Next Article in Journal
Feature Mining: A Novel Training Strategy for Convolutional Neural Network
Next Article in Special Issue
Benchmarking of Load Forecasting Methods Using Residential Smart Meter Data
Previous Article in Journal
Creating a Remote Choir Performance Recording Based on an Ambisonic Approach
Previous Article in Special Issue
Benchmarking Various Pseudo-Measurement Data Generation Techniques in a Low-Voltage State Estimation Pilot Environment
 
 
Article
Peer-Review Record

Identification of Typical and Anomalous Patterns in Electricity Consumption

Appl. Sci. 2022, 12(7), 3317; https://doi.org/10.3390/app12073317
by José Nuno Fidalgo 1,2,* and Pedro Macedo 1
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
Appl. Sci. 2022, 12(7), 3317; https://doi.org/10.3390/app12073317
Submission received: 28 January 2022 / Revised: 15 March 2022 / Accepted: 21 March 2022 / Published: 24 March 2022
(This article belongs to the Special Issue Data Science Applications in Medium/Low Voltage Smart Grids)

Round 1

Reviewer 1 Report

It appears that Figure 1 is intact; however, under Section 2.1.1, there is an error message, “Error! Reference source not found.” Is something missing? The logical progression of the paper is very good. There are various formatting inconsistencies (e.g., font), such as on page 8 of 15 under Section 2.4, last paragraph. Figure 14 is partially cut-off on the right-hand side; however, I was able to glean the essence. The caveats in the conclusion nicely frame the practicalities of meter tampering, etc, that would obviate the presented approach.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The reviewed manuscript is at an average level when it comes to the originality of the solutions presented and the detail of the research carried out as well as the final conclusions formulated.

Despite the above reservations, the manuscript brings valuable comments to the discussion and indicates directions for further research related to the identification of typical and anomalous patterns of electricity consumption and will certainly meet with the interest of readers.

This manuscript includes the analysis of exploratory data and the identification of typical patterns in electricity distribution networks, followed by the detection of unusual electricity consumption, clustering and ranking of consumers by degree of anomaly. However, the analysis tools known from the literature are used here, containing significant simplifications.

The manuscript presented in this form is practically ineligible for publication in Applied Sciences and requires some changes, in particular as regards linguistic and stylistic correctness. Also requires corrections and additions to the presented text. For example, some of them are given below:

  • Nomenclature should be supplemented with all markings appearing in the manuscript, e.g. following markings are missing: DSO, P_week, P_month etc.
  • Subsection 2.2 is missing
  • Tables 1 and 2 are unreadable and should be reformatted
  • For Figure 1 and Figure 8, an error occurs i.e. reference source has not found
  • Section 4 Conclusions should be supplemented by quantitative conclusions resulting from the studies carried out.
  • In the manuscript is missing: Author Contributions, Funding, Data Availability Statement, Conflicts of Interest etc.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This paper may be of interest to many industry partners.

 

Technical and high level feedbacks:

  1. The literature review paper in the introduction section is short. Do we need review state of art research in this topic area? To reveal a gap so there is need of this paper to be published as an archival journal paper to contribute to human knowledge.

 

  1. What is the difference between LVN and LVS? Why are they different?
  2. The methodology section needs more work, such as re-organisation and editing. For example,

-  if the section starts with 3 overall steps, please consider move description about exploratory data analysis to the next sub-section 2.1.

- a suggestion is to review the methodology and results section to see how the story flows

  1. Please consider to review consistency and coherency for the whole manuscript. For example,

- prototypes, classes, clusters, typical patterns are used to describe more or less the same meaning. However, it can be confusing for future readers to comprehend.

- The methodology section mentioned annual consumption and the description below Figure 13 still mentions annual consumption. However, the caption and legend for Figure are “scale classes”. To make the paper more friendly to juniors/wide audience, do we need consider improve consistency in naming and ideas’ representation?

 

 

Other feedbacks:

  1. In Table 2, on Page 7 of 15, please consider to make the first letter of each month upper case. The 5th month in English is “May”.
  2. Figure 1 and Figure 8 in-text reference have an error message: Error! Reference source not found.
  3. The last paragraph in Section 2.4 has different style of font compared to other sections.
  4. From Page 12, paragraphs use different indent style to previous pages. Please consider to make it consistent.

 

Thanks a lot for great work on such an interesting topic.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

  1. Are the authors pioneers in identifying patterns of electricity consumption and using them to detect, in particular, non-technical energy losses? If not, it is advisable to present the contribution of other researchers into this field.
  2. Presented prototypes for week and month patterns of electricity consumption (Figure 11 and 12 respectively) are not clear because the groups of consumers are not defined.
  3. The criteria or examples of criteria of atypical consumption are not given in section 3.2 Anomalies detection. Without that, the above section is very vague. Also the information on building of potential anomaly score are ommited.
  4. Figures 15 and 16 are left without any comments and explanation. There are not clear relations of this figures to the previously given information. The y-axis is described in all cases in the same way. It would be interesting to know what were real causes of presented anomalous.
  5. In the section named Nomenclature, the authors explained a number of abbreviations. It is worth to add and explain one more: DSO. By the way, the abbreviation LVN is not used in the manuscript.
  6. It is advisable to add the unit of Gap dimension on Figure 4 and 5.
  7. Percentage of cases per class are omitted on figure 11. It also would be good to mention the absolute number of cases used to define each of the classes presented on Figure 11 and 12.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I would like to thank the authors of the manuscript for taking into account my comments in the review (round 1).

The manuscript in its current form is eligible for publication in Applied Sciences.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

Thank you the authors for their answers and small improvements in the manuscript.

Please consider the following changes:

228
base stings -> base settings

302 and 305
The authors are suggested to use the full name of the abnormality filters in lines 302 and 305 instead of their abbreviations. This will be more in line with the descriptions of the corresponding figures.

308
Probably the Enter key was pressed accidentally and the line breakdown should be removed.  

Author Response

Please see the attachment.

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.


Back to TopTop