the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Identification of electro-mechanical interactions in wind turbines
Abstract. Large direct drive wind turbines with a multi-megawatt power rating face design challenges due to scaling laws for high-torque generators. This work proposes to extend the design space by moving towards a more system-oriented approach. This requires an extension of the state-of-the-art wind turbine models with additional degrees of freedom to include electro-mechanical interactions. To limit the computational effort of such models, a profound understanding of possible interaction mechanisms is required. This work aims to identify interactions of an additional degree of freedom in the radial direction of the generator with the wind turbine structure, the aerodynamics and the wind turbine controller. Therefore, a Simpack model of the IEA 15MW RWT is implemented and coupled to a quasi-static analytical generator model for electromagnetic forces. The analytical model, sourced from literature, is code-to-code validated against a finite element model of the generator in Comsol Multiphysics. Electro-mechanical simulation results do not show interactions with the aerodynamics or the controller. However, interactions with the wind turbine structure occur. It is shown that the modelling approach can affect the system's natural frequencies, which can potentially impact the overall system design choices.
- Preprint
(1467 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2024-13', Anonymous Referee #1, 16 Feb 2024
Please consider the following comments:
- Is the main contribution related to eccentricity modelling when it comes to electrical-mechanical model? Then it should be clearer in the title or abstract.
- There are different types of eccentricity - static, dynamic, and mixed. These exist also cases with axial misalignment. It seems that static eccentricity is the topic of the paper. Please elaborate.
- More details regarding the generator characteristics should be included (e.g. pole and slot number).
- Numerical model should be presented with more details as well. Is model 2D or 3D? Please include a cross-section of the model
- Please show where z-axis and y-axis are. It was not clear for me. Normally (in electrical engineering literature) net force in eccentricity cases is characterized with Fx and Fy, assuming that airgap is uniformed in axial (z direction). It seems that in this paper the axis definitions are different. Please elaborate.
Citation: https://doi.org/10.5194/wes-2024-13-RC1 -
RC2: 'Comment on wes-2024-13', Anonymous Referee #2, 19 Mar 2024
- Line 15: Not true, especially onshore where MW/m^2 is decreased for low wind speed turbines
- Line 88: A source or a figure would be good for a proof and how it was done
- Line 130: How did you avoid overconstraining the system with two rigid axial constraints?
- Line 139: Is the drivetrain tilt considered here?
- Line 195: Please elaborate why no axial slicing was done and thus the effect of rotor tilting on the airgap was neglegted
- Line 239: Which loadcase was used here?
- Line 325: Doesn't this rather reveal the necessity to consider ambient stiffnesses in WTs due to their complex dynamics? How does it change without the electromagnetic radial forces?
- Line 385: Which load case was chosen here? Were tangential emag loads considered in both models?
- General: How are rotorblades and tower modeled? What was the cut off frequency for eigenmodes?
- General: If the outcome was that the additional DOF is mainly responsible for a change in the coupled dynamics, then the effect of the axial DOF on the WT dynamics need to be studied as well
Citation: https://doi.org/10.5194/wes-2024-13-RC2 -
RC3: 'Comment on wes-2024-13', Anonymous Referee #3, 25 Mar 2024
General comment:
The main objective of the paper is to present a physical mechanism behind electro-mechanical interactions in WTs. The paper described the modeling approach and setting in detail. There are general notes as follows:
- However, the paper claimed that the Simpack model was validated against OpenFAST; no references or result comparisons are presented.
- It is not clear what the bearing arrangement is in this study. The paper just mentioned the terms fixed and floating bearings without any more information. More detail on the arrangement and specification of the bearings is needed.
- A sensitivity analysis of the bearing stiffness against the response of the structure is needed.
- Please consider the definition of the turbine and floater as a table in Wind Turbine Section 2.1. The coordinate system should be included too.
- Please add the definition of the load cases in each section.
- It seems none of the cases are real DLCs in Section 3. In order to have a better understanding of this added DOF, it’s better to consider a DLC with NTM (wind) and NSS (wave) and study the load on the generator or bearings.
- The work is limited to a very specific condition in the generator. It is recommended to include this limitation in the title of the paper.
Specific comment:
- In Section 2.1.1, line 139, the maximum allowable radial eccentricity of the generator is assumed to be 2 mm according to the design. Please provide references for this assumption. Keeping a 2 mm radial distance for a 10 m generator needs reference.
- In Section 2.1.1, line 140, it is assumed that gravity loading should only cause a maximum of 10% of the allowed eccentricity. It is not clear how this assumption was validated, as this assumption is fundamental to the study. Further investigation in this regard or acquiring other references is needed.
- In Section 2.1.1, line 141, it is assumed that the bearings have the same stiffness. The internal and type of the bearings are not the same; therefore, the assumption is not correct.
- In Section 2.2.2, line 222, it is stated that the coupling of the FEM and WT models proved to be too computationally expensive. It is better to present in a paragraph the quantitative cost of the FEM and WT models and compare the differences.
- In Figure 10 (b, d, f), it is not easy to notice different graphs. However, the plots show the similarity of the curves; using markers could be helpful.
Citation: https://doi.org/10.5194/wes-2024-13-RC3 -
AC1: 'Comment on wes-2024-13', Fiona Luedecke, 06 Apr 2024
We thank all reviewers for their constructive comments that helped improve the quality of the manuscript and appreciate the time and effort they put into this. Please find attached our answers to the comments on a point-by-point basis. On behalf of the authors, Fiona D. Lüdecke
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
378 | 58 | 12 | 448 | 8 | 4 |
- HTML: 378
- PDF: 58
- XML: 12
- Total: 448
- BibTeX: 8
- EndNote: 4
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1