Rectifier Fault Diagnosis Based on Euclidean Norm Fusion Multi-Frequency Bands and Multi-Scale Permutation Entropy
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis paper proposes a novel fault diagnosis method, which has certain research value. However, the details of the methodology, the comprehensiveness of the experimental design, and the in-depth analysis of the results need further enhancement.
1. In the introduction, although the paper mentions the application of single-phase PWM rectifiers in urban rail transit, it lacks specific case analyses of the impact of faults on power systems and train safety. It is recommended that relevant literature or examples be added to enhance the urgency and importance of the research. Additionally, there is a lack of coherence and logic. It is suggested that the content of the introduction be reorganized, the logical connections between paragraphs strengthened, the shortcomings of current research clarified, and the significance and innovation of this study highlighted.
2. For the fault diagnosis algorithm used in this paper, it is recommended to consider or discuss other advanced algorithms proposed in the last two years (https://doi.org/10.1016/j.oceaneng.2024.119665, https://doi.org/10.1007/s43236-024-00916-z) for comparison.
3. In Section 2, when simulating different short-circuit faults, the authors should explain the reasons for setting these faults, the probability of fault occurrence, and the scenarios.
4. In Section 4.1, the authors established the rectifier model, which is the basis for the data source of the proposed method. The authors should provide the validation process and results of the model.
Comments on the Quality of English LanguageNo comments.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors The authors proposed a fault diagnosis method for single-phase PWM rectifiers based on Euclidean norm fusion of multi-frequency bands and multi-scale permutation entropy features. The paper explores the use of wavelet packet decomposition for signal processing, multi-scale permutation entropy for feature extraction, and SVM for fault classification. Results demonstrate high diagnostic accuracy and robustness under varying noise levels, showcasing the method's effectiveness. However, certain aspects of the paper could benefit from further refinement: 1. The paper mentions the use of db wavelet functions for signal decomposition but does not provide sufficient reasoning for why specific functions (e.g., db4, db6) were optimal for different fault modes. Adding a detailed explanation or comparison of the wavelet function selection process could enhance the transparency and reproducibility of the study. 2. To better highlight the advantages of the proposed approach, it would be helpful to include a comparison with other commonly used fault diagnosis methods, such as deep learning-based techniques (e.g., CNNs or DBNs) or simpler entropy-based methods (e.g., single-scale entropy without fusion). Presenting comparison results in additional tables or figures would provide readers with a clearer understanding of the method's relative performance in terms of accuracy, robustness, and noise resistance. 3. The robustness of the method under strong noise is well demonstrated, but the paper does not sufficiently discuss potential limitations, such as its applicability to other types of faults or systems beyond single-phase PWM rectifiers. Including a brief discussion on the method's limitations and suggestions for future work could strengthen the paper's impact and applicability.Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsRectifier fault diagnosis based on Euclidean norm fusion multi-frequency bands and multi-scale permutation entropy
Review
A fault diagnosis technique based on Euclidean norm fusion multi-frequency bands and multi-scale permutation entropy for single-phase Pulse Width Modulation (PWM) rectifier is proposed in this paper.
By using optimal wavelet function, the optimal multi-frequency bands information of the fault signal is selected after the wavelet packet decomposition. Then, are calculated the multi-scale permutation entropy of each frequency band, and multiple fault feature vectors. Euclidean norm is used to fuse the multi-scale permutation entropy into an entropy value, so that each frequency band uses an entropy value to characterize the fault information features.
The topic is interesting.
The manuscript contains system analysis and fault analysis, simulation and results obtained from matlab.
Also, the manuscript contains a good review of the state of the art and related literature. The selected references are relevant.
Recommendations for improving the content and accuracy are below:
1. To validate the simulation results some experimental tests and measurements should be added.
2. It is unclear how the presented simulation can be used in a real electric traction drive of an electric locomotive supplied by a PWM rectifier-inverter.
3. In the Conclusions are summarized the previous ideas of the manuscript. Nevertheless, real applications of the proposed method must be added. Such applications could be retrieved from literature and added in a section Discussion.
4. In the List of References, in many lines, there is are symbols [J] or [C] which must be removed.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsI recommend that this paper be accepted. In addition, I suggest including relevant algorithms in Section 4 for comparison.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsNo further comments
Author Response
Thanks again for your review of my paper.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe revised manuscript can advance for publication.
Author Response
Thanks again for your review of my paper.