Composite Panel Damage Classification Based on Guided Waves and Machine Learning: An Experimental Approach
Round 1
Reviewer 1 Report
Dear authors,
I have been through the article and my comments are as follows:
1. The title needs to be revised, the order of the terms is not inline (Optional).
2. The statement, "The proposed approach can be used for accurate and efficient damage diagnosis and classification in composite structures, with potential applications in aerospace, automotive, and civil engineering." should better be in the abstract.
3. Introduction is dragged, better target the topic you have worked on. Remove redundancies.
4. The paper suffers from a lack of clarity in presenting the methodology used for classifying damage in the composite panel. The sequence of the experimental setup and data processing is mixed up, making it challenging for readers to follow the procedures accurately. The authors should provide a clearer and more organized description of their experimental approach to enhance the reproducibility and reliability of the study.
5. One major flaw in the paper is the absence of information about the numerical model used in conjunction with the experimental setup. A numerical model is essential in understanding the theoretical basis of the study and validating the experimental results.
6. The paper fails to sufficiently explain the data sets used for the specific parts of the research and how they were validated. Without a clear understanding of the data sources and validation methods, readers cannot assess the reliability and generalizability of the results. The authors should elaborate on the data collection process and the criteria used to validate the accuracy of their machine learning-based classification system.
7. The conclusion section of the paper is fragmented and lacks a cohesive overview of the findings and their implications. The authors should present a comprehensive summary of the experimental results, discuss the significance of their contributions, and highlight the limitations and future directions for further research.
Moderate changes required in introduction.
Author Response
Please, find in attached author's replies
Author Response File: 
Author Response.docx
Reviewer 2 Report
The subject of this paper is interesting and relevant to the journal topics. The authors consider the relevant subject of knowledge, which is non-destructive diagnostic. The study is based on the Ultrasonic Guided Waves, which cause the signal analysis (signal classification). This problem is decided based on the machine learning methods as
- The analysis of the problem can be extended by considering the application of the fuzzy classifier in non-destructive diagnostics. In particular, I’d like to propose consider the publications:
- Rabcan, J., Levashenko, V., Zaitseva, E., Kvassay, M., Subbotin, S., Application of Fuzzy Decision Tree for Signal Classification, IEEE Trans Industrial Informatics, 15(10), pp. 5425–5434, 8666793
- Agrawal, P., Jayaswal, P., A Review of Fault Detection, Diagnosis, and Prognosis of Rolling Element Bearing Using Advanced Approaches and Vibration Signature Analysis, Lecture Notes in Mechanical Engineering, pp. 207–219, 2020
- The experiments need explaining. Why does the accuracy of the training data consider (Table 6)? In the ML methods, the accuracy of the tested data is considered in the evaluation.
- Why do you consider accuracy only for the method evaluation There are other metrics for the evaluation of the ML-based methods (sensitivity, availability F1 scope etc).
Author Response
Please, find in attached author's replies
Author Response File: Author Response.docx
Reviewer 3 Report
The paper is very interesting and very well written. The visual quality of the figures could be improved though.
Author Response
Please, find in attached author's replies
Author Response File: Author Response.docx
Reviewer 4 Report
This research showcases empirical findings that illustrate the promise of this method for precise and effective identification and categorization of damage in diverse sectors.
However, it is essential to address the following inquiries:
- What are Ultrasonic Guided Waves (UGW) and how do they contribute to Structural Health Monitoring (SHM) systems?
- In what way does the post-processing of signals collected by piezoelectric sensors aid in extracting characteristics associated with the location and extent of damage?
- What do R-Classification and S-Classification entail, and how do machine learning models employ them to differentiate fault location and magnitude?
More a basic architectural diagram is required to add here in the paper for better understanding.
The review is comparatively less, do add more scientific content and information related to your work nearly.
The experimental accuracy needs to be clearly justified and also one basic methodology to be added which is missing.
English is fine, but grammatically few places needs to be crosschecked again and there is some typo errors, do check once again by the authors.
Author Response
Please, find in attached author's replies
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Recommended for publication.
Minor.
Author Response
Authors thank reviewer for his comments and suggestions.
Reviewer 2 Report
I thank the authors for the consideration of the comment. According to the "Authors' response" document these comments were accepted. But I didn't find these modifications in the text of the manuscript. I would like to see these changes in the text of manuscript too (in particular, i.1 and i.3).
Author Response
Authors thank reviewer for his comments and suggestions. The specified sections have been highlighted in green, including the two citations in the reference section and the portions related to the evaluation metrics.
Reviewer 4 Report
the author has incorporated all the query responses and I accept the submission in its present form.
English moderate changes are required, still, there are some typographic errors and grammatical errors.
Author Response
Authors thank reviewer for his comments and suggestions.
