Optimal Placement and Sizing of an Energy Storage System Using a Power Sensitivity Analysis in a Practical Stand-Alone Microgrid

Round 1

Reviewer 1 Report

This paper proposes ESS sizing algorithm based on the power flow model sensitivity. The proposed algorithm  was implemented on an islanded power system.  The paper contribution needs to be clarified and a comparison to more sizing techniques is required.

• P2, L47, the mentioned problems may be power quality or power reliability problems, not all power stability problem; please clarify
• In-depth literature review is required for the works [9] to [20]
• In eq(3), what will be the case if a bus has both PV and ESS?
• Eq(13) needs more explanation
• How the proposed algorithm takes into account the ESS cost?
• Comparison to other ESS sizing techniques is required, either heuristic sizing algorithm such as PSO, GA or gradient-based methods.

Author Response

Please see the attached response to the reviewer file.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper is well written and easy to follow. There are some parts where more careful presentation of the work done should be done. For instance, the relations (1) and (2) are not clear, should be more careful explained e.g., signs minus and plus involved, see also the phrase from lines 114 – 116. Terms involved in equation (3) are not completely explained, they should be. Then it is not clear if the paper refers to the installation of a single ESS or multiple ESSs at the same tee, with respect to the involvement of the optimization algorithm proposed. Another formulation is not clear and should be deeper explained: “priority according to analysis of all candidate placements” (rows 157-159). It is not clear if in the optimization process the defined objective function (relation 13) should be maximized or minimized – please detail. Following the flow-chart from Figure 1 it seems that the algorithm refers to installation of only one new ESS at a time so it cannot be used for multiple ESS installation at the same time – meaning a single design variable. The algorithms in fact do a set of repeated analyses and keep the best solution that maximizes the objective function. From this perspective, the algorithm proposed is not an optimization algorithm in the sense known in the scientific literature. With this respect, the question is, having in fact a single design variable (location to install ESS), would have not been more effective to implement, for instance, a gradient optimization method to find out rapidly the optimal solution?

Author Response

Please see the attached response to the reviewer file.

Author Response File: Author Response.pdf

Reviewer 3 Report

The first sentence of the paper is not well thought-out. Renewable energy sources such as biomass or wind generate pollution no less than the fossil fuels. E.g. biomass: exhaust gases, wind: noise pollution & vibration, solar panels: lots of dead solar panels after 20 yrs etc. Authors should stick to facts and not mindlessly repeat ecological propaganda phrases.

Sentences about AI in lines 83-88 are contradicting. The first says AI can give optimal solutions, and the second that it cannot. The comparison of methods to propose EES size and location should be improved and more in depth. Why AI cannot give optimal solution? Please give some references on that.

 

Near line 88, the authors should mention that AI does typically require large amount of data to learn. This is a disadvantage, because learning on big data can be time and power-consuming, and requires large-scale sensor installations in the first place.

 

Line 122: "The subscripts, ESS and MG mean that the values of the ESSs connected buses and the values of the other buses excluding the ESSs connected buses."

should be corrected, I guess, to:

"The subscripts, ESS and MG denote the values of the ESSs connected buses and the values of the other buses excluding the ESSs connected buses, respectively."

 

In line 123 it is noted that the value for the actual slack bus is not considered. Authors should explain is this the result of Taylor expansion or the value is ignored by the authors by an arbitrary decision? If the latter is true, how does this affect the accuracy of the optimization method?

Explanation of K_EES and K_MG matrix in lines 127-129 should be rewritten (language is not best, the sentence seems to be chopped and unfinished). 

 

The phrases "like as (5)", "like as (6)" etc. should be replaced with "as presented in equation (5)" or "as in (6)"

 

The indices in matrices K_ESS and K_MG are not clearly explained. Why the first element is (2,2) and not (1,1) or (0,0) ?

 

Paragraph starting at line 183 begins with a phrase: "It was determined by the proposed algorithm for optimal installation placement and sizing for a newly installed ESS."

But that sentence is poorly placed, it should explain what was determined by the algorithm. The style of writing in this paper should be improved by some good English speaker.

Author Response

Please see the attached response to the reviewer file.

Author Response File: Author Response.pdf

Reviewer 4 Report

Dear Authors,

I have no significant comments on the article. My critical remarks concern:
1. The abstract should be completed with the quantitative and qualitative results of the optimization performer,
2. The introduction is written correctly. The introduction defines the research problem and sufficiently reviews the literature on the topic. The last paragraph of the introduction (lines 99 - 103), which have the character of a summary of the article, should be removed,
3. Conclusions should be supplemented with quantitative and qualitative results of the optimisation performed.

 

Author Response

Please see the attached response to the reviewer file.

Author Response File: Author Response.pdf

Reviewer 5 Report

The paper deals with the optimal positioning and sizing of ESS in a microgrid. However, the paper is not organized very well. The methodology and information are not explained fluently. The description of each step for obtaining the results is not well explained and the results are not compared with other methodologies. All in all, for acceptance of the paper, the authors must rewrite almost all the parts of the paper and they must add some more technical information related to the data of the grid and optimization procedure.

Please find here some technical reviews:

• In the introduction, you need to connect the state of the art to your paper goals. Please follow the literature review with a clear and concise state of the art analysis. This should clearly show the knowledge gaps identified and link them to your paper goals. Please reason both the novelty and the relevance of your paper goals.
• The manuscript needs a thorough revision of its language and style. Overall, this paper is not clear. I suggest a thorough overhaul of the text for a clearer understanding of the reader.
• The methodology is not clear, the problem formulation is summarised (the description is not in detail), the procedure and a clear description of methodology should be explained. E.g., Figure 1 shows the flowchart of methodology, which is not very comprehensive, I would suggest providing a new figure with more details.
• Case of study and dataset are not explained in detail, e.g., the characteristics of the system are missed. Please introduce the case of study in a section.
• The novelty of the work, considering the sensitivity analysis for the placement of ESS and sizing of it, is not clear. The results should be compared with similar works, and also the methodology should be applied to the different case studies. Or at least another methodology should be applied to the same case and the outcomes should be compared.
• The system constraints are not explained. The ESS constraints and charging-discharging equations are not explained. A detailed explanation of system constraints should be described.
• The ESS type and its characteristics are not explained. Also, it would be mentioned that the exact sizing and placement of ESSs is time depending, since, for the planning of ESSs long period should be taken into account considering the seasonal variation of loads. Also, for the optimal usage of ESSs, a multi-period power flow should be considered since a single period can not represent an optimal sizing and placement of ESS in a power system.
• In general, the figures are not well organized, the quality, caption description, legends should be improved.
• In the conclusions, in addition to summarising the actions taken and results, please strengthen the explanation of their significance. It is recommended to use quantitative reasoning comparing with appropriate benchmarks, especially those stemming from previous work.

Author Response

Please see the attached response to the reviewer file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The paper contribution needs to be clarified and a comparison to more sizing techniques is required.

• state clearly the main contribution of this work compared to other works?
• Comparison to other ESS sizing techniques is required, either heuristic sizing algorithm such as PSO, GA or gradient-based methods?
• provide a detailed response to reviewer document, manuscript comments are not acceptable 

 

Author Response

Please refer to the attached authors' response file. Thank you.

Author Response File: Author Response.pdf

Reviewer 5 Report

Thank you for your detailed overview of the comments.

Author Response

Please refer to the authors' response file. Thank you.

Author Response File: Author Response.pdf