Experimental Validation of a High Precision GNSS System for Monitoring of Civil Infrastructures

Round 1

Reviewer 1 Report

1. The authors describe the problem clearly and carry out verification, but they lack a coherent description of the discovered laws.

2. The authors do not describe the standard quantities used in the experimental procedure.

3. The verification under complex conditions such as unstable signal or interference needs to be supplemented. If it cannot be measured, it can be simulated.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper presents a comparison of GNSS-based and accelerometric structural monitoring of deflections and accelerations. The topic of the presented paper and the methodology is quite interesting. The structuring and some evaluation approaches should be reformulated. The aspect of infrastructure monitoring (ref. title) comes quite short with just a qualitative comparison of two diagrams.

Comments in detail:

- The introduction starts with the acknowledgments and the funding declaration. This is already given at the end of the paper and is more suitable there.

- The state of the art is presented quite shortly. There is a lot of research about GSNN-based structural monitoring or other satellite-based monitoring (e.g. radar interferometry). The novelty of this paper should be pointed out more detailed.

- ln. 38 pp.: Why do you refer to seismic monitoring? In the remainder of the paper, you focus on any deflection. Seismic loading is quite specific.
The mentioned benefit to compare two measurements before and after an earthquake should rather be mentioned in the conclusions.

- ln. 58-84: This is a long summary of the paper. It repeats the abstract and is partially repeated in the beginning of section 2.

-> The introduction should be revised:
Acknowledgment and funding at the end.
State of the art in satellite-based monitoring of structures.
Summary is not necessary.
Conclusions belong at the end.

- ln. 120: "appropriate configuration" is quite general and not comprehensible. Please give more details.

- ln. 123-124: What means "appropriately connected"? Please rephrase.

- ln. 149: How could you ensure that there is a static condition? One would assume a minimum of vibration at any time.
Does the vibration/movement of the building have any impact on the results since it is more dominant in the global deflection monitoring of the GNSS system?

- fig. 2/ln. 153: Please give more details about the static system. What is the static system? What is the cross-section of the vertical beams and how are they connected in horizontal direction?

- Fig. 6 and 7 show the same results, only the values on the vertical axis change. One figure would be enough.
In Fig. 6 are some (small) vibrations in x visible during the excitations (ux). They are not visible in Fig. 7. I would expect them here if only the mean value is shifted.

- ln. 233-234: How accurate can the resting position be determined? Is it measured at a specific time or determined from an average deflection over a certain period?

- ln. 236: What are the parameters of the Butterworth filter?

- ln. 315: The difference between FEM and measurement is quite high. The model should be improved. Otherwise it has no benefit to the study.

- ln. 320: The caption "results" is not suitable here. You presented more results in the previous section. This is rather an application.

- ln. 325: Why were no FEM analyses carried out? It is difficult to assess the method and the measurement accuracy without any data to compare.

- ln. 332: Please add the country where the bridge is located. Orte isn't well-known worldwide.
What means "(VT)"?

- ln. 333 pp.: Please give more details about the structure. A schematic view of the cross-section and the longitudinal view would be helpful.
Is it a sequence of 4 single-span beams? Are all the 4 box girders connected in the transverse direction?

- ln. 339: The depicted mode shapes (Fig. 12) are the first two for a 2D system. Is this also true for a 3D system or are there relevant modes perpendicular to the plane?

- Fig. 13: It would be helpful to show the position of the sensors in the photo.

- ln. 348-350: You mounted the sensor on the guardrail. Here I would expect a dominant vibration of the guardrail itself (especially under seismic loading or wind). How do you ensure that it has no impact? In Sect. 2 you argue vice versa.
The mentioned "verification" is not explained and not comprehensible.
Was the accelerometer also mounted on top of the guardrail?

- ln. 371 pp.: The "Validation" is quite short. The results are presented but not validated.
The comparison of the deflections from GNSS vs. accelerometer is reasonable only for position no. 2. Because of the different positions on the structure different deflections must be expected. Of course, they are smaller in position no. 1 and as heavy trucks normally drive on the right side (in most countries), the deflections are also smaller on the left.

-> Section 3.2 you be expanded to a real validation and evaluation of the results. This is rather a short qualitative comparison of the deflections. Why could the vibrations / natural frequencies not be evaluated?

- Sect. 4: Discussion: The section is quite short and a detailed or general result to the accuracy of the measurement is missing. -> Section 4 should be expanded.

- Fig. 17: The line of the accelerometer is very difficult to see.

- The conclusions are missing.

- ln. 441: Statement about the data availability: "Not applicable" is not true. There is data. Is it available?

ln. 494: The URL is missing.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. Accelerometer instruments specification have been provided in detail that allow reproducing experiments. Provide the same detail for GNSS instrumentas well as for the antenna (Line 120) i.e. model, number of frequency, sensitivity, etc

2. Describe the manual excitation applied to the benchmark structure (Line 155)

3. There is a slight discrepancy between the results of the accelerometers (Line 400). Is there any possibility of identifying the actual source of noise i.e. satellite signal or double numerical integration

4. A recent review by Wang, et. al "Review of Bridge Structural Health Monitoring Based on GNSS: From Displacement Monitoring to Dynamic Characteristic Identification" provides an overall view of the application of GNSS from monitoring structural health. Author should consider comparing their work in context of the broader development in the field.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf