the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Feb 2025
Airborne wind energy system test bench electrical emulator
Abstract. Airborne Wind Energy Systems (AWES) offer a promising alternative to conventional wind turbines, but their commercialization is hindered by challenges in efficiently converting the highly dynamic mechanical power of tethered flight into stable electrical energy. While extensive research has focused on optimizing AWES flight trajectories and control strategies, the power conversion stage, which is critical for integrating AWES into electrical grids, is relatively under-researched. To bridge this gap, reliable and flexible electrical test bench emulators are needed to replicate AWES dynamics under controlled conditions, enabling systematic evaluation and optimization of power electronics and control strategies.
This paper presents a validated electrical test bench emulator and a torque ripple-optimized Model Predictive Control (MPC) strategy designed to enhance the performance of AWES ground station generators. The proposed emulator accurately reproduces the mechanical-electrical interactions of a real AWES by simulating the variable tether forces and reeling dynamics encountered during optimal crosswind flight. Two electrical topologies are introduced: a separated DC bus configuration that closely mimics real AWES energy storage dynamics and a common DC bus topology that minimizes battery requirements for extended control testing. The proposed MPC strategy ensures precise generator speed and torque regulation, achieving less than 1 % root mean square error (RMSE) in torque tracking while optimizing energy efficiency.
Using experimental flight data, the test bench demonstrates an overall energy efficiency exceeding 80 % and peak conversion efficiencies up to 93 %, with Permanent Magnet Synchronous Generators (PMSG) outperforming Induction Machines (IM) by 2–6 % in instantaneous efficiency. These findings establish electrical test benches as essential for AWES development, offering a scalable platform for optimizing power conversion and control, which advances AWES as a viable renewable energy technology.
- Preprint
(2289 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2025-19', Anonymous Referee #1, 29 May 2025
Reviewer remarks:
The text is well written. The figures are clear.
The authors have referred to the state of the art. However, the introduction could be strengthened by explicitly highlighting how this work significantly advances the field beyond existing emulators and control strategies. (Emphasizing the specific limitations of previous emulators, mention unique benefits of the proposed approach). Perhaps a sub-section in the introduction which addresses the novelty, scope and limitations of the presented work.
Furthermore on scope, it will be helpful in knowing what part was modelled, what part was measured from experiments, and what information was taken from another source.
The authors chose PMSG and IM for the case study, are these and the parameters for the emulator representative of the state of the art AWES prototypes? The AWES schematic shown in Fig.1 is specific to the presented application? Maybe first define or illustrate the state of the AWES systems and its main components and operational framework and then show the presented AWES system and emulator.In my opinion, adding a flow chart in the methodology section showing the overall procedure and work flow of the presented work will be of great help. Example showing the models that were developed, the experimental data, load estimation, validation cases, results, error comparison.... Main idea is for the reader to get a complete understanding of what was done, and how was it done without reading the text.
The experimental validation details need more detail. The authors only gave reference to a previous study from where the flight time series and data were obtained. More detail on the experimental setup should be provided.
In the discussion or conclusion section, the authors are requested to highlight limitations of the presented emulator and control strategy. For example, how do the assumptions taken for estimating the torque loads hold in real-world application.
Minor remarks regarding acronyms, they should be defined when first used. Some acronyms were defined multiple times (example RMSE and PMSG).
Citation: https://doi.org/10.5194/wes-2025-19-RC1 -
RC2: 'Comment on wes-2025-19', Anonymous Referee #2, 04 Jun 2025
The article presents research on the development of a test bench designed to emulate airborne wind energy (AWE) systems. The introduction is well-written, providing a solid review of relevant literature and a clear explanation of the objectives and scope of the research. It demonstrates that the authors possess a strong understanding of the field.
The research methodology is described in detail, and additional technical specifications for the control and emulator systems are appropriately included in the appendices. Overall, the paper is a valuable contribution and is suitable for publication after addressing the following concerns:
Controller Benchmarking:
The authors justify the selection of model predictive control (MPC) by referencing the fast-changing dynamics of AWE systems. While the MPC performance appears promising, it is essential to include a comparative analysis with at least one other control strategy, such as a conventional PID or another baseline controller, to validate the superiority of MPC in this context.Novelty of the Test Rig:
The paper lacks a clear articulation of the novelty of the proposed test bench. The authors should elaborate on how their emulator differs from and improves upon existing AWE test platforms discussed in the literature.Validation with Real Data:
Although the authors use real AWE test data to validate their numerical model, relying on a single data set may not be sufficient. Additional validation using at least one more independent AWE operation dataset would strengthen the credibility of the model's performance.Physical Implementation:
The paper does not clarify why only a numerical/simulation model was developed and no physical test bench was built. The authors should justify this decision and discuss the implications for the practical application of their work.DC Bus Power Observation:
In Figure 9, the power at the common DC bus is nearly zero throughout the operation. This suggests that most of the generated power is consumed internally for kite emulation, which significantly diverges from real-world AWE system performance. If grid integration is planned for future work, this could pose a challenge, as the generated power would need to be dispatched to the grid rather than being fully consumed by the emulator. The authors should clarify the rationale for selecting a common DC bus topology and discuss its practical implications.Topological Comparison:
A more detailed comparative analysis between the common DC bus and separate DC bus topologies is recommended. This would provide deeper insights into the trade-offs and performance implications of each configuration.Enhancement of Results Section:
The results section would benefit from additional data and analysis that reflect the above concerns. This includes more extensive validation, clearer discussion of control performance, and practical considerations related to topology and physical implementation.Citation: https://doi.org/10.5194/wes-2025-19-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 165 | 52 | 16 | 233 | 17 | 18 |
- HTML: 165
- PDF: 52
- XML: 16
- Total: 233
- BibTeX: 17
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1