An Integrated Cost-Aware Dual Monitoring Framework for SMPS Switching Device Diagnosis
Round 1
Reviewer 1 Report
The section 2 titled “Motivation” is just a general explanation of SMPS, but not a motivation of the paper, I think. Such an explanation should be brief and can be described in the introduction.
Why did the authors use “functional blocks” instead of circuit diagram in Figure 1? I think that circuit diagram of switching circuits is more common. In the first place, can I ask what is “functional blocks”?
The subsection 4.2 is a general explanation of MOSFET. I think it is easily found in any textbook about electronics. This journal is named electronics. The authors can explain it for reader’s convenience but they should make it clear where they use it for their research. The authors should not refer to [37] and [38] for such general explanation of MOSFET.
In Table 5, voltage and current were sampled at 1000Hz. But I did not find at which element or node the authors sampled voltage and current in the circuit. There are so many candidates of voltage and current. The authors must explain the circuit diagram they used and show which voltage and current they should measure for diagnosis.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Dear Authors,
Based on the 1st round review of the manuscript entitled An SMPS Switching Devices Fault Diagnosis Approach using Support Vector Machine and k-Nearest Neighbors Classifier, the reviewer has the following comments:
- The contributions are not clear. Please highlighted in the revised manuscript.
- Which types of faults (sensor, system, or actuator) is analyzed in this work? please mention about it in the revised manuscript.
- Why the rest of the work is explained in 4 parts. You can put it in the paragraph.
- Based on Figure 2, what is your novelty in this research? please explain it deeply in the revised manuscript.
- As you know the reliability plays an important role, so, how you can validate the reliability of the proposed method (the proposed method is a combination of feature extraction and machine learning approach)? Please explain about it deeply in the revised edition.
- How about the robustness, which part is used to improve the robustness? and how you can validate it (one of the main challenge in data driven/AI techniques is robustness). Please explain about it deeply in the revised manuscript .
- Based on Table 1, why these features are selected? How you can feature selection?
- Based on the results, the reviewer believes that this is fault detection technique not fault diagnosis? In which part the author is identified the types of the fault?
- What is limitation of the proposed method? please explain about the limitation in the conclusion before the future work.
Regards,
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors revised the paper according to the reviewer's comments.
Reviewer 2 Report
Dear Authors;
Regarding the 2nd round reviw of the manuscript it can be accepted for further processing .
Regards,