A New Drive-by Method for Bridge Damage Inspection Based on Characteristic Wavelet Coefficient
Round 1
Reviewer 1 Report
This paper presents a drive-by method for bridge damage inspection based on characteristic wavelet coefficients. Some general conclusions are given and potentially helpful for related researchers. In the reviewer's opinion, the paper should be minor modified before the paper is accepted, and the authors were suggested to clarify the following aspects:
- Several abbreviations have been used in this paper. Authors should give a list of abbreviations and clearly define or explain as each abbreviation first appears in the content. Once defined, the text content can be represented by abbreviations, and it is not necessary to repeat the complete text in the article content; the author should recheck the entire article content for abbreviations.
- Signal analysis processing methods are usually well-validated in numerical models. However, applying it to the actual structure will be a real challenge. It is suggested that in the conclusion section, the authors should add more information about how the methodology of the study can be effectively applied to the practical structure, the difficulties that may be encountered, and the areas where this method can be improved in the future
Author Response
The detailed response is documented in the attached file named "Response Letter-final".
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This study proposed a drive-by based method to detect damage in bridge beams using characteristic wavelet coefficient. The CWC is obtained by analyzing the response via CWT and CEEMDAN. The paper is interesting regarding the noise filtering capability of the proposed method. However, several issues need to be addressed regarding the proposed method. This reviewer recommends a minor revision.
1. The main issue(s)/problem(s) of the study should be highlighted in the abstract.
2. What are the previous WT methods applied to solve these issue(s)/problem(s)?
3. What are the drawbacks of these previous methods?
4. “CWT” in line 14 should be written in full (continuous wavelet transform) for the first time.
5. Line 16 has a repetition of “of the”.
6. The following research related to the topic of the study, I suggest benefiting from and citing it.
doi.org/10.1016/j.measurement.2019.07.027
doi.org/10.1080/24705314.2020.1862969
doi.org/10.1016/j.jsv.2018.07.011
7. Line 43 and 44 states “…which is costly and could affect the traffic operation.”. This statement is not correct, Output-Only methods of SHM does not affect operations/traffic of bridge. In fact, the traffic serves as the source of excitation.
8. GNSS and UAV should be written in full.
9. “key challenges still remain with the modal parameter-based drive-by method for bridge damage inspection, including how to extract the modal parameter efficiently and accurately, and how to eliminate the effect of roughness and noise.” What methods have been previously applied and what are the drawbacks of these methods?
10. “CWT” in line 98 should be written in full.
11. Line 127, how is CEEMDAN overcoming these problems?
12. Your parametric study considers scale factor, vehicle speed, noise and boundary effect, can the author justify why the mass of vehicle was not considered?
13. “It is noteworthy that the WC contains a variety of high-order frequency contents of bridge, which has adverse effects on the damage inspection accuracy.” How?
14. What is the difference between damage Case 1, Case 6 and Case 9?
15. What method was applied in the preliminary analysis to select Mexican hat wavelet?
16. Recheck the scale factor of Figure 5b.
17. In Figure 6a, why does only 40% damage severity has a peak on the right-hand-side?
18. In Section 3.3, why did the authors apply high damage severity of 40% to show the improvement?
19. “Fig. 10” in Line 341 should be “Fig. 9”.
20. What is “wo” in Line 362?
21. How was boundary effect applied in the parametric study?
22. How is border distortion taken care in the proposed method?
23. The authors should provide an example indicating damage at the 1st or 30th element.
24. The authors should justify why they applied only 2 speed values in the parametric analysis.
25. Line 416: “For example, when the vehicle speed is 4 m/s, it is impossible to extract the damage”. Is this due to the low speed or noise?
26. In the second paragraph of page 15, how come speed of 1 m/s can detect damage in the 2nd element but not with speed of 4 m/s? Fig. 13 shows that speeds of 8 m/s and 12 m/s was able to detect damage in the 2nd element.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
