the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Jan 2025
Non-linear interaction between synchronous generator and GFM controlled wind turbines – Inertial effect enhancement and oscillations mitigation
Abstract. The integration of grid-forming (GFM) controlled wind turbines into AC grids introduces complex dynamic interactions that significantly influence its behavior on the AC side. This study explores the nonlinear coupling between wind turbines and AC grids and propose strategies for the enhancement of the inertial effect and the mitigation of oscillations which can arise in case of an AC event. A simplified synthetic model is developed to elucidate the physical insights of these interactions. The findings reveal that wind turbine dynamics has an impact on the inertial contribution and introduce oscillatory behavior under certain conditions. Advanced control strategies are then proposed. They include the integration of input shaping filters and lead-lag compensation to optimize inertial response and dampen mechanical oscillations. The theoretical analysis, validated through simulation, demonstrates the effectiveness and limitations of these methods in enhancing the AC side behavior without compromising the performance of the mechanical system.
- Preprint
(3325 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2025-10', Anonymous Referee #1, 22 Apr 2025
Explain why there might be a rate of change of frequency limit? What if the value is too high?
Looks as if a wind turbine power curve has been used rather than a wind farm power curve. Something that should at least be noted in the paper
Section 2 of the paper could be shortened, since there is lots of familiar material here
What is the electrical distance between the synchronous machine, load and wind farm, and how might this affect the results, and the parameter selection process?
“hit and trial” should read as “trial and error”. Can the authors recommend a better approach than “trial and error” to select the parameters, particularly for application to a real system, and where the system configuration, demand levels, etc. may vary over time?
Fig. 3b – “region” and “zone” terms are both used interchangeably – best to stick with one term
Which tool is used to run the EMT simulations?
Fig. 7c and 8b show noticeable differences between the different approaches, but the text doesn’t clearly explain why
How should a zero vibration filter be designed?
How are H(GFM) and D(GFM) chosen? Could they become variables during transients?
One reason to provide an inertial (or fast frequency) response is to obtain a revenue stream by satisfying the particular system service definition. The authors may want to quantify the magnitude and duration of the power injection phase, and the duration and depth of the recovery phase after the frequency nadir when evaluating parameter choices. The authors should look at fast frequency reserve definitions in different countries.
How should the time delay, td, be best chosen, and would it need to vary with system loading conditions, wind turbine (farm) output, share of generation from renewable (converter-based) sources, etc.? It is not clear that the preferred parameter values given in the paper are robust for a much wider range of system conditions, and for other systems.
What would happen to the frequency response if the wind share was increased, larger load disturbance, weaker grid connection, fewer synchronous machines online, etc. ? What would be the implications for controller configuration and parameter selection?
Various time delays are considered, but how robust are the choices against changes in system conditions, demand levels, generator locations, grid strength, renewable share of demand, …?
Be careful when using the term “optimise”. Are the presented results “optimal”, or just “better”? It looks as if they are simply “better”
Figure 15 – are the authors stating that an input shaping filter is not applied when the turbine is operating in zone 2? What happens if the input wind speed is such that the turbine varies continuously between zone 1 and zone 2?
Appendix A – 50 Hz and 60 Hz are both mentioned – what is happening here?
English is rather clunky throughout the paper. It would be helpful to tidy up the English, and to expand upon the explanations of what the results mean, and why they are considered significant.
Define all acronyms
Kirchhoff is spelt like so
From a presentational perspective, the results figures should be placed much closer to the relevant text. For example, at the moment, the figures associated with Section 3 actually appear in Section 4. Consequently, the second half of the paper is not easy to follow, with constant switching between “text” pages and much later “figure” pages.
Harvard referencing uses 2 formats, e.g. Wagh et al. (2025) or (Wagh et al., 2025). The first format should be used if the reference forms an “active” part of the sentence, and the second format should be used if the reference doesn’t form an active part of the sentence. There are many places in the paper where the reference formatting is incorrect.
Citation: https://doi.org/10.5194/wes-2025-10-RC1 -
RC2: 'Comment on wes-2025-10', Anonymous Referee #2, 10 Jul 2025
The topic of this paper is not at the core of my expertise, so it is very difficult for me to evaluate the contribution of this paper.
However, I think as a reviewer, I can pose some useful questions and make some useful comments.
- I would appreciate it if the authors make more clear what they see as the innovative contribution(s) of the paper.
- Some of the values and equations seem to be arbitrary: Where does equation (2) come from? Are the values for K_dSM of 10 and 30 pu based on trial and error?
- What is an AC event (abstract)?
- Could delta omega in equation (6) be defined?
- Is the wind speed for all wind turbines assumed to be constant? I would assume so, because besides the load change, everything seems to be constant.
- Is it correct that in the end the only effect of the load change of 3% or 20% is that the synchronous generator in the conventional power station increases its load?
- In the graphs in figure 7 and 8, I see that the frequency decreases, but I do not see that the power generated by the synchronous generator increases. Why (not)?
- The paper suggests some solutions for the problems. Are these the only possible solutions, the best solutions, or just solutions that (accidentally) happen to work?
- In parts of Europe, there are times that significantly more than half of the generation comes from Inverter Based Resources (IRBs). Therefore, it seems that this research is describing things that are already used in the industry. Is that correct? Does this mean that science is lagging with the application? Or do we apply things we do not yet sufficiently understand?
- Do the problems (the oscillatory behavior) described in the paper also show up in practice? If yes, comparison with experimental results would be a valuable form of experimental validation of the work. If no, what does that say about this paper?
- If we want to go through energy transition, the challenge is not to keep a stable power system while there is always enough reserve power form thermal power stations, but to keep a stable power system fully based on renewable energy (IRBs) with all the intermittency, so that is where I see the real challenges.
Citation: https://doi.org/10.5194/wes-2025-10-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 182 | 72 | 17 | 271 | 14 | 20 |
- HTML: 182
- PDF: 72
- XML: 17
- Total: 271
- BibTeX: 14
- EndNote: 20
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1