An Ensemble Transfer Learning Spiking Immune System for Adaptive Smart Grid Protection

Round 1

Reviewer 1 Report

This work presents an artificial immune system (AIS) for smart grid protection. Overall the manuscript is well organized and supported by experimental evidence, but I have a few suggestions to improve:

1) The authors claim that the contribution of the work is an innovative AIS, but many other solutions can be easily found in the literature. I will only give a few examples:

[R1] Chowdhury MM, Tang J, Nygard KE. An artificial immune system heuristic in a smart grid. In The 28th International Conference on Computers and Their Applications 2013.

[R2] Song K, Kim P, Tyagi V, Rajasekaran S. Artificial immune system (AIS) based intrusion detection system (IDS) for smart grid advanced metering infrastructure (AMI) networks.

In view of the above, the authors must clearly state the effective contribution of their study at the end of section 2. What are the improvements compared with the most recent state of the art? What are the research gaps?

3) Why did the authors choose CSA as an optimization tool? The inherent advantages and limitations of the algorithm must be clearly stated. Please, refer to [R3] and elaborate.

[R3] Haktanirlar Ulutas B, Kulturel-Konak S. A review of clonal selection algorithm and its applications. Artificial Intelligence Review. 2011 Aug;36(2):117-38.

4) The description given in section 3 is too short. Either incorporate it into section 4 or improve the discussion. A flowchart must be added to demonstrate the operation of the AIS, but it has nothing to do with the one in Fig. 2.

5) The discussion of results is not deep enough. Please, improve it by adding plots that allow verifying the performance of the assessed algorithms. Another important aspect that should be incorporated into the analysis is the computational burden.

Non-technical aspects include the need to replace the figures with high-quality vector graphics and adopt the same style for the tables.

Author Response

Dear respected reviewer

We express our sincere appreciation for your insightful comments and constructive suggestions, which have significantly helped us improve the manuscript. For clarity, we have uploaded a copy of the original manuscript with all highlighted changes. Appended to this letter is our point-by-point response to the comments raised by the reviewers. The words are reproduced, and our answers are given directly afterward in a different color (red).

We would also like to thank you for allowing us to resubmit a revised copy of the manuscript.

Reviewer 1

This work presents an artificial immune system (AIS) for smart grid protection. Overall, the manuscript is well organized and supported by experimental evidence, but I have a few suggestions to improve:

1) The authors claim that the contribution of the work is an innovative AIS, but many other solutions can be easily found in the literature. I will only give a few examples:

[R1] Chowdhury MM, Tang J, Nygard KE. An artificial immune system heuristic in a smart grid. In The 28th International Conference on Computers and Their Applications 2013.

[R2] Song K, Kim P, Tyagi V, Rajasekaran S. Artificial immune system (AIS) based intrusion detection system (IDS) for smart grid advanced metering infrastructure (AMI) networks.

In view of the above, the authors must clearly state the effective contribution of their study at the end of section.

ANS: Thank you for this constructive comment that allows us to clarify things further.  The revised manuscript includes the suggested scientific literature papers based on your suggestions. It must be noted that there isn't another comparable model in the literature to use as a benchmark. Specifically, the proposed approach employs optimally spiking neural networks, Clonal Selection Algorithm (CSA), ensemble theory, and advanced Transfer Learning algorithms in an efficient hybrid method, presented for the first time in the literature review. From this point of view, the contribution of the work is an innovative AIS. In addition, to avoid bias or incorrect impressions, we perform the proposed model without comparing it with other alternative models except those shown in the paper's tables.

The above appropriate explanations according to your suggestions were added at the end of introduction section.

2) What are the improvements compared with the most recent state of the art? What are the research gaps?

ANS: Thank you for this comment. The revised introduction section improved the motivation of the proposed approach and highlighted the research gaps filled by the method. Also, the improvements in state of the art are described in the conclusion section.

3) Why did the authors choose CSA as an optimization tool? The inherent advantages and limitations of the algorithm must be clearly stated. Please, refer to [R3] and elaborate.

[R3] Haktanirlar Ulutas B, Kulturel-Konak S. A review of clonal selection algorithm and its applications. Artificial Intelligence Review. 2011 Aug;36(2):117-38.

ANS: Thank you for this helpful comment.  The revised manuscript includes the suggested explanations and references. Specifically, “The suggested CSA algorithm is a biologically inspired technique for multimodal search and learning. Their coding schemes and evaluation functions are identical to those of other approaches, such as genetic algorithms. Nonetheless, the evolutionary search methods vary in inspiration and evolutionary step sequencing, which creates significant transparency of the process and various configuration options. Furthermore, the CSA method is highly parallel and has excellent tractability regarding computing cost and resources. In any event, it should be noted that the CSA can achieve a diversified range of local optima solutions without being trapped in pointless retrospectives. The other strategies tend to polarize the whole population toward the best candidate answer, while in the CSA algorithm, convergence is implemented in a much more specific solution.”

4) The description given in section 3 is too short. Either incorporate it into section 4 or improve the discussion. A flowchart must be added to demonstrate the operation of the AIS, but it has nothing to do with the one in Fig. 2.

ANS: Thank you for this comment. Section 3 was incorporated into section 4 according to your valuable suggestion.

5) The discussion of results is not deep enough. Please, improve it by adding plots that allow verifying the performance of the assessed algorithms. Another important aspect that should be incorporated into the analysis is the computational burden.

ANS: Thank you for this comment. According to your suggestions, the appropriate explanations and discussion were added to the revised 4. Dataset and Results section. We believe the revised version gives the apparent meaning of performance results values, and the new plots allow verifying the performance of the assessed algorithms.

6) Non-technical aspects include the need to replace the figures with high-quality vector graphics and adopt the same style for the tables.

ANS: Thank you for this constructive comment. The tables were rearranged according to the journal template and were added figures in high-quality resolution (300dpi).

Author Response File: Author Response.docx

Reviewer 2 Report

This paper proposed an ensemble transfer learning spiking immune system for adaptive smart grid protection, and the proposed system fully automates the strategic security planning of energy networks with computational intelligence methods.The research topics closely follow the reality, and the work has a certain degree of innovation. However, there are several things that the author should pay attention to, and I hope the author can improve the article for the following points.

1. The Introduction part only briefly introduces the research background and purpose, and does not clearly show the ideological contribution of the author of this study and the innovation of the proposed system.

2. Since the author chooses to use The Proposed Artificial Immune System as a separate chapter to introduce the concept source of the system in detail, it is suggested that the advantages and disadvantages of this system and previous studies should also be introduced together to pave the way for the subsequent conclusion.

3. The result comparison of relevant model cases in the Dataset and Results section can be displayed by using the statistical analysis chart, which is more intuitive than the simple grid data.

Author Response

Dear respected reviewer

We express our sincere appreciation for your insightful comments and constructive suggestions, which have significantly helped us improve the manuscript. For clarity, we have uploaded a copy of the original manuscript with all highlighted changes. Appended to this letter is our point-by-point response to the comments raised by the reviewers. The words are reproduced, and our answers are given directly afterward in a different color (red).

We would also like to thank you for allowing us to resubmit a revised copy of the manuscript.

This paper proposed an ensemble transfer learning spiking immune system for adaptive smart grid protection, and the proposed system fully automates the strategic security planning of energy networks with computational intelligence methods.The research topics closely follow the reality, and the work has a certain degree of innovation. However, there are several things that the author should pay attention to, and I hope the author can improve the article for the following points.

1. The Introduction part only briefly introduces the research background and purpose, and does not clearly show the ideological contribution of the author of this study and the innovation of the proposed system.

ANS: Thank you for this comment. The revised introduction section improved the motivation of the proposed approach and highlighted the research gaps filled by the method. Also, the improvements in state of the art are described in the conclusion section.

2. Since the author chooses to use The Proposed Artificial Immune System as a separate chapter to introduce the concept source of the system in detail, it is suggested that the advantages and disadvantages of this system and previous studies should also be introduced together to pave the way for the subsequent conclusion.

ANS: Thank you for this comment. The manuscript was written and enhanced to a level appropriate for the prestigious journal. Section 3 was incorporated into section 4 to be self-consistent.

3. The result comparison of relevant model cases in the Dataset and Results section can be displayed by using the statistical analysis chart, which is more intuitive than the simple grid data.

ANS: Thank you for this comment. Following your suggestions, we have improved the section by adding relevant comparison plots. Also, the appropriate explanations and discussions were added to the revised 4—Dataset and Results section. We believe the revised version gives the apparent meaning of performance results values, and the new plots verify the assessed algorithms' performance.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript presents an interesting topic of an advanced standard for securing energy infrastructure to automate operational cyber security. The authors claim that the model uses advanced computational intelligence methods in a hybrid system first introduced in the literature. Some concerns have been found and they need to be handled before the manuscript can be considered for publishing, as follows,

- The most significant findings of this work should be highlighted in the abstract.

- Has the schematic representation of CSA in Figure 2 been taken from a specific reference, if so, please mention this reference?

- The numeric results of different scenarios have not been discussed in a way that can give the obvious meaning of these values or compare them to others.

- The conclusion section can be improved by including some numeric results and their importance to this work.

Author Response

Dear respected reviewer

We express our sincere appreciation for your insightful comments and constructive suggestions, which have significantly helped us improve the manuscript. For clarity, we have uploaded a copy of the original manuscript with all highlighted changes. Appended to this letter is our point-by-point response to the comments raised by the reviewers. The words are reproduced, and our answers are given directly afterward in a different color (red).

We would also like to thank you for allowing us to resubmit a revised copy of the manuscript.

Reviewer 2

The manuscript presents an interesting topic of an advanced standard for securing energy infrastructure to automate operational cyber security. The authors claim that the model uses advanced computational intelligence methods in a hybrid system first introduced in the literature. Some concerns have been found and they need to be handled before the manuscript can be considered for publishing, as follows,

- The most significant findings of this work should be highlighted in the abstract.

ANS: Thank you for this constructive comment. The most significant findings of this work are highlighted in the abstract as suggested. Specifically, the revised abstract includes “The most significant findings of this work are located in that the transfer learning architecture's shared learning rate significantly adds to the speed of generalization and convergence approach. Also, the ensemble combination improves the accuracy of the model because the overall behavior of the numerous models is less noisy than a comparable individual single model. And last, the Izhikevich Spiking Neural Network used, due to its dynamic configuration, can reproduce different spikes and triggering behaviors of neurons, which models precisely the problem of digital security of energy infrastructures, as proved experimentally”.

- Has the schematic representation of CSA in Figure 2 been taken from a specific reference, if so, please mention this reference?

ANS: Thank you for this helpful comment. The schematic representation of CSA in Figure 2 has not been taken from any specific reference but drawn by the authors. Also, according to your suggestions, we have improved the references of the method.

- The numeric results of different scenarios have not been discussed in a way that can give the obvious meaning of these values or compare them to others.

ANS: Thank you for your helpful remarks. We have added an appropriate discussion according to your suggestions. We believe that the revised 5. Dataset and Results section gives the apparent meaning of performance results values.

- The conclusion section can be improved by including some numeric results and their importance to this work.

ANS: Thank you for this comment. We have added an appropriate discussion according to your suggestions that improves the conclusion section by including some numeric results and their importance to this work.

Author Response File: Author Response.docx

Reviewer 4 Report

Authors have presented a nice work.

However authors are encouraged to include more latest scientific literature published recently in 2022 for the study in question and compare the results to validate the significance of the proposed advanced computational intelligence methods.

Author Response

Dear respected reviewer

We express our sincere appreciation for your insightful comments and constructive suggestions, which have significantly helped us improve the manuscript. For clarity, we have uploaded a copy of the original manuscript with all highlighted changes. Appended to this letter is our point-by-point response to the comments raised by the reviewers. The words are reproduced, and our answers are given directly afterward in a different color (red).

We would also like to thank you for allowing us to resubmit a revised copy of the manuscript.

Reviewer 3

The authors have presented a nice work. However, authors are encouraged to include more latest scientific literature published recently in 2022 for the study in question and compare the results to validate the significance of the proposed advanced computational intelligence methods.

ANS: Thank you for this constructive comment. Based on your suggestions, the revised manuscript includes the latest related scientific literature published in 2020 - 2022 for the study in question It must be noted that there isn't another comparable model in the literature to use as a benchmark. Consequently, to avoid bias or incorrect impressions, we present the performance of the proposed model without making any comparisons with any other alternative models except that shown in the tables of the paper.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

The manuscript has been revised properly based on the provided comments.