Next Article in Journal
Beyond the Ecological Boundary: A Quasi-Natural Experiment on the Impact of National Marine Parks on Eco-Efficiency in Coastal Cities
Previous Article in Journal
Green Concrete with Glass Powder—A Literature Review
Previous Article in Special Issue
Transient Stability Analysis for Grid-Forming VSCs Based on Nonlinear Decoupling Method
 
 
Article
Peer-Review Record

A Non-Intrusive Identification Approach for Residential Photovoltaic Systems Using Transient Features and TCN with Attention Mechanisms

Sustainability 2023, 15(20), 14865; https://doi.org/10.3390/su152014865
by Yini Ni, Yanghong Xia *, Zichen Li and Qifan Feng
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
Sustainability 2023, 15(20), 14865; https://doi.org/10.3390/su152014865
Submission received: 26 August 2023 / Revised: 23 September 2023 / Accepted: 12 October 2023 / Published: 13 October 2023
(This article belongs to the Special Issue Power Generation Systems for Green Sustainable Energy)

Round 1

Reviewer 1 Report

Dear authors,

Congratulations for your research regarding the evaluation of the photovoltaic systems by using advanced mathematical techniques. The transient features and temporal convolutional network with Attention mechanisms used, combined with the experimental analysis results into a very interesting paper.

The introduction is well structured and the reference list is recent.

I propose to publish the paper in the present form.

Author Response

The authors thank the reviewer for this comment. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

 

The authors, in this paper, have presented a non-intrusive method for accurately identifying residential photovoltaic (PV) systems' grid-connection states and switching events in real-time by leveraging transient features and a temporal convolutional network (TCN) model with attention mechanisms, finally demonstrating its precision and practical efficiency on an experimental platform.

In my opinion, the best way to determine which model works best for your application is to consider factors like the amount of data available, the complexity of the problem, and the computational resources at your disposal.

I have the following comments:

1. Did the authors compare their model (TCN) with other models such as Long Short-Term Memory (LSTM) networks, Gated Recurrent Unit (GRU) networks, and Convolutional Neural Networks (CNNs) with 1D convolutions, or other?

2. Please check Section 4 in Page 10. “Based on the feature selection method using semi-Fisher score and MIC, as described in Chapter 1, the discrimination and redundancy of different transient features in various devices are computed and ranked. According to the ranking results, the features are added to …. “. What do you mean by « as described in Chapter 1 » ?

3. In my opinion, all sections must be improved. More new references must be added. Comparisons with other works must be well established. Discussion section must be improved.

4. Please check Figures 4 and 7.

 

Minor corrections of English are needed.

Author Response

The authors thank the reviewers for their comments, Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript presents and analyzes an improved NILM (Non-Intrusive Load Monitoring) method for PV grid connection recognition. The study has an actual and well-presented literature revision and shows an appreciable amount of work.

Next, certain aspects that the authors may correct or clarify for paper improvement:

1- The manuscript has many acronyms, so I recommend the authors include a glossary.

2- The authors could improve the validation of the method through experimental analysis with photovoltaic systems that use different inverters.

3- It is not clear what the authors intend to present in Figure 10.

Author Response

The authors thank the reviewers for their comments, Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The idea and results in the paper look promising and interesting. Kindly execute the followings to transform the paper towards publication:

  1) What type of PV array architecture has been utilized in this paper. For example, string-level inverter PV, centralized inverter PV with series-parallel architecture. A block or single line flow diagram would be enough.

  2) What would be the impact of MPP algorithm and partial-shading/mismatch effects of PV. Kindly just add few lines about this in the text.

 3) Can you please compare your method with some other. Even qualitative comparison in tabular format would be sufficient. 

 4) Kindly cite the following articles to improve the Reference Sec..

    a)  A. F. Murtaza, H. A. Sher, “A reconfiguration circuit to boost the output power of a partially shaded PV string”, Energies, 2023. 

   b) A. F. Murtaza, H. A. Sher, F. Usman, A. Nasir, F. Spertino, “Efficient MPP tracking of photovoltaic (PV) array through modified boost converter with simple SMC voltage regulator”, IEEE Transactions on Sustainable Energy, 2023

 

 

Negligible errors 

Author Response

The authors thank the reviewers for their comments, Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

According to the authors' responses, I can make a positive decision to publish their paper.

The authors need to review the manuscript's English in order to refine it.

Reviewer 4 Report

All comments are well-answered. Accepted. 

Back to TopTop