Energy Storage Scheduling in Distribution Systems Considering Wind and Photovoltaic Generation Uncertainties
Round 1
Reviewer 1 Report
The discussed topic is interesting and the article is quite clear in illustrating both methodology and results. However, in the reviewer’s opinion, some integrations are required before accepting the manuscript for the final publication. In details:
A) In the introduction, the authors specify that generation uncertainties come from PV units, wind generators and unregulated small-scale hydropower plants. It is not clear why the last type of DG units is no longer considered in the paper; thus, the paper should be integrated with this technology specific characterization.
B) The paper seems to consider ESSs owned by the DSO (in general, ESS could be owned by private end-users or by DSOs). It should be clarified.
C) In general, optimization methods identify a global solution or fall to partial solution (local minimum solution different from the global solution). Please discuss the reported methods (Table 1) in terms of ability in identifying the global solution of the optimization problem. Additionally, the authors are called to discuss this point considering that they are proposing to use “fmincon”. Have other solvers been investigated by the authors?
D) Please discuss the level of complexity of the proposed method. How the computational effort grows with the size of the problem (i.e., number of network nodes, number of end-users, number of storage devices, etc.)?
E) Rows 173 reports “the cost of grid losses is implicitly accounted for through these terms”. This point requires a deeper explanation. How the grid losses are accounted in these terms?
F) Function f(v_t) reported in formula (17) has to be represented graphically, taking into account typical wind-to-power correlations for real wind turbines compatible in size with the connection at the medium voltage level.
G) Since a standard network (e.g., IEEE networks, etc.) is not used, please specify the network main characteristics (e.g., distribution feeders’ length, line types, etc.): the LV domain is single-phase or multi-phase? With neutral conductor? These aspects have a significant impact on distribution management, unbalancing, losses, etc. (e.g., please see “Effects of Distributed Generation on Power Losses in Unbalanced Low Voltage Networks” @ 2018 IEEE Power and Energy Society General Meeting, or “Losses management strategies in active distribution networks: A review” @ Electric Power Systems Research, 163, pp. 116-132)
H) In the case study, it is not clear if the charging/discharging efficiencies include the inverter losses or not
I) Is it possible (and reasonable) to correlate the output of the methodology (in terms of reduction of the total cost for operating the distribution network) with the storage system capital cost (around 500-700 € per kWh?) This analysis could be interesting to understand how much this technology is near to be a concrete solution for improving the distribution network management.
J) In the reviewer’s opinion, the conclusions could be further summarized.
K) Some English errors are present (e.g., row 170, “corresponds to optimizing”) and some orthographic errors require attention (e.g., row 438. “scenario: In a part”, row 483 “follows: We consider”, row 630 “energy: For the case”, etc.)
L) Font height in Figure 4 should be increased
Author Response
We would like to thank the reviewer for very thorough and thoughtful comments on our manuscript. Please see the attached document for detailed responses to all comments in the review report.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The proposed manuscript titled “Energy storage scheduling in distribution systems considering wind and photovoltaic generation uncertainties” proposes a framework for model accounting for uncertainties due to distributed resources such as wind or solar photovoltaic generation.
In general terms, the article is correct and well explained. Only, as minor criticism, some parts of this are a bit wordy. As an example of this criticism, section 5 (conclusions) could be simplified, expressing the ideas in a more summarized and concise way, not repeating the ideas already expressed in the documents so extensively. Additionally, few typos has been detected (e.g.; Page 7, Line 192, “where where”).
Thus, the proposed document is technically interesting and shows an appropriate approach to solve the ESS scheduling. Moreover, the results of this work are validated through a study based on data from a real network, highlighting the advantages of the proposed approach.
Author Response
We would like to thank the reviewer for the reviewing and providing feedback on the manuscript, and we refer to the attached document for detailed responses to the comments in the review report.
Author Response File: Author Response.pdf
Reviewer 3 Report
This manuscript presented methods and models that solving the ESS scheduling optimization problems for wind and PV in a system. The literature review and introduction provided thoroughly relevant references and backgrounds. The methods and models include MPOPF, value function, stochasticity of wind and PV system were adequately described. The case study of a Norway DG system demonstrated the proposed methods and models, and the results showed the effectiveness of solving the ESS scheduling with DG uncertainties. Only a few items are suggested for authors:
1. Reduce or condense the manuscript, so readers could get the whole picture easily.
2. Grammar checks through the manuscript, for example, line 141.
Author Response
We would like to thank the reviewer for the reviewing and providing feedback on the manuscript, and we refer to the attached document for detailed responses to the comments in the review report.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Questions have been discussed and the paper quality is remarkably enhanced in comparison with the first submission. I am suggesting to accept the paper in the present form.
