Damage Detection of Gantry Crane with a Moving Mass Using Artificial Neural Network

Round 1

Reviewer 1 Report

The research objective of this paper is meaningful, and the presentation of the paper is very detailed. However, the current version of the paper is difficult for readers to discover innovative points. The paper is not suitable for publication without thorough revisions. The main issues are listed below:

1. This paper seems to only establish a two-dimensional in-plane (vertical and horizontal direction) finite element model of the gantry crane but in fact, the dynamic behavior of out-of-plane is also important. Please supplement the simulation cases of finite element including out-of-plane or explain why the out-of-plane cases should be ignored.

2. Section 2.3, although reference 1 may only present the time history results of horizontal displacement, the reviewer suggests providing the calculation results of vertical displacement, stress, and acceleration in this paper. Even without references for comparison, it can still be used to help readers understand the effectiveness of finite element simulation results.

3. Page 7, Line 216, where is the ‘Appendix’?

4. Page 8, Whether the moving speed of the moving mass block (m23) has influence on the structural response and the defect’s detection results needs to be discussed. It is suggested adding multiple commonly used classes for different moving speeds.

5. Perhaps adding a framework graph to Section 3 would be more effective in helping readers understand the process of defect detection.

6. Page 9, Line 274, ‘Nonlinearity Detection’ is not a common word in structural engineering, and its explanation here should be more detailed. 

7. Page 9, Line 280, not all readers of Buildings understand the meaning of ‘Robustness’. Please cite some relevant references here to help the authors understand its meaning, such as: https://doi.org/10.1109/TIM.2023.3343742

8. The readability of Figure 6 is poor. It is recommended listing the calculation formulas of LVQ neural network in the text separately, and improving the quality of this figure.

9. Figures 7 and 8, the lines in the graphs cover each other, please use different line shapes (e.g., solid line, dotted line, dashed line, and dash-dotted line) for different classes. Especially for Figure 8, it may be better to display more detailed comparison by adding a small window of graph.

10. The section of Conclusions only consists of a long paragraph. It is suggested discussing it point by point by several paragraphs.

Minor editing of English language is required.

Author Response

The response to reviewer 1 is attached.

Author Response File: Author Response.pdf

Reviewer 2 Report

This article uses the finite element method and neural network method to identify defects in the structure of gantry cranes. The author first used the finite element method to establish a dynamic model of the gantry crane, divided different parts of the structure, simulated structural defects by changing the stiffness parameters of different parts of the structure, and graded the degree of defects in different parts for structural analysis. This article utilizes a learning vector quantization neural network for training and defect identification while introducing noisy data to simulate environments closer to reality. The research results indicate that the proposed method can identify the location and degree of defects in the structure of gantry cranes.

1. The literature cited in lines 87 to 89 of the article mentions that the combination of AR models and ANNS is an effective tool for dam age classification and estimation. The previous and subsequent content is related to the study of cranes, which is somewhat abrupt. Please make modifications.

2. The article mentions from lines 223 to 224 that the simplification of the finite element model resulted in displacement errors compared to the original model. Please provide a detailed introduction to these simplifications. For example, how much the number of units has decreased and how the shape has changed.

3. The damage level of the model is divided in lines 248 to 254 of the text. May I ask if the author has any basis for dividing the degree of injury?

4. The 328th line of the article mentions that according to Figures 7 and 8, it can be seen that as the severity of the defect increases, the peak in the figure shifts to the left. Can the author explain why this phenomenon occurs?

5. The content from lines 327 to 331 in the text is extremely similar to the content from lines 341 to 344. Please make deletions and modifications.

6. Several relative studies should consider to be mentioned: Combined joint and member damage Identification of semi-rigid frames with slender beams considering shear deformation, Buildings. Damage identification of large-scale space truss structures based on stiffness separation method, Structures. Parameter Identification of Frame Structures by considering Shear Deformation, International Journal of Distributed Sensor Networks. Non-destructive damage evaluation based on static response for beam-like structures considering shear deformation, Applied Sciences.

7. Lines 372 to 378 of the text mention that as the severity of defects increases, the success rate of defect identification also increases. However, according to the data in Figure 9, such a conclusion cannot be drawn whether in the absence or presence of noise. Please make modifications to this section.

Author Response

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper presents a method for the damage detection of gantry crane system with a moving mass using neural network. The paper will contribute the pertinent literature.

The English writing of the paper should be improved.

Author Response

The response to reviewer 3 is attached.

Author Response File: Author Response.pdf

Reviewer 4 Report

At the beginning of the article, the authors state that the problem solved in this article has received little attention in the literature. This may indicate either that this problem is not relevant, or that the authors simply did not find literature on this problem due to an insufficiently deep search. In the first case, one should doubt the relevance of this article, since, perhaps, there is no need to study the wear of cranes of this type, since they are already quite reliable, and the accumulated defects either do not lead to accidents or make themselves felt much earlier than a breakdown occurs. In the second case, the authors should still work on searching the literature.

However, the authors further provide links to confirm their statements, although, apparently, these links, such as [6-11] and [12], are not related to this type of equipment. The relevance of modeling the state of this type of equipment is not confirmed by the reasoning given in the article. It can be assumed that this type of technology is quite reliable. To prevent an accident of such a crane, it seems that the best solution would be to regularly inspect its actual condition. This inspection can be carried out visually, also by monitoring the condition of bolted and welded joints, and also, for example, by the acoustic method, creating sound pulses using vibration or sound sources, or using blows at control points, followed by sound at other control points, with taking into account the amplitude, phase, nonlinear distortions and delays, it is possible to reliably distinguish serviceable equipment from faulty equipment of the same type, since a violation of the strength of bolted or welded connections will certainly lead to changes in the acoustic properties of the structure. At the very least, such a method should be considered as an alternative and the advantages of the proposed method should be explained.

The conclusion of the article is written more like an abstract should be written. In conclusion, it is customary to present specific scientific results and report exactly how they can be used, and not only in the problem being solved, but also in other similar problems. It should be explained what the scientific novelty and practical usefulness of these results are, and how the reliability of these results is ensured.

The statement in the conclusions that the important problem solved by this article is often overlooked is not justified. To make such a statement, you should carry out a detailed search of the literature and prove that this is exactly the case, with bibliographic references. Otherwise, this looks like an unfounded accusation of absolutely everyone, and the authors pitting themselves against all other researchers.

This article needs to be edited to clearly present the novelty and contribution of this research to science.

Author Response

The response to reviewer 4 is attached.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Good revision. Minor revision needs to be done before recommending publication:

1. Line 203, there are two duplicate FEM here.

2. Lines 313~314, the authors have added enough explanation for ‘Nonlinearity Detection’. However, some related references need be cited to support the statement ‘dynamic behavior of complex systems often exhibits intricate, non-linear patterns that may not adhere to conventional linear relationships’.

3. Line 398, please give ‘m23’ the correct form of writing (whether it is subscript, whether it is italicized, whether it is bold?). Similar issues in other places of the whole paper also need to be corrected together.

4. The effects of temperature on the deflection of the Gantry crane may influence the dynamic behavior of the system of gantry crane–moving mass. It is suggested adding some prospects of research at the end of the Conclusions about this topic. 

Minor editing of English language is required.

Author Response

Attached

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript has been revised, and the quality has been improved.

Author Response

The authors are grateful for the valuable and helpful comments from esteemed reviewer 2.

Reviewer 4 Report

After corrections have been made by the authors, the article can be published

Author Response

The authors appreciate the thoughtful and constructive feedback provided by the respected reviewer 4.