Fault Diagnosis of HV Cable Metal Sheath Grounding System Based on LSTM
Round 1
Reviewer 1 Report
The paper with title “Fault Diagnosis of HV Cable Metal Sheath Grounding System Based on LSTM” presents a study on fault diagnosis of metal sheath of high voltage (HV) cables.
Is the first time that fault diagnosis method is introduced?
How this work is different from previous reported studies using LSTM?
Figure 1 shows HV cable grounding system model. Which tool was used to simulate the model?
Is the model practically implemented and tested? Are the results verifiable?
It is suggested to incorporate the details in the manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
This work describes about a fault diagnosis of HV cable metal sheath grounding system based on LSTM. This article is interesting and useful topic in the real industry. In order to have your paper ready for publications, however, you should answer and revise the paper following comments:
1. Organize the advantages and disadvantages of the proposed method into a table and add it in the chapter 1 or 2
2. There are many kinds of neural network algorithms (MNN, CNN, LSTM, Boosting etc.) for fault classification, but why did you use LSTM for clustering?
3. Why did you use the 14 feature vector [a1 a2 .... b6 b7] for inputs of the NN?
What is the relationship between the number of input variables (features) and fault diagnosis performance?
4. What are the Confusion matrix of classification results (Figure 14) for other methods (see Table 4)? Especially SVM and KNN!
5. Include the analysis of the experimental results and the further study in the “Chapter 5 Conclusion”.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Authors have provided answers to the questions and added the information to improve the manuscript.
Reviewer 2 Report
The comments of the reviewer were well reflected and revised.