A Spectral Model of Grid Frequency for Assessing the Impact of Inertia Response on Wind Turbine Dynamics

Round 1

Reviewer 1 Report

The paper presents a spectral model of grid frequency for studying the impact of inertia response on the wind turbine dynamics. The manuscript is interesting and well written.

On line 84, please correct section 2 instead of section 1 (data source and statistical analysis).

How was tuned the PI controller for the blade pitch control?

The case study section is well presented and commented.

The conclusions must be reformulated for a better highlighting of the authors contributions versus other similar studies.

Minor English language improvements should be done.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The Authors come up with a very interesting yet useful research on inertial response assessment of wind turbines based upon grid frequency simulation through spectral model. The research subject is a current topic and it deserves publication.

There are some minor flaws however, that the Authors might want to check and/or explain, just in the benefit of an improved document to be a reference for other interested readers. There they go:

• The Authors might want to go back over the manuscript to take a final English style and grammar minor corrections: to avoid long sentences if possible, to clarify the meaning of certain sentences (Line 29 "Balancing the future power system with higher share of whether-depend renewable generations is challenging, since they not only draw more uncertainties in the stochastic generations but also reduce system inertia by replacing conventional synchronous generators (SGs)"), to take a look at minor typos here and there (for example: Line 41 "...can be considered as be...", Line 406 "However, an significant increase..."), etc.
• The SI acronym (Line 46) appears just before the term "sythetic inertia" is introduced (Line 81). It would be desirable to have the acronyms defined in their corresponding text lines. In any case, the acronyms list at the end of the manuscript is of course greatly appreciated. 
• In opinion of this reviewer, the bulk concept of the document and its foundation is the SI and how it can help to achieve a better performance. Indeed the proposed spectral model works in that direction. For that reason, and accounting for the fact that Section 2 jumps straight into the spectral rationale of the research, it would be desirable if the paragraph starting in Line 33 be more explanatory. Or if preferred by the Authors, a paragraph at the very beginning of Section 2 might be introduced, clearly stating how the SI works, why the interest on it and how to deal with it through the spectral model.
• While it is clear the effectiveness of the model in terms of spectral performance, as deduced from Figure 11, at the sight of the RoCoF vs. time and mostly at the sight of frequency vs. time, this reviewer wonders if the variability of frequency with time (generated) really addresses the measured one. One thing is that the spectral pattern is correctly simulated, and a different thing is that the high and not-so-high frequency events occur under the same pattern. That time distribution in turns affect (or might affect) the in-time response depending on the nature and distribution of previous responses (just like an infinite impulse response filter, so to speak). Could it be possible to clarify that point?
• Related with filters and spectral analysis, have the Authors considered de viability of using multi-resolution analysis (wavelets) to extract very different time-variability patterns from the interested signals? This reviewer has long experience on that kind of analysis and it could provide with good results too (as a pre-analysis, instead of standard spectral analysis). Perhaps the Authors might want to include some comment (probably the Conclusions section would be the right place) bringing to front the efficiency of the proposed methodology with respect to MRA.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Authors present the results of study that deals with a spectral model of grid frequency for assessing the impact of inertia response on wind turbine dynamics. Article and its results are based on assumptions and on proposed scenarios. Some minor errors need to be modified in paper.

Row …, Page …: Add missing units for quantities.
Row n, Page n: Correct some typing errors: “whether”, …
Row 113, Page 3: Missing indication of IEC standard (also in references).
Row 122, Page 4 (Figure 2): Why is there peak value of time lag at value 60 (min)?
Row …, Page …: There are some mismatches because authors have in introduction and conclusion results of 50 Hz and 60 Hz systems; so what was the real frequency that they were deal with it?
Row 148, Page 5: Is it effective to use 1-day data block and not 1-month or 1-year continuous data block?
Row 166, Page 5: Is the frequency deviation of 0,1 Hz in accordance with standard EN 50160?
Row 331, Page 14: Why authors simulated wind turbine only in 600 s time interval?
Row 358, Page 15: Why the generator speed is approximately 5 seconds faster as the rotor effective wind speed? Is it correct?

The positive benefit of the paper is good introduction to assessment of the impact of synthetic inertia response on wind turbines dynamics.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Thanks for the answers to this reviewer's comments.