Synthetic PMU Data Generator for Smart Grids Analytics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper a synthetic PMU generator has been proposed to be used in smart grids. The performance of the proposed PMU generator has been analyzed under the conditions specified in the Standard IEC 60255-118-1 for the P-class PMU performance and different real datasets.

 I have the following remarks on this work:

 1. The accuracy of the proposed PMU generator should be analyzed also in the case of missing data.

 2. The minimum number of nominal cycles required by the proposed PMU generator to comply with the Standard IEC 60255-118-1 for the P-class PMU performance should be specified.

 3. For the performed tests it should be specified the values used for the thresholds theta_SNR and theta_OoB. Also, it should be specified how the above thresholds should be determined in a given application.

 4. The reference(s) for the TFM algorithm used by the proposed PMU generator should be given. 

Author Response

Reviewer #1

In this paper a synthetic PMU generator has been proposed to be used in smart grids. The performance of the proposed PMU generator has been analyzed under the conditions specified in the Standard IEC 60255-118-1 for the P-class PMU performance and different real datasets.

We thank the Reviewer for the insightful comments and advices. In the following, the specific remarks are addressed and the corresponding modifications in the manuscript have been highlighted in BLUE.

 I have the following remarks on this work:

1. The accuracy of the proposed PMU generator should be analyzed also in the case of missing data.

We thank the Reviewer for the interesting remark. This is indeed a very important functionality for an easier and wider portability of the proposed PMU synthetic data generator on real-world datasets that may include interruptions in the data stream or inconsistent time-stamps. In the revised manuscript, we propose an interpolation technique to substitute the missing data. In any case, the corresponding PMU synthetic data are flagged as derived from interpolated data. Thank you for the opportunity to include a new functionality in the generator!

2. The minimum number of nominal cycles required by the proposed PMU generator to comply with the Standard IEC 60255-118-1 for the P-class PMU performance should be specified.

We do see the Reviewer’s point. The number of nominal cycles is fixed to 4, as this has been proven to be a good trade-off between estimation accuracy and responsiveness in the presence of transients or fast dynamics. In the revised manuscript, we clarified this aspect and we explained how to test it with shorter window lengths (e.g. 3 or 2 nominal cycles). This later solution comes with some limitations in terms of estimation accuracy and capability of detecting Out-of-Band components. For this reason, a warning is included if this processing approach is employed.

3. For the performed tests it should be specified the values used for the thresholds theta_SNR and theta_OoB. Also, it should be specified how the above thresholds should be determined in a given application.

We thank the Reviewer for pointing out an aspect that may have been misleading in the previous version of the manuscript. In the revised version of the manuscript, we clarified which value of the two thresholds was used to obtain the presented results. The section “Parameter setting” is intended to provide indications on how to carry out a sensitivity analysis in the specific application scenario under analysis.

4. The reference(s) for the TFM algorithm used by the proposed PMU generator should be given. 

Thank you for the suggestion. We added the reference to a similar implementation of TFM algorithm for PMUs. It is important to note that the PMU synthetic data generator applies a slightly different approach from the provided reference. Nevertheless, the measurement principle and the theoretical background are the same and this reference makes the present manuscript much more complete. Thank you!

Reviewer 2 Report

Comments and Suggestions for Authors

I have reviewed your manuscript titled "SYNTHETIC PMU DATA GENERATOR FOR SMART GRIDS ANALYTICS" and found it to be a well-written and technically sound contribution to the field of smart grid analytics. However, to further strengthen the paper and ensure its clarity and impact, I would like to suggest the following revisions and questions for your consideration:

1. The introduction effectively outlines the issue of PMU data scarcity in smart grids, but after presenting the background, it is recommended to more clearly articulate the research objectives. Summarizing the innovations of the synthetic PMU data generator in the introduction would help readers quickly grasp the core contributions.

2. Although the paper describes the synthetic data generation process in detail, it lacks a more concrete description of its application in real-world grid scenarios. It is recommended to include specific case studies on how synthetic data can support real-world grid operations, such as fault detection and load balancing, to enhance the practical relevance and persuasiveness of the article.

3. The paper introduces the algorithm for the synthetic PMU data generator, but it is suggested to provide more detail on how the input parameters affect output accuracy, especially in different grid scenarios.

4.The conclusion section of the paper, while mentioning future work directions, lacks specific details. It is suggested to further clarify the specific implementation paths and challenges for future work.

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Reviewer #2

I have reviewed your manuscript titled "SYNTHETIC PMU DATA GENERATOR FOR SMART GRIDS ANALYTICS" and found it to be a well-written and technically sound contribution to the field of smart grid analytics. However, to further strengthen the paper and ensure its clarity and impact, I would like to suggest the following revisions and questions for your consideration:

We thank the Reviewer for the insightful comments and advices. In the following, the specific remarks are addressed and the corresponding modifications in the manuscript have been highlighted in BLUE.

1. The introduction effectively outlines the issue of PMU data scarcity in smart grids, but after presenting the background, it is recommended to more clearly articulate the research objectives. Summarizing the innovations of the synthetic PMU data generator in the introduction would help readers quickly grasp the core contributions.

We thank the Reviewer for the insightful and helpful advice. We revised the end of Introduction to underline the novelty points of the PMU synthetic data generator.

2. Although the paper describes the synthetic data generation process in detail, it lacks a more concrete description of its application in real-world grid scenarios. It is recommended to include specific case studies on how synthetic data can support real-world grid operations, such as fault detection and load balancing, to enhance the practical relevance and persuasiveness of the article.

We thank the Reviewer for pointing out an aspect that may have been unclear or not properly addressed in the previous version of the manuscript. One of the main applications would be the generation of synthetic yet realistic datasets of PMU measurements to test machine learning and big data applications (e.g., state estimators, fault locators). We better clarified this aspect and we provided an example of a statistical distribution of synthetic data as derived from a real dataset.

3. The paper introduces the algorithm for the synthetic PMU data generator, but it is suggested to provide more detail on how the input parameters affect output accuracy, especially in different grid scenarios.

We do see the Reviewer’s point. In the revised manuscript, we better clarified the selection of the input parameters. The estimation accuracy is mostly dependent on the level of noise and on the dynamics on-going. An example of synthetic data distributions is given to provide an example of the possible application of the PMU data generator. Thank you for the advice that makes much clearer the potential benefits of the proposed generator!

4.The conclusion section of the paper, while mentioning future work directions, lacks specific details. It is suggested to further clarify the specific implementation paths and challenges for future work.

We understand the Reviewer’s point. We revised the Conclusions to reflect more practical aspects of the future development of this research activity, with a specific focus on the improvement of the synthetic dataset generation.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have only the following minor remark:

1. Line 609, it should be written

"Fig. 8 provides..."

instead of

"Fig. provides..." 

 

Author Response

Comment 1: I have only the following minor remark: Line 609, it should be written "Fig. 8 provides..." instead of "Fig. provides..."

Reply: We thank the Reviewer for the careful reading. This was indeed a typo. We included the cross-reference in the final manuscript. Thank you!