A Density Clustering RAPID Based on an Array-Compensated Damage Index for Quantitative Damage Diagnosis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper discusses a density clustering-RAPID based on array-compensated damage index algorithm. On the whole, the paper is well structured. Through using array compensation and density clustering techniques, the research methodology can improve the accuracy of damage localization and quantify the degree of damage.

However, the clarity of expression could be enhanced in certain sections, making the content more accessible to readers. The major points to address are:

1. This article requires some formatting proofreading.

2. In order to facilitate readers' understanding and universality of the algorithm, the description of the number of pixels (NOP) should be specified and explained in the diagram.

3. The result of NOP for bolt loosening has not been provided and needs to be supplemented.

4. Some charts may need further optimization to improve clarity, and the font and size of the figure captions in Figure 1, Figure 2, Figure 5, Figure 7, Figure 10, and Figure 11 should be consistent with the article.

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Comments to the Author(s)

This manuscript presents A Density Clustering-RAPID Based on Array-Compensated Damage Index for Quantitative Damage Diagnosis. This paper contains a good effort related to SHM and NDT community because it presents an interesting method based on lamb waves for visualizing damages. In general, the manuscript is organized well. However, the authors have to improve the paper by resolving the following issues (All the answers should be included in the manuscript):

1.      Page 3, line 117, please add a reference to the Eq.1 and Eq.2.

2.      Page 4, Table 1, add references to the presented methods.

3.      Fig. 2, how did you measure the reference signals? Did you measure it before creating the plate damage or measure it from a separate pristine plate?

4.      Section 3.1, please add the type of excitation signal.

5.      Page 8, add space between quantity value and its unit.

6.      Page 8, line 249, rewrite the “500mm*500mm*2mm” to be “500 mm × 500 mm × 2 mm”. “2700kg/m^3” to be “2700 kg/m³”. 

7.      Fig. 12, please mention the total number of sensing paths that be used for the construction of the damage images.

8.      Does the number of used sensors affect the resolution of visualizing the damage? How the number of used sensors can be determined?

9.      Fig. 12, please explain the red circle and what it means.

10.  Please mention the limitation of the proposed method (Density Clustering-RAPID).

11.  In Table 2, you have added a note ((unit: mm) to the table caption, so you should remove the units from the values in the table.

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have improved an existing technique (RAPID) by incorporating a damage index and have quantitatively produced damage detection/localization. The methodology has been presented in sufficient detail and validated experimentally by applying on an aluminum plate undergoing damages. There are a few comments that represent the questions arising during the review of the manuscript, presented below:

The authors have used a factor 'beta' in equation 1 which is a shape factor controlling the range of the elliptical weight function. Further explanation of this parameter is needed how the value is varied/chosen for this work.

 

In experimental setup, it is understandable that the PZT13 was attached on the back side of the 2mm thick plate. because the stiffeners were attached on the front face. From the authors perspective, the wave path has its importance and is affected by the presence of the discontinuities. The authors have kept all the sensors on the back of the plate to have a perfectly planar arrangement. Please provide your comments on placement of the PZT13 (actuator) on the back of the plate.

 

The authors provided an explanation of differences in the voltage for two similar wave paths at equal distances from damage. The justification cannot be understood and needs to be explained more clearly. Probably, based on this explanation, the calculation of DI is also carried out and has shown similar values for PZT 1, 7 and 12 compared to reduced value in PZT 4. The explanation need to be provided more clearly.

In figures 8, 9 and 10. There are pairs of vertical lines for each waveform. One of the figures show a label of cross-talk. What do they indicate has not been explained.

 

The method is capable of locating as well as quantifying damage. The types of damage mentioned are crack and screw loosening. Can we differentiate between these two damages using this methodology? In conclusion, some more explanation is needed to justify this claim by the authors.

 

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf