Artificial Intelligence Driven Smart Hierarchical Control for Micro Grids—A Comprehensive Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents a comprehensive review of Artificial Intelligence (AI) techniques applied to the hierarchical control structure of microgrids. It covers the transition to renewable energy, the classification of control levels (primary, secondary, tertiary), and details various computational intelligence methods—such as Machine Learning and Fuzzy Logic—to enhance system stability and economic management. In my opinion, the article needs major revisions before being published.

1. The authors are strongly advised to undertake a thorough proofreading of the manuscript to address semantic and formatting errors. Specifically, they should substitute incorrect phrasing such as "In controversy" (line 556) with "In contrast" and "extinct problem" (line 32) with "resource depletion".
2. Instead of describing each study sequentially, it is suggested that the authors group the reviewed research by specific themes or technical challenges solved. This approach will significantly improve the narrative flow, preventing the section from resembling a mere list of references.
3. The manuscript currently includes tables that summarize the reviewed literature (listing references, techniques, and applications), the authors should go a step further by including tables that critically compare the performance of these AI techniques against each other. It would be highly beneficial to include a comparative analysis of factors such as computational burden, convergence speed, robustness against uncertainty, or implementation complexity (e.g., comparing ANN vs. Fuzzy Logic), rather than limiting the tables to a bibliographic inventory.
4. It is recommended that the authors include a brief discussion regarding the applicability of these techniques beyond simulation environments. Clarifying which methods have been validated on physical hardware and which remain purely theoretical would add significant value to the review.
5. The authors should condense the elementary definitions of Artificial Intelligence and focus more intensively on why and how these specific architectures are advantageous for energy management and control within microgrids.

Author Response

Reviewer 1:

 Comments and Suggestions for Authors

Comments 1: The manuscript presents a comprehensive review of Artificial Intelligence (AI) techniques applied to the hierarchical control structure of microgrids. It covers the transition to renewable energy, the classification of control levels (primary, secondary, tertiary), and details various computational intelligence methods—such as Machine Learning and Fuzzy Logic—to enhance system stability and economic management. In my opinion, the article needs major revisions before being published.

The authors are strongly advised to undertake a thorough proofreading of the manuscript to address semantic and formatting errors. Specifically, they should substitute incorrect phrasing such as "In controversy" (line 556) with "In contrast" and "extinct problem" (line 32) with "resource depletion".

Response: With reference to this comment, we have modified those lines and removed the words controversy and extinct. Also in Pg. 18, similar modification is done.

Comments 2: Instead of describing each study sequentially, it is suggested that the authors group the reviewed research by specific themes or technical challenges solved. This approach will significantly improve the narrative flow, preventing the section from resembling a mere list of references.

Thanks for the suggestions.  As the review is grouped with the specific theme of various hierarchical control levels, ie Primary, secondary and tertiary.  Each of these themes is analyzed under the sub-theme of application of Fuzzy logic and Artificial intelligent techniques for the same.  However, we have not grouped this under the technical challenges because each of the method involves different challenges and hence the groupings are retained under common challenges.

Comments 3: The manuscript currently includes tables that summarize the reviewed literature (listing references, techniques, and applications), the authors should go a step further by including tables that critically compare the performance of these AI techniques against each other. It would be highly beneficial to include a comparative analysis of factors such as computational burden, convergence speed, robustness against uncertainty, or implementation complexity (e.g., comparing ANN vs. Fuzzy Logic), rather than limiting the tables to a bibliographic inventory.

Thanks for the wonderful suggestion.  Based on this suggestion, Table 1 is newly added in Page 6.

Comments 4: It is recommended that the authors include a brief discussion regarding the applicability of these techniques beyond simulation environments. Clarifying which methods have been validated on physical hardware and which remain purely theoretical would add significant value to the review.

Thanks for your clarification.  Based on the literature survey, it was evident that among the four techniques, fuzzy logic and expert systems are the only ones consistently validated in physical microgrid hardware, whereas ML and metaheuristic remain largely simulation-centric with limited real-world deployment.  This information has been now included in the paper in Page 6.

Comment 5: The authors should condense the elementary definitions of Artificial Intelligence and focus more intensively on why and how these specific architectures are advantageous for energy management and control within microgrids.

Based on this suggestion, the advantages and limitations of these techniques in control of microgrids have been included in the Table 1 itself.

Reviewer 2 Report

Comments and Suggestions for Authors

This paper presents a review on the application of artificial intelligence models and methods to the smart hierarchical control for microgrids.

The authors identify a three-level control strategy (primary, secondary and tertiary control) and then four groups of computational intelligence (CI) models used in power systems control (machine learning, fuzzy logic, expert systems and metaheuristic methods.

The main concern of the reviewer addresses the fact that although there are four CI models identified by authors, they further present the analysis for the three types of control applied only for two types of CI models, namely the fuzzy logic and the artificial neural network models. Please, explain the reason for this choice.

Also, the authors often use acronyms that are not explained. Please, define acronyms at their first use. Example of such unexplained acronyms are: WECS, HPGS, VSI (page 2), UG, MGCC (page 3), SVPWM (in Fig. 2, on page 4), and so on.

Author Response

Reviewer 2:

Comment 1: This paper presents a review on the application of artificial intelligence models and methods to the smart hierarchical control for microgrids.

The authors identify a three-level control strategy (primary, secondary and tertiary control) and then four groups of computational intelligence (CI) models used in power systems control (machine learning, fuzzy logic, expert systems and metaheuristic methods.

The main concern of the reviewer addresses the fact that although there are four CI models identified by authors, they further present the analysis for the three types of control applied only for two types of CI models, namely the fuzzy logic and the artificial neural network models. Please, explain the reason for this choice.

Thanks for the query.  It has been now made clear from the Table 1, that these two models are more suitable for the control of microgrids when comparing with the other models and hence reviewed further for listing their application in microgrid control

Comment 2: Also, the authors often use acronyms that are not explained. Please, define acronyms at their first use. Example of such unexplained acronyms are: WECS, HPGS, VSI (page 2), UG, MGCC (page 3), SVPWM (in Fig. 2, on page 4), and so on.

Thanks for the crucial review.  The acronyms WECS, HPGS, VSI, UG, MGCC are already listed in Page 27 and Page 28.  However, SVPWM alone is not listed, but now included in the abbreviation list in page 28.  Besides these, other acronyms present in the diagrams are also included in the abbreviations now.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents a comprehensive review of Artificial Intelligence techniques applied to the hierarchical control of Microgrids. The authors have structured the review logically and the content is up-to-date.

The authors should add a discussion section or expand the existing sections to compare the performance of these AI techniques.

Figures 4, 5, and 6 are currently presented as bubble charts/word clouds. These figures do not add significant instructional value or structural insight. These figures should be replaced with taxonomy diagrams or flowcharts.

The conclusion notes a gap in "real-time implementation". The body of the text should analyze why this gap exists.

Author Response

Reviewer 3

Comment 1: The manuscript presents a comprehensive review of Artificial Intelligence techniques applied to the hierarchical control of Microgrids. The authors have structured the review logically and the content is up-to-date.

The authors should add a discussion section or expand the existing sections to compare the performance of these AI techniques.

Thanks for the suggestion.  This is now included in the Table 1 in page 6

Comment 2: Figures 4, 5, and 6 are currently presented as bubble charts/word clouds. These figures do not add significant instructional value or structural insight. These figures should be replaced with taxonomy diagrams or flowcharts.

Thanks for the suggestion.  Based on the comment, Figure 4, 5 and 6 are modified as taxonomy diagrams.

Comment 3: The conclusion notes a gap in "real-time implementation". The body of the text should analyze why this gap exists.

Thanks for the suggestion.  The information on the reason for gap in real-time implementation is now included in Page 6 and it is highlighted

Reviewer 4 Report

Comments and Suggestions for Authors

1-DescribeBeyond qualitative descriptions, we encourage the inclusion of comparative tables that report key performance indicators (e.g., settling time, frequency deviation, THD, cost savings, computational burden) to objectively assess AI-based controllers versus conventional methods. 

2- Beyond qualitative descriptions, we encourage the inclusion of comparative tables that report key performance indicators (e.g., settling time, frequency deviation, THD, cost savings, computational burden) to objectively assess AI-based controllers versus conventional methods.

3- Most works cited rely on simulations or HIL tests; please expand the discussion on existing pilot or field deployments, the barriers to full-scale implementation, and lessons learned that could guide practitioners (Hybrid deep learning model for improving the operational efficiency of microgrids, JOURNAL OF SCIENCE AND TECHNOLOGY, VOL. 23, NO. 6A, 2025)

4- Readers would benefit from a standard set of metrics (stability margins, communication overhead, scalability, computational complexity, robustness indices) that future studies should report to enable fair cross-comparison.

Comments on the Quality of English Language

 English could be improved to more clearly express the research

Author Response

Reviewer 4:

Comment 1: Describe beyond qualitative descriptions, we encourage the inclusion of comparative tables that report key performance indicators (e.g., settling time, frequency deviation, THD, cost savings, computational burden) to objectively assess AI-based controllers versus conventional methods.

Thanks for the wonderful suggestion.  However, only a few papers have discussed on the settling time, THD cost minimization etc. and hence it is difficult to include a separate column for this as it would be sparse.  However, we have mentioned in the conclusion in general and it is highlighted

Comment 2: Beyond qualitative descriptions, we encourage the inclusion of comparative tables that report key performance indicators (e.g., settling time, frequency deviation, THD, cost savings, computational burden) to objectively assess AI-based controllers versus conventional methods.

Question is repeated

Comment 3: Most works cited rely on simulations or HIL tests; please expand the discussion on existing pilot or field deployments, the barriers to full-scale implementation, and lessons learned that could guide practitioners (Hybrid deep learning model for improving the operational efficiency of microgrids, JOURNAL OF SCIENCE AND TECHNOLOGY, VOL. 23, NO. 6A, 2025).

Thanks for your recommendation – The information about the barriers to full scale implementation is now included as paragraph in page 6.  The suggested paper deals with AI techniques in microgrid and hence included as reference [16]

Comment 4: Readers would benefit from a standard set of metrics (stability margins, communication overhead, scalability, computational complexity, robustness indices) that future studies should report to enable fair cross-comparison.

Thanks for the suggestion.  This is now included in the Table 1 in page 6

Reviewer 5 Report

Comments and Suggestions for Authors

This paper systematically reviews recent applications of artificial intelligence (AI) technologies in the hierarchical control of microgrids (MGs), covering the fundamental theories, typical methods, and application cases of AI in primary, secondary, and tertiary control levels. There are some comments:

1. As a review article, the authors primarily present existing research results; however, the discussion on the advantages and disadvantages of different AI methods, their applicable conditions, and system limitations is relatively weak.
2. Although the hierarchical control structure is clearly described, the paper does not provide an in-depth analysis of the synergistic effects of AI across the three control layers or the information flow interactions.
3. The paper should cite some of the most recent publications in the microgrid field.

Hierarchical Robustness Strategy Combining Model-Free Prediction and Fixed-Time Control for Islanded AC Microgrids," IEEE Transactions on Smart Grid, vol. 16, no. 6, pp. 4380-4394, Nov. 2025.

Dynamic Event-Triggered-Based Privacy-Preserving Secondary Control for AC Microgrids with Optimal Power Allocation via State Decomposition, IEEE Transactions on Industry Applications, doi: 10.1109/TIA.2025.3618806.

4. The section on future research directions and technical challenges should be strengthened, including issues such as stability under large-scale renewable integration, communication delays, and cybersecurity.

Author Response

Reviewer 5:

 Comment 1: This paper systematically reviews recent applications of artificial intelligence (AI) technologies in the hierarchical control of microgrids (MGs), covering the fundamental theories, typical methods, and application cases of AI in primary, secondary, and tertiary control levels. There are some comments:

As a review article, the authors primarily present existing research results; however, the discussion on the advantages and disadvantages of different AI methods, their applicable conditions, and system limitations is relatively weak.

Thanks for the suggestion.  This issue is  now addressed in the form of a Table 1 in page 6

Comment 2: Although the hierarchical control structure is clearly described, the paper does not provide an in-depth analysis of the synergistic effects of AI across the three control layers or the information flow interactions.

Thanks for the wonderful suggestion.  The cross layer synergic effects of AI across the three layers of microgrid grid control is now included in Page 7 which gives the readers, a clear view on how it makes AI superior when compared to the conventional techniques.

Comment 3: The paper should cite some of the most recent publications in the microgrid field.

Hierarchical Robustness Strategy Combining Model-Free Prediction and Fixed-Time Control for Islanded AC Microgrids," IEEE Transactions on Smart Grid, vol. 16, no. 6, pp. 4380-4394, Nov. 2025.

Dynamic Event-Triggered-Based Privacy-Preserving Secondary Control for AC Microgrids with Optimal Power Allocation via State Decomposition, IEEE Transactions on Industry Applications, doi: 10.1109/TIA.2025.3618806.

Thanks for your recommendation – However, these papers do not utilize any AI techniques for the Microgrid control and hence these two papers are not included in the references

Comment 4: The section on future research directions and technical challenges should be strengthened, including issues such as stability under large-scale renewable integration, communication delays, and cyber-security.

Thank you for the suggestion.  As per the comment, the conclusion is now modified.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript is suitable for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

No more comments.

Reviewer 4 Report

Comments and Suggestions for Authors

The revised manuscript was improved by authors.

Comments on the Quality of English Language

 English could be improved to more clearly express the research

Reviewer 5 Report

Comments and Suggestions for Authors

No