to model simplification than to optimisation.

Lines 388-390: This description is ambiguous. In region 2.3 the blade has variable pitch, so there is no unique c_P for this region. Could this be clarified?

Lines 400-402: It is described that a conventional look-up table was not found to perform sufficiently well. Could it be clarified whether this means that something else has been implemented? This seems to be the implication, since this section is about the controller design, and not about its evaluation. Therefore, this doesn’t seem to be simply an observation of performance, but a reason for change.

Lines 403-404: Could the authors explain what is meant by 'minimal' and 'reduce' compared to what? The previous descriptions of prescribed pitch do not seem to relate to the region where RBM load control is needed, or is it (dynamically)? The later text (lines 412-413) implies that ‘minimal’ refers to the steady-state pitch angle that was previously discussed. It would be helpful to get this information first. Having said that, lines 457-459 state that this controller is not used. Therefore, I would recommend removing this entire description of the (dynamic) load controller.

Lines 468-470: Are the ‘quasi-steady loads’ determined by dynamic simulation with uniform and constant wind speed? That is not the same as quasi-steady (even though the outcome might be similar). Could the procedure for this assessment be described with a little bit more detail?

Lines 478-480: This statement seems to contradict the earlier description. Does this only apply to the tip deflection? Why wouldn't the same argument apply to flapwise RBM and thrust?

Lines 487-488: This describe the normalisation of the loads. It doesn’t mention that a different normalisation is used for operational load cases and storm load cases. It would be useful to mention this up front, to avoid confusion with interpretation of the results later. This use two different normalisation values might even be reconsidered, even though I can see arguments for its use. Nevertheless, in the discussion and conclusions the authors now need to warn the reader that values for operational load cases and storm load cases cannot be compared directly. On line 678, they state that this is due to using relative values, but it is actually due to using different reference values for each.

Lines 500-501, 504-505, (680-682,) 734 and 738: It is stated that the increase in DLC 6.3 is significant compared to the reference turbine. If I’m correct, this is confusing if not misleading, since DLC 6.3 is not assessed for the reference wind turbine. After this observation of increased loading, it is nevertheless claimed that the slender blade design shows benefits (= load reduction?) in storm events. This is also confusing. Perhaps it is meant that the increase in loading is not as large as it would have been in case no slender blade design was used. However, this is not what is compared here (a Hybrid-lambda rotor and a conventionally upscaled rotor). Along similar lines, on line 738, it is concluded that the Hybrid-lambda rotor shows advantages in reducing loads. Especially here, out of context, this seems somewhat misleading. In absolute sense, the loads are not reduced. I probably agree with the point that might have been intended, if it is about combating the load increase with the design. Could this be rephrased?

Lines 522-525: I agree with the effect of the longer tower (higher lever arm, for almost equal thrust). However, the second argument seems flawed to me. Soft towers have a lower dynamic amplification factor for excitation frequencies that are above the natural frequency. They can have larger displacements with the same or even lower (internal) moments, which is why they are 'soft' (low stiffness). Thus, the effect of softness is more complicated and can go either way (depending on the excitation frequencies).

Lines 574-576: This statement seems in line with visual observations from the graph. However, the lever arm is increased for the Hybrid-lambda rotor, while it decreases for the reference turbine. Doesn’t that correspond to an increased contribution of the outer part?

Line 595: I suggest removing the reference to the aspect of market value here. At this point (the model for) market value is not yet introduced to the reader.

Line 611-612: To some extent the limitation of the flapwise RBM will oppose this effect of geometric scaling. Although I agree that the mass will increase stronger than for the Hybrid-lambda rotor, it doesn't seem fair to model the structure of the Hybrid-lambda rotor and only hypothesise for more conventional scaling. Furthermore, line 359 states that the mass of the new blade is only 14% lighter than that of a scaled blade. Is 14% considered to be ‘strongly increased’?

Figure 18: Why are results shown for the non-optimised tower? The optimised design seems to be the only sensible design, which fulfils the constraints with the actual (quasi-steady) loads.

Technical corrections
Overall: ‘Sec.’, ‘Sect.’ and ‘section’ are used, without consistency. Same for ‘Fig.’ and ‘Figure’.

Line 135: Considering line 175-176, probably 'adjusted' is meant here. 'Adopt' implies that it is kept the same (in dimensionless spanwise coordinates). Alternatively, it could have been meant that the same 'order' was adopted, instead of the distribution.

Line 170: The use of ‘maximum’ is confusing here (especially for a low-induction rotor, which doesn’t operate at maximum power coefficient in design conditions). Is it meant at TSR 11 (and at which pitch)?

Line 187: ‘choice for’ would be more appropriate than ‘assumption of’. The authors are not addressing an unknown aspect here.

Line 201: ‘planed’ -> ‘planned’.

Line 213 and 215: ‘blade design’ -> ‘aerodynamic blade design’.

Line 220-221: ‘which … moments’ would be more appropriate as an argument on line 151.

Line 273: Probably ‘that’ is meant, instead of ‘which’.

Line 228 (Heading section 3.3): It is not the ‘model’ that is designed and optimised.

Line 357: ‘up’ -> ‘down’.

Line 530: ‘The unsteady event [add: starts after 200 seconds and] lasts for 12 seconds, …’.
Citation: https://doi.org/10.5194/wes-2023-72-RC1
RC2:
'Comment on wes-2023-72', Anonymous Referee #2, 14 Sep 2023
The manuscript presents a design and optimization methodology for a novel wind turbine rotor concept the authors call ‘Hybrid-Lambda’. The work aims to a design rotor where (i) the outer part of the rotor is set to be optimal at low wind speeds operating at high TSR, and the inner part is designed for higher wind speeds at a lower TSR, and (ii) the increased loads are managed through a peak shaving controller close to rated conditions. The authors target to achieve this while constraining the mean blade flapwise bending moment loads below the max value of the reference turbine (IEA 15MW). As stated by the authors, the economic motivation for the design is to take advantage of energy pricing at low-wind conditions.
The work presented is scientifically significant and proves to challenge the conventional design of horizontal axis wind turbine rotors. The motivation and objectives of the work is presented clearly. But when presenting the methodology and results the ideas/concepts/fundamentals are difficult to follow. I do acknowledge that the body of work presented here is immense and there are a lot of moving parts to the novel rotor design. Light restructuring of concepts will help the readers appreciate the value of the manuscript. As an example, moving the controller strategy outlined in section 3.2 and figure 6 to line-125 would strengthen Section 2.
Overall, the manuscript is well structured and provides significant work that will be valuable to the broader wind energy community. Detailed comments and minor corrections are shared below:
Detailed Comments:
Section 1, line 58-59: Similarity to Wobben’s work is presented, but it is not clear how the current work differentiates from itself until section 4. Please include details on how this work sets itself apart from previous works in the introduction.

Section 1, line 63-64: I do not agree with the terminology “zero pitch” used in-lieu of “fine pitch”. Typically, the blade tip is set to a pitch angle of zero and is a reference orientation for the geometric twist of the blade. “Fine Pitch” is the additional pitch offset added during operation such that the tip of the blade is at the optimal design twist. Please make the necessary changes here and through the manuscript.

Section 2.1, line 119-120: The concept of peak-shaving is introduced but it is not clear what the procedure entails. Please provide a brief description.

Please comment on how peak shaving influences the design of the blade. Reading though the manuscript, it feels like a control strategy and not something influencing the aerodynamic design of the rotor.

Section 2.3: In this section the free variables are defined as chord, twist, radial airfoil positions, and spar cap thickness. But in section 2.1, the transition position and rotor radius are also discussed as design variables. Please clarify in the manuscript which variables are set/pre-determined and which ones are free variables.

Section 2.3: The load case for the optimization is defined at a wind speed of 6.9m/s, the following sentence on line 180-181 does not justify why this case was selected. If the rotor radius is a free parameter, then, the inflow for the load case is going to be a function of radius as the TSR is set to 11, this is confusing. How was this predetermined?

Section 2.3, line 204: OpenFAST provides a large set of options in its aerodynamic module AeroDyn. Please elaborate on what aerodynamic options were used when carrying out the aero-elastic simulations. Was it the same as the reference wind turbine? This will help guide discussing the load comparisons.

Section 3.1, line 216-217: It is not clear how the specific rating and rotor diameter is determined? Was it a design variable? If so, please define in Section 2.2/2.3. If not, please clarify on how this was determined.

Section 3.1, line 221-222: I find it difficult to follow the need for the twist offset in the inner section of the blade. The discussion related to this in previous and future sections feel fragmented. Please try re-organizing and better explain the need for the twist offset.

Section 3.1, line 254-269: This paragraph emphasis and extensively discusses the rotor radius as a varying parameter, this leads the reader to believe that it is a design parameter, but it has not been highlighted as such in Section 2.3.

Section 3.1, line 276: This is the first time the transition point for lambda is presented as a design choice and not a free variable. There are a lot of variables and moving parts in the optimization to follow. Presenting the optimization/design workflow in a flow diagram would help guide the reader through the whole optimization process better, in fact it will help the authors be more clear in their discussion of the optimization process. Using XDSM (eXtended Design Structure Matrix) might be a good approach.

Section 3.2, line 289: Presenting figure 6 in section 2.1, around line 125 would help the readers better understand the unique speed and pitch schedule, and peak shaving that is discussed extensively up until line 289.

Section 3.2: Please discuss the limitations of using BEM specifically for the hybrid-lamda rotor. Given the step change in induction at the 70% blade span. Does using higher fidelity method like free-vortex or CFD change the load distribution near the 70% blade span?

Section 3.3, Line 350: In addition to presenting the relative thickness and the spar-cap thickness, it would be valuable to compare the flapwise and edgewise stiffness, and mass distribution of the blade vs the IEA 15MW. The rapid transition in stiffness at the 70% location of the blade will be a point of concern especially for extreme loads. The optimization routine uses a steady inflow condition at relatively low wind speeds (as discussed in Section 2.3) this will not be representative of the stiffness distribution at the TSR transition region of the blade.

Section 3.3, Line 361-362: Using an exponent of 3 for geometrically scaling the reference blade for comparison is unfair. More recent publications (Griffith 2014, SNL100-03) have shown that the mass scaling exponent is realistically between 2.1 to 2.5.

Section 3.3, Line 366: What is the tower design driver for the IEA 15MW turbine? How does that contrast to the design driver for the current design? The reduction in tower diameter from 10m to 8.54m is significant especially given the 13% increase in blade mass (based on Line 362).

Section 3.4.1, Line 390: The equation is typically used for a constant Cp region. Since this value is not unique for the hybrid-lambda rotor how is the generator torque determined?

Section 3.4.2: Can you comment on the increased pitch activity due to the newer controller as compared to the reference? This will be important when determining the scaling of components (like pitch bearing/pitch actuator) costs for the final cost function.

Section 3.4.3: Does ‘quasi-steady loads’ refer to the loads experienced by the turbine due to steady inflow? If so please replace with ‘steady state loads’ or ‘steady-inflow loads’.

Section 3.4.3, Line 504-505: In storm cases, it is not only the slenderness of the blade that determines the load or reduction in loads. It is the complex interactions arising due to the blade geometrical twist, azimuthal angle, and yaw error that determines the loading of the turbine. Attributing the lower storm loads to planform area is assuming the inflow to the blades are primarily in at 90-deg to the airfoils, this is far from the case.

Section 3.5: Generally, any discussions regarding CapEx increases/decreases in components other than blade/rotor and tower are neglected. It will add value if the authors share why CapEx change of other components are significant (or not) to COVE.

Section 4: This section generally reads well.

Section 4, 656: The authors contrast their work to that of Wobben, please consider moving this discussion to the literature review to make a stronger argument about the novelty of the Hybrid-Lambda rotor.

Section 4, 665-666: what does “way more than 100m length” mean in this context? Is it a mis-phrased sentence?

Section 4, 685-670: Yes, I strongly agree with the authors the value of considering the torsional degree of freedom for the blade. Especially given its slender nature. Consequently, the aero-elastic stability of the blade will be interesting given how close to stall the inner part of the blade is at certain operational conditions.

Section 5, lines 710-712: After reading the paper it is not yet clear to me how the peak-shaving is integrated into the design process of the rotor, or how the aerodynamic parameters are influenced by it. The aforementioned flow diagram for the design/optimization process will help guide the reader to this conclusion.

Minor corrections:
Line 157: Citation for Buhl might be missing.

Line 170: The source code ….. as described in the following (sections).

Line 201: Typo, ‘planed’

Section 3 title: ‘Design and optimization of the blade structure’?

Line 451: avoid using the word ‘slight’ when discussing quantitative values like RBM.

Line 593: ‘Figure’ is used to reference figure 17, whereas in the previous sections ‘Fig. XX’ has been used. Please maintain consistency.

Line 599: ‘Sect. 1’ is used to refer to a Section, whereas ‘Sec. XX’ was used previously. It is clear that different authors have contributed to the sections, hence the change in style, but please maintain constancy throughout the manuscript as it is a single body of work.
Citation: https://doi.org/10.5194/wes-2023-72-RC2

Daniel Ribnitzky et al.

Viewed

Total article views: 446 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
352	82	12	446	9	6

HTML: 352
PDF: 82
XML: 12
Total: 446
BibTeX: 9
EndNote: 6

Views and downloads (calculated since 17 Jul 2023)

Month	HTML	PDF	XML	Total
Jul 2023	111	42	4	157
Aug 2023	129	21	2	152
Sep 2023	112	19	6	137

Cumulative views and downloads (calculated since 17 Jul 2023)

Month	HTML	PDF	XML	Total
Jul 2023	111	42	4	157
Aug 2023	129	21	2	152
Sep 2023	112	19	6	137

Viewed (geographical distribution)

Total article views: 431 (including HTML, PDF, and XML) Thereof 431 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 22 Sep 2023

Short summary

This paper provides an innovative blade design methodology for offshore wind turbines with very large rotors compared to their rated power, which are tailored for an increased power feed-in at low wind speeds. Rather than designing the blade for one single optimized operational point, we include the application of peak-shaving in the design process and introduce a design for two TSRs. We describe how the enlargement of the rotor diameter can be realized to improve the value of wind power.


Total:	0
HTML:	0
PDF:	0
XML:	0