Bridge Health Monitoring via Displacement Reconstruction-Based NB-IoT Technology

Round 1

Reviewer 1 Report

The article is well structured. The abstract describes the content well. The intorduction is clear and describes the state of the art. Experimentation is appropriate and the results are well presented. Therefore the article is accepted in this form.

Author Response

Reviewer #1:

The article is well structured. The abstract describes the content well. The introduction is clear and describes the state of the art. Experimentation is appropriate and the results are well presented. Therefore, the article is accepted in this form.

Response:

We deeply appreciate reviewer’s recognition for our study.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper entitled Bridge Health Monitoring via Displacement Reconstruction-based NB-IoT Technology describes the use of acceleration measurements on a bridge to perform structural health monitoring assessment. The topic is very interesting. There has been quite some investigation on the bridge SHM in the past, however, the field has still opportunities to implement novel approaches especially in damage localization and service life estimations.

While reading I noticed that there are some instances where singular is used instead of plural and some capital letters are missing at the beginning of the sentence. I am kindly suggesting the authors provide another read of the paper to correct these minor spell and grammar mistakes.

My comments are the following:

• I am suggesting that instead of saying artificial model use numerical model.
• Please describe the direction of the sensitive axis of the mounted instruments.
• Can you please explain how did you consider the type of damage and location? Did you consider real failure modes for the bridge under investigation?
• How do you consider environmental factors?
• I could not find section A-A marked anywhere, figure 10 mentions section A-A, where is the location of the bridge?
• Laboratory tests: what do you consider light and heavy mass of the vehicle?
• Which similarity rules did you consider when building the model in the laboratory?
• You are describing the effect of the added mass on the model. What is the added mass with respect to the mass of the bridge? Instruments are measuring the response of the system and a part of the system is also the vehicle, which contributes to the system's mass matrix. How do you mechanically explain the effect of the velocity of the vehicle and mass on the fundamental frequencies? In your opinion is this true only for the model or prototype as well?
• Figures 12, 13, and its explanation is very confusing to the reader. If I understand correctly the cart is once on position A and once on Position B. And the cart is not moving? Where is the response measured? It is very brave to claim that frequency is not a reliable bridge health indicator based on a scaled model where similarity rules are not explained and where the reader can see that only 2D scenario was considered, and measurements seem to be only in the direction of the gravity, while bridges exhibit strongly 3D behavior. Please consider this.
• Figures 14, 15, and others hard to notice anything on a grayscale plot.
• Are directions H1 and H2 marked anywhere?
• Explain boundary conditions on the numerical model?
• How did you consider the fact that the actual bridge moves around 3 axes? Damage on the bridge affects movements in all directions.
• Please explain more about the Wanshou bridge. How old is it, when was it build? Which data is considered the intact state and why? How long did the experiment take place? How were environmental factors considered? Is there any seasonality on the data or how do you explain various indexes for different months? Is this elastic deformation, plastic? Was the bridge examined before the test?
• Figures 28-31, please explain more with respect to my previous comment.
• 6/15, 6/30 this is the date?
• In line 386 do you mean the first fundamental frequency of the bridge is 4Hz?

I am guessing your aim was to introduce the new method and not a complete SHM system for the bridge. I am guessing you aim to further develop the method and application. However, I must emphasize that in my opinion aging of infrastructure must be to at least some extent supported with numerical modeling. The sensors indicate there is damage, but the actual consequences on the structure cannot be determined without any mechanical background.

I believe the paper is good. However, it needs some improvement and some clarification. I am also enclosing a pdf with my comments, which are basically the same as the one I wrote here. But maybe it will help the authors understand what I meant.

Comments for author File: Comments.pdf

Author Response

Please refer to the attached file for details.

Author Response File: Author Response.docx

Reviewer 3 Report

• How does this method compare (in theory and in performance) with the many efforts using influence line-based monitoring strategies. The authors should, at very least, include in the intro mention and references to this field of study. See, for example:

Zhang, S. and Liu, Y Damage Detection in Beam Bridges Using Quasi-static Displacement Influence Lines. Applied Sciences 2019, 9, 1805.

Zeinali, Y.; Story, B.A. Impairment localization and quantification using noisy static deformation influence lines and Iterative Multi-parameter Tikhonov Regularization. Mech. Syst. Signal Process. 2018, 109, 399–419.

Zeinali, Y.; Story, B.A. Framework for Flexural Rigidity Estimation in Euler-Bernoulli Beams Using Deformation Influence Lines. Infrastructures 2017, 2, 23.

Zaurin, R.; Catbas, F.N. Integration of computer imaging and sensor data for structural health monitoring of bridges. Smart Mater. Struct. 2009, 19, 1–15.

Zaurin, R.; Catbas, F.N. Structural health monitoring using video stream, influence lines, and statistical analysis. Struct. Health Monit. 2011, 10, 309–332.

 

• Why is an earthquake engineering example used in Fig. 2 and for finding Nw? The field of bridge monitoring, and especially vehicle-bridge interaction, has many examples pertinent to the topic of study in this paper.
• I suggested consistent formatting in Tables 2 and 3 (lit. on left side and proposed on right side, etc.)
• Significant detail is missing in the conversion from acceleration to displacement. This is a difficult task and the details of exactly how  you achieve such excellent results are unclear.
• What is the increase in performance by using Newmark-Beta over Central Difference?
• What is the motivation behind using the windowing method and not reconstructing the entire displacement time history using Newmark’s method?
• What is the motivation behind using the middle point and not another value, such as the mean of the displacement over the window?
• Figures 16 and 17 seem to show the same data, but arranged differently. Figure 17 seems to be more descriptive – why keep Figure 16?
• Table 8 – for a 40 m single span, the given value for I (0.00327 mm^4) seems very low. Verify that the units are correct.
• It seems like the captions for Figures 20-25 need to be updated, as the Level 1 damage indicator is not included in the figures but is referenced in the captions.
• Figure 28b should have the same x-axis as figure 28a (same for Figs. 29-31)
• Is the presented methodology sensitive to changes within the bridge system apart from damage? In other words, using the proposed damage indicators, is damage uniquely identifiable, or would the damage indicators also be high if the temperature changed, etc?
• A couple additional sources for feature-based damage detection approaches:

2019. Mei, M. Gül, and M. Boay, “Indirect health monitoring of bridges using mel-frequency cepstral coefficients and principal component analysis,” Mechanical Systems and Signal Processing, vol. 119, pp. 523–546, 2019.
2020. Liao, A. S. Kiremidjian, R. Rajagopal, and C.-H. Loh, “Structural damage detection and localization with unknown post-damage feature distribution using sequential change-point detection method,” Journal of Aerospace Engineering, vol. 32, no. 2, p. 04018149, 2019.

There are minor editorial/grammar mistakes in the manuscript throughout. A thorough editorial review is required.

Author Response

Please refer to the attached file for details.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear all,

The authors have replied and considered all my comments, suggestions, and dilemmas. I suggest the paper to be published in Applied Sciences.

I wish the authors all the best with their future work.

Kind regards

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors have clarified many aspects of the paper. I have a few remaining comments:

-The literature review is improved, but can you provide any comparisons of the basic performance of your method vs. those methods? This would improve understanding of which methods may be more appropriate in specific situations.

 

-The technical content is clarified sufficiently.

 

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx