Fatigue Life Predictions Using a Novel Adaptive Meshing Technique in Non-Linear Finite Element Analysis
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsREVIEW
on article
Proposing a Novel Separative Morphing Adaptive Re-Meshing Technology for Predicting the Fatigue Life of FRP Strengthened Pre-cracked Steel Beams using Nonlinear FE Analysis
M. Thiruvannamalai, P. Vincent Venkatesan, Maheswaran Chellapandian
SUMMARY
The paper proposes a new adaptive mesh rebuilding technique, Separative Morphing to predict the fatigue life of pre-cracked steel beam reinforced with FRP using nonlinear FE analysis. The relevance of the paper is due to the fact that such problems of steel beams are an important issue in the design and prediction of the durability of the structure. All this has a high degree of importance for the construction and construction of buildings and structures. The authors applied three-dimensional finite element models, studied the behavior of cracked steel beam by applying constant crack growth and fatigue load based on amplitude. The obtained data are important for future issues of predicting the fatigue life of cracked steel beams. Despite this, the reviewer believes that this paper has critical shortcomings. They are listed below.
COMMENTS
1. It is necessary to remove the abbreviations FRP and FE from the title. They should be replaced with full definitions of the terms. This will be more correct.
2. The abstract is poorly written. There is a description of the research methods and a quantitative expression of the results obtained. However, the abstract should also present the scientific problem, the applied problem and the purpose of the study. The abstract should also formulate a general conclusion about the value for construction science and engineering tasks.
3. The "Introduction" section is presented very briefly and does not contain a full analysis of the current state of the problem. The section mentions 33 sources of literature. But many of them are simply listed, without reflecting the essence of the study.
4. Also, in the "Introduction" there is a very large number of abbreviations. This is somewhat confusing and can negatively affect the interest of readers. Abbreviations should still be written in full wording. But this remark is advisory in nature and can be left to the discretion of the authors.
5. It is proposed to rename Section 2 to "Methods". It would also be nice if the authors changed Figure 5a. Firstly, this figure should be divided into three separate figures. Secondly, it shows graphical dependencies that should be presented in the format of bar charts. This would be more correct.
6. The color scheme of Figure 2 is poorly chosen. The symbols are difficult to read. It is proposed to replace this figure by changing the color scheme to lighter tones.
7. Figure 8a looks logical, and Figure 8b has dots connected by lines. 8b should be brought into line with 8a.
8. A similar remark on the correctness of graphical dependencies applies to Figures 14a, 14b, 17a, 17b, 18a, 18b, 20a, 20b.
9. It is necessary to separate the "Discussion" section into a separate paragraph. The comparison of the obtained results with the results of other authors should be clearly formulated. A more clearly expressed scientific and practical result obtained in this article is needed.
10. The section "Summary and Conclusions" should be renamed to "Conclusions". It looks too large. It is necessary to provide only the formulations of scientific novelty, applied result and prospects for future development of the research. It would also be nice to see more specific recommendations from the authors on the application of the obtained results using real cases as an example.
11. 41 titles of literature in the list "References" do not seem representative enough. The number of analyzed literature should be increased to 55-60.
12. Sources 4, 5, 6 raise some doubts. Unfortunately, the dates of these sources are very old. There is a possibility that this information may not be relevant at the moment. Since science is already at a new stage of its development, the list should be supplemented with works from the last 5 years. This will help this article to become more relevant and modern.
Overall, the reviewer believes that this paper presents interesting results, but their validation is unclear. The novelty is also not stated properly.
Comments on the Quality of English Language
Minor editing of English language required.
Author Response
The authors thank the reviewer for the positive comments and valuable suggestions to enhance the quality of our work. As suggested, the following comments were addressed and filed as attachment using following link. Also, the comments given are incorporated in our revised submission.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsOne of the objectives to be achieved from this manuscript is to "Predict the Fatigue Life of the FRP Strengthened Pre-cracked Steel Beams".
But the contents of this manuscript do not provide real evidence and descriptions of "predicting the fatigue life".
In the experimental program, the specimen should be loaded with a fatigue load based on the fatigue loading history by applying several different stress levels, to:
1). Control steel beam (without pre-cracked).
2). The steel beam (with pre-cracked)
3). The FRP strengthened pre-cracked steel beam.
It should be the final result of this experimental work is the hysteretic loop for several different stress levels. And based on these hysteretic loops, it is possible to predict the fatigue life for 1), 2) and 3). Then, all test results can be compared with each other. It can also be seen the influence of the FRP strengthening of the pre-cracked steel beam.
After that, validation can be carried out using a proposed novel SMART (Separative Morphing Adaptive Re-Meshing Technology). It can also be compared if you use the usual (default) meshing from Ansys. Explain what the difference is between the two.
After validation is complete, the proposed novel SMART (Separative Morphing Adaptive Re-Meshing Technology) needs to be verified again against the results of experimental work from several different literatures to review or asses the reliability of the proposed novel SMART.
Comments for author File: Comments.pdf
Improve the English Language
Author Response
The authors thank the reviewer for the positive comments and valuable suggestions to enhance the quality of our work. As suggested, the following comments were addressed and incorporated in our revised submission. The details can be found in the attachment below.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsFrom the results of the parametric study, it is evident that the FRP’s tensile stress is 553
maximum for the beams reinforced with FRP having longitudinally oriented fibers 554
(=0º) and it is minimal for those reinforced with FRP having transversely oriented 555
fibers, which helps to the delay the crack growth, reduce SIF, and increase fatigue life 556
for all the reinforced beams till the last step of crack growth.
Author Response
Question: From the results of the parametric study, it is evident that the FRP’s tensile stress is maximum for the beams reinforced with FRP having longitudinally oriented fibers (q=0º) and it is minimal for those reinforced with FRP having transversely oriented fibers, which helps to the delay the crack growth, reduce SIF, and increase fatigue life for all the reinforced beams till the last step of crack growth.
Response to reviewer comments:
The authors completely agree with the observations made by the reviewer. The orientation of FRP laminates along the direction of loading is considered a parallel alignment (i.e, q = 90º). Similarly, the orientation of fibers perpendicular to the direction of loading is considered a horizontal alignment (i.e. q = 0º).
From the Figure 14 (in revised manuscript), it can be seen that the crack propagation rate is smaller for beams CB02 and GB02 and higher for beams CB08 and GB08, which gradually increases from B02 to B08 irrespective of the type of FRP. Therefore, the FRP laminate is highly effective in reducing the SIF when the fibers are oriented in the longitudinal direction (i.e, q = 0º), and is least effective when the fibers are oriented parallel to load application (i.e, q = 90º).
The crack growth retardation as per the semi-empirical approach based on Paris law is reflected in Figure 17 of revised manuscript, which depicts the SIF distribution over crack extension for CFRP and GFRP reinforced beams. From the results, it can be observed that the use of carbon/glass FRP could help in reducing the stress intensity factor with the fibers oriented in a perpendicular direction (i.e., q = 0º). However, with the increase in the orientation of fibers, the SIF was found to increase in almost linear proportions corresponding to the increase in crack extension (Δa).
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsI recommend the article for publishing.
Comments on the Quality of English LanguageMinor
Author Response
The authors thank the reviewer for the positive comments and valuable suggestions.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have not addressed my previous comments, which include:
I). In the experimental program, the specimen should be loaded with a fatigue load based on the fatigue loading history by applying several different stress levels, to:
1). Control steel beam (without pre-cracked).
2). The steel beam (with pre-cracked)
3). The FRP strengthened pre-cracked steel beam.
It should be the final result of this experimental work is the hysteretic loop for several different stress levels. And based on these hysteretic loops, it is possible to predict the fatigue life for 1), 2) and 3). Then, all test results can be compared with each other. It can also be seen the influence of the FRP strengthening of the pre-cracked steel beam.
II). After that, validation can be carried out using a proposed novel SMART (Separative Morphing Adaptive Re-Meshing Technology). It can also be compared if you use the usual (default) meshing from Ansys. Explain what the difference is between the two. This method will also be able to prove whether a proposed SMART novel gives better results or not.
The latest comments are:
Line 151: Mention the size of pre-crack given to the steel beam.
Line 275: Experimental setup for Fig. 3(a) and Fig. 5(b) uses the four-point bending test. This is not the same as what is shown in Fig. 6(a). Clarify it.
Line 314: The crack pattern generated by the FE analysis result is not similar to the experimental results.
Line 315: Provide also the image of fatigue testing on steel beams strengthened with FRP. And the actual comparison of the failure modes that occur between steel beams without FRP and with FRP.
Line 336: Show also the original image of the fatigue test results.
Line 351: The fatigue loading history has been established in this manuscript, so the fatigue test should be based on different stress level values, which will be finally obtained the hysteresis loop curve and fatigue life prediction of the steel beam under this study.
Line 385: All crack patterns resulted by FE analysis should be compared with the experimental results in order to evaluate whether the FE analysis results are reliable or not.
Line 576: Rearrange after revisions.
Comments for author File: Comments.pdf
Improve it
Author Response
The authors thank the reviewer for the positive comments and valuable suggestions.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsAccept
Author Response
The authors thank the reviewer for the positive comments and valuable suggestions.
Author Response File: Author Response.pdf
Round 3
Reviewer 2 Report
Comments and Suggestions for AuthorsLine 20: The purpose of this study is not appropriate. Actually, a more appropriate research objective in this study is to test the feasibility of applying the FEM approach involving the proposed SMART to predict the crack growth, fatigue life and SIF on pre-cracked steel beams strengthened by FRP.
Line 27: This sentence seems to say that the FEM approach used can prolong the service life. In fact, what can extend the service life is the use of FRP strengthening on pre-cracked steel beams. FEM is only used as a tool for prediction. Please correct it.
Line 32: Add some literatures related to "the FRP strengthening on the steel beams".
Line 115: Add some literatures related to "the strengthening steel beams using FRP under fatigue loads". There is quite a lot of literature that studies it, including:
1). A review on strengthening steel beams using FRP under fatigue.
2). Experimental study on the fatigue behavior of steel beams strengthened with different fiber-reinforced composite plates.
3). Fiber-reinforced polymer strengthening of steel beams under static and fatigue loadings.
4). Fatigue strengthening of cracked steel beams with different configurations and materials.
5). Fatigue crack growth in steel beams strengthened by CFRP strips..
6). Fatigue strengthening of cracked steel beams with CFRP plates.
7). Fatigue design of CFRP strengthened steel members
8). etc.
Line 150: The authors' response states that "The current study is based on numerical simulations and theoretical models rather than experimental work". The authors should explain the description of the theoretical model in question and how to use it to predict the crack growth, fatigue life and SIF.
Line 156: In the analysis results, it should be reported that the application of "the proposed SMART" shows the FEM simulation process is faster compared to the use of default meshing. In addition, the crack growth prediction is also more accurate.
Line 171: Equations 1-7 are actually the formulas used by Ansys for modeling purposes.
There are 2 things that need to be clarified by the authors, those are:
1). In which section have the authors described and outlined the theoretical models in this study. If what is meant is Equations 1-7, then this is not approriate, because these equations are the approach methods used by Ansys.
2). If the authors do propose SMART as a novelty in this study, there should be an algorithm displayed in the manuscript which will later be used as a sub-routine that is synchronized on Ansys, so that it can be used for modeling purposes.
Line 251: Show also the analytical equations for "the steel beams strengthened by FRP", so it can be seen the difference.
Line 317: If a higher SIF number is obtained, does it guarantee more accurate analysis results? Clarify it.
Line 324:
Fig. 5(a): 1). Explain and describe in the manuscript, how to get the figures on this bar chart.
Especially for "analytical Dunn", display the calculation procedure in a table. Display the related parameters and numbers used or involved in the calculation, so that the calculation results can be understood and their accuracy can be checked.
2). Compare the results of the "present study" and "analytical" to the "experimental" in a table. Calculate the error in percentage and explain it.
Fig. 5 (b): Validation should be done against experimental results from several literatures. Perform FEM analysis with the existing default meshing and with the proposed SMART. Compare the analysis results for the prediction of the crack growth, fatigue life and SIF.
Fig. 5(b): One of the purposes of using the proposed SMART is to minimize the computing time. So it should also be possible to display a comparison of the computing time of the proposed SMART with the default meshing based on the FEM analysis in the reviewed literature.
Fig. 5(c): There is no citation information on this image. So it seems as if this experimental work was done by the authors.
Fig. 5(c): The collapse character is like a result of monotonic load. Dominant crack occurs until it reaches the upper flange. If the steel beam is loaded cyclically, usually the crack length can be controlled, because the actuator machine can be stopped at a certain cycle.
Fig. 5(d): 1). The crack pattern generated by the FE analysis result is not similar to the experimental results.
2). Display the crack pattern or failure mode of the steel beam based on the FEM analysis results, for 1 full steel beam.
Line 336: 1). Find the experimental results for the pre-cracked steel beams without FRP strengthening (at least 5 different literatures). Provide and display the crack growth and crack pattern (or failure mode), fatigue life and SIF.
2). By using the proposed SMART, the FEM analysis can simulate and predict the crack growth and crack pattern (or failure mode), fatigue life and SIF for No.1. And then, compare the FEM analysis results to the experimental results. The proposed SMART can be said to be reliable and accurate if a small error is obtained.
Line 367: 1). Find the experimental results for the pre-cracked steel beams with FRP strengthening (at least 5 different literatures). Provide and display the crack growth and crack pattern (or failure mode), fatigue life and SIF.
2). By using the proposed SMART, the FEM analysis can simulate and predict the crack growth and crack pattern (or failure mode), fatigue life and SIF for No.1. And then, compare the FEM analysis results to the experimental results. The proposed SMART can be said to be reliable and accurate if a small error is obtained.
Line 552: Explain why the SIF value of CB02 is greater than the others.
Line 553: Explain why the SIF value of GB02 is greater than the others.
Line 602: Rearrange after revisions
Comments for author File: Comments.pdf
Improve it
Author Response
The authors thank the reviewer for spending his valuable time and putting in repeated efforts as a result of which the paper has shaped up really well when compared to the original submission. We are very grateful for the insightful comments for the last three rounds of reviews. The reviews were highly encouraging and have been considered in the revised manuscript.
Author Response File: Author Response.pdf
Round 4
Reviewer 2 Report
Comments and Suggestions for Authors1). Actually, this research would be better and stronger if the results of the FEM analysis were more compared to several experimental results from related literature. So that the reliability of the proposed FEM can be better.
2). Author's response: "Similarly, the analysis takes more than 2 hours for the model with standard meshing but is reduced to minutes when using SMART adaptive meshing. Therefore, SMART is also computationally efficient" is still just a statement. There should be authentic evidence that can be shown.
3). All authors' responses and explanations should be included in the manuscript.
Comments on the Quality of English Languageminor
Author Response
The authors thank the reviewer for spending his valuable time and putting in repeated efforts as a result of which the paper has shaped up really well when compared to the original submission. We are very grateful for the insightful comments for the last three rounds of reviews. The reviews were highly encouraging and have been considered in the revised manuscript.
Author Response File: Author Response.pdf
