Defense Strategy against False Data Injection Attacks in Ship DC Microgrids

Round 1

Reviewer 1 Report

The article presented shows a novel method for mitigating cyber attacks and, more precisely, defending from FDIA on a ship's DC microgrid. The methodology is presented appropriately, and 6 case studies are used to present the results, which seem to be satisfactory.

Lines 22-23. Please do not cite 5 sources simultaneously, as presented in the paper. Take the time and explain the ideas supported by the citations.

Throughout the paper, there are some abbreviations issues. Please address the name and abbreviation at the beginning of the text (0. Introduction, but not the abstract. Start like it is a new text). And afterwards, be consistent with it (see line 185). It is important to use abbreviations when you are explaining ideas.

Lines 107-108. The syntaxis of the phrases seems strange. Could you please rewrite the paragraph?

Figure 3, it seems that the description does not apply to the image.

Do you plan to apply this technique on board and check whether the system's behaviour is the same as in Matlab?

Author Response

Point 1: Lines 22-23. Please do not cite 5 sources simultaneously, as presented in the paper. Take the time and explain the ideas supported by the citations.

Response: Thank you for your advice. It has been modified as “Compared to Alternating Current (AC) grids, Direct Current (DC) grid has been an increasingly popular choice for ship electric system applications [2], due to its flexible operation, high power density, space saving, and free of synchronization. After consideration, one article is enough to confirm a common sense point of view, so the remaining four articles are deleted.”

Point 2. Throughout the paper, there are some abbreviations issues. Please address the name and abbreviation at the beginning of the text (0. Introduction, but not the abstract. Start like it is a new text). And afterwards, be consistent with it (see line 185). It is important to use abbreviations when you are explaining ideas.

Response: It has been modified. For example:” Compared to Alternating Current (AC) grids, Direct Current (DC) grid has been an increasingly popular choice for ship electric system applications.

The data of the communication network used by the supervisory control and data acquisition (SCADA) system is vulnerable to attacks because there is no firewall (because of the delay problem),”

Line 185 was changed. This is the changed content.

“In order to implement the ANN online, the ANN is trained offline.”

Point 3. Lines 107-108. The syntaxis of the phrases seems strange. Could you please rewrite the paragraph?

Response: It has been modified as “Formula (1) is expressed in the form of matrix. Define  as the weight matrix of the hidden layer of ANN. Define  is the bias vector.”

Point 4. Figure 3, it seems that the description does not apply to the image.

Response: It has been modified as “Figure 3. Control Hierarchy of DC microgrid.” The picture is consistent with the content.

Point 5. Do you plan to apply this technique on board and check whether the system's behaviour is the same as in Matlab?

Response: In consideration of time, this technology has not been implemented on the board, but this work will be carried out in the follow-up study.

Author Response File: Author Response.pdf

Reviewer 2 Report

1. In abstract ..."Results shows that the method can identify and mitigate the FDIA effectively." can be briefly detailed to be more clear than this general equation

2. "1." Introduction not"0"

3. the introduction must be improved, for instance, the authors can add the types of faults for the microgrid in ships, their solutions, method used, etc.

4. DC microgrid in the first sentence, DC must be defined; the authors can check all symbols and their appearances in the manuscript. Also, the authors can describe what a microgrid means in the ship for non-specialist readers.

5. Symbols such as (Vri) in sentences must be written as text to be aligned with the sentence.

6. I think all the sections that describe the model can be combined together into one section. let's say a numerical model with different subsections describing the model.

7. The authors need to describe the primary and secondary control units more.

8. The cases require more description and details.

9. Conclusion must be improved.

10. The authors can add a nomenclature section.

 

Author Response

Point 1. In abstract ..."Results shows that the method can identify and mitigate the FDIA effectively." can be briefly detailed to be more clear than this general equation.

Response: Thank you for your advice. It is changed to the following: ”Results shows that the method can identify and mitigate the FDIA effectively, the error of identifying FDIA by ANN is less than 0.5V. Therefore, the deviation between the actual bus voltage and the reference voltage is less than 0.5V.”

Point 2. "1." Introduction not"0"

Response: It has been modified as”1. Introduction”. And adjusted the overall title number of the article.

Point 3. the introduction must be improved, for instance, the authors can add the types of faults for the microgrid in ships, their solutions, method used, etc.

Response: This is what was added. “The data of the communication network used by the supervisory control and data acquisition (SCADA) system is vulnerable to attacks because there is no firewall (because of the delay problem), the communication protocol lacks strong encryption, and it is not updated to deal with the latest network security threats. Modern ship control systems use commercial off the shelf computing platforms [4]. In [4], the tertiary control of the ship's power grid is attacked by FDIA, the objective of an attacker is to maximize load curtailment during cyber attack by redispatching the generators. To detect signs of malicious data, a multiagent system (MAS) that checks commands from the central energy management system is employed.”

The added content has been marked in the updated manuscript

Point 4. DC microgrid in the first sentence, DC must be defined; the authors can check all symbols and their appearances in the manuscript. Also, the authors can describe what a microgrid means in the ship for non-specialist readers.

Response: This is what was added. ”Compared to Alternating Current (AC) grids, Direct Current (DC) grid has been an increasingly popular choice for ship electric system applications [2], due to its flexible operation, high power density, space saving, and free of synchronization.

 

The ship's microgrid is a small power generation and distribution system composed of distributed power supply, energy storage device, energy conversion device and some important loads of ship power grid. Among them, distributed power generation refers to small power generation system mainly using clean energy or renewable energy [1].”

 

Point 5. Symbols such as (Vri) in sentences must be written as text to be aligned with the sentence.

Response: Align symbol with text in manuscript. “Where  is the reference voltage of the j-th unit;  is the reference DC bus voltage of the j-th unit;  is the droop coefficient of the j-th unit;  is the output current of the j-th unit.”

 

Point 6. I think all the sections that describe the model can be combined together into one section. let's say a numerical model with different subsections describing the model.

Response: It has been modified as

“2. Anti-FDIA Model of Ship DC Microgrid

2.1. Basic Features of ANN

2.2. FDIA to DC Secondary Control

2.3. Detection and Mitigation Strategy”

 

Point 7. The authors need to describe the primary and secondary control units more.

Response: Detailed description in the introduction. “Hierarchical control consists of three layers. The goal of the primary control is to maintain the voltage stability. The droop control is used to adjust the output voltage and the power distribution between each converter. The secondary control is used to compensate the voltage drop generated by the primary control. The goal of the secondary control is to ensure that the power is transmitted according to the scheduling plan and the average voltage is adjusted. Based on the secondary control, the tertiary control introduces management strategy and optimization algorithm, sets the priority of load, and formulates optimal operation strategies for different working conditions to regulate power quality [2,3].”

Figure 3 is the schematic diagram of primary control and secondary control.

 

Point 8. The cases require more description and details.

Response: Some details have been added, For example:

“The simulation step is 5e-5s.

The topology of bidirectional DC/DC converter used is two-way Buck/Boost circuit.

The droop coefficient () in droop control is 0.5. The proportional parameter and integral parameter of the secondary controller are set to 1.

The output capacitance of the converter is 1000 μF.

The switching frequency of the converter is 5kHz.”

 

Point 9. Conclusion must be improved.

Response:

The conclusion has been updated to

 

All updated content:

“Under various disturbances, the estimated deviation of ANN to bus voltage is less than 0.5V, while the influence of 0.5V DC bus voltage deviation on ship power grid can be ignored.  use abbreviations when you are explaining ideas.”

Also, We use abbreviations when we are explaining ideas.

For example: Use “ANN” instead of “artificial neural network”.

 

Point 10. The authors can add a nomenclature section.

Response: Thank you very much for your constructive comments. According to your comments, the following contents will be added.

NOMENCLATURE

Indices and Sets

FDIA	False Data Injection Attacks
DC	Direct Current
ANN	Artificial Neural Network
AC	Alternating Current
SCADA	Supervisory control and data acquisition
MITM	Man-In-The-Middle attack
DoS	Denial of Service
MAS	Multiagent system
CPS	Cyber-physical system
STL	Signal Temporal Logic
RNN	Recurrent Neural Network
DER	Distributed energy resources
	Weight matrix of the hidden layer of ANN
	Bias vector of ANN
	The reference DC bus voltage
Vdc	DC bus voltage
	The ANN's estimate of the DC bus voltage

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Accepted