A Novel NODE Approach Combined with LSTM for Short-Term Electricity Load Forecasting

Round 1

Reviewer 1 Report

The manuscript titled 'A novel NODE approach for electricity load forecasting' has been completely reviewed. I have some suggestions that may help to improve paper quality further:

1- The title of the article is quite general. It is better to use the predictive method used in the article in the title. 2-  Table 1 (Parameter setting of the proposed models) should be revised and determined in a better way. As a reviewer, I cannot evaluate the proposed method with this table. The presented format of the table should be changed and more explanations should be provided for the references. 3- To what extent do you believe that the proposed method can work under different study cases and conditions?
4- The background and literature review of this paper is alittle limited. More quality papers regarding forecasting algorithms should be reviewed and used in this section, for example: i) 'A new prediction model of electricity load based on hybrid forecast engine'; ii)  'A New Spinning Reserve Requirement Prediction with Hybrid Model'. The reviewer highly recommends reading and using the above mentioned papers in literature review of your article to provide better understanding this issue. 5- How have you considered and minimized the three sources of error including noise, bias and variance in the simulation?
6- After Figure 14, it is better to provide an explanation and discussion about the obtained results.

Author Response

Please see the attached

Author Response File: Author Response.docx

Reviewer 2 Report

1.     The manuscript presents a novel NODE approach for electricity load forecasting, which is interesting. The subject addressed is within the scope of the journal.

2.     However, the manuscript, in its present form, contains several weaknesses. Appropriate revisions to the following points should be undertaken in order to justify recommendation for publication.

3.     Full names should be shown for all abbreviations in their first occurrence in texts. For example, OCBL in p.2, OBBL in p.2, OBBLL in p.2, etc.

4.     For readers to quickly catch your contribution, it would be better to highlight major difficulties and challenges, and your original achievements to overcome them, in a clearer way in abstract and introduction.

5.     It is shown in the reference list that the authors have several publications in this field. This raises some concerns regarding the potential overlap with their previous works. The authors should explicitly state the novel contribution of this work, the similarities, and the differences of this work with their previous publications.

6.     p.1 - a neural ordinary differential equation method is adopted for electricity load forecasting. What are other feasible alternatives? What are the advantages of adopting this method over others in this case? How will this affect the results? The authors should provide more details on this.

7.     p.1 - three groups of models are adopted in the experiments. What are the other feasible alternatives? What are the advantages of adopting these groups of models over others in this case? How will this affect the results? More details should be furnished.

8.     p.1 - long short-term memory and bidirectional LSTM are adopted to combine with NODE. What are the other feasible alternatives? What are the advantages of adopting these soft computing techniques over others in this case? How will this affect the results? More details should be furnished.

9.     p.6 - the structure as shown in Figure 4 is adopted for the NODE block. What are other feasible alternatives? What are the advantages of adopting this structure over others in this case? How will this affect the results? The authors should provide more details on this.

10.  p.6 - three evaluation metrics are adopted to assess the performance of the models. What are the other feasible alternatives? What are the advantages of adopting these evaluation criteria over others in this case? How will this affect the results? More details should be furnished.

11.  p.7 - Queensland Australia is adopted as the case study. What are other feasible alternatives? What are the advantages of adopting this case study over others in this case? How will this affect the results? The authors should provide more details on this.

12.  p.7 - the window size as 20 and processing STLF with step 1, 3, 5 and 7 are adopted in the experiments. What are the other feasible alternatives? What are the advantages of adopting these parameters over others in this case? How will this affect the results? More details should be furnished.

13.  p.10 - specific parameters as shown in Table 1 are adopted for the proposed models. What are the other feasible alternatives? What are the advantages of adopting these parameters over others in this case? How will this affect the results? More details should be furnished.

14.  p.11 - “…Howerve, models with NODE seem to perform similar to model without NODE when step is 1. The reasons is that.…” More justification should be furnished on this issue.

15.  p.13 - “…There are three reasons why NODE can improve the prediction accuracy of LSTM and BiLSTM. Firstly.…” More justification should be furnished on this issue.

16.  Some key parameters are not mentioned. The rationale on the choice of the particular set of parameters should be explained with more details. Have the authors experimented with other sets of values? What are the sensitivities of these parameters on the results?

17.  Some assumptions are stated in various sections. Justifications should be provided on these assumptions. Evaluation on how they will affect the results should be made.

18.  The discussion section in the present form is relatively weak and should be strengthened with more details and justifications.

19.  There are some occasional grammatical problems within the text. It may need the attention of someone fluent in English language to enhance the readability.

20.  Moreover, the manuscript could be substantially improved by relying and citing more on recent literature about real-life case studies of contemporary soft computing techniques on electricity load forecasting such as the following. Discussions about result comparison and/or incorporation of those concepts in your works are encouraged:

l   Chen, Z., et al., “A novel short-term load forecasting framework based on time-series clustering and early classification algorithm,” Energy and Buildings 251: 111375 2021.

l   Han, J.Y., et al., “A Task-based Day-ahead Load Forecasting Model for Stochastic Economic Dispatch,” IEEE TRANSACTIONS ON POWER SYSTEMS 36 (6): 5294-5304 2021.

l   Fallah, S.N., et al., “Computational Intelligence on Load Forecasting: A Methodological Overview,” Energies 12 (3): article no. 393 2019.

21.  Some inconsistencies and minor errors that needed attention are:

l   Replace “…to propagates through the Eq. (11)…” with “…to propagate through Eq. (11)…” in line 153 of p.5

l   Replace “…in time series And another is dynamicly adaptive to…” with “…in time series and another is dynamically adaptive to…” in line 159 of p.6

l   Replace “…easier for the follow LSTM or BiLSTM layer to learn the serie features…” with “…easier for the following LSTM or BiLSTM layer to learn the series features…” in line 173 of p.6

l   Replace “…Howerve, models with NODE seem to perform…” with “…However, models with NODE seem to perform…” in line 267 of p.11

l   Replace “…The reasons is that the combination…” with “…The reason is that the combination…” in line 268 of p.11

l   and more…

22.  In the conclusion section, the limitations of this study and suggested improvements of this work should be highlighted.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thanks for your response and clarifications.

Reviewer 2 Report

The revised paper has addressed all my previous comments, and I suggest to ACCEPT the paper as it is now.