The Prediction of Abrasion Resistance of Mortars Modified with Granite Powder and Fly Ash Using Artificial Neural Networks
Round 1
Reviewer 1 Report
The abrasion resistance of mortars modified with granite powder and fly ash is studied, and an artificial neural networks model is established to predict the abrasion resistance of mortar, which is meaningful and innovative. However, the writing and organization of this paper is very poor, it is difficult to understand, and many crucial issues have not been fully explained and discussed. If overhaul cannot be carried out, it is difficult for the paper to be accepted. The comments are as follows:
1. The research background should be briefly involved in the Abstract, and there should be quantitative data to show that the model is accurate.
2. The source of the data in figure 1 should be provided.
3. Error bars should be added in figures 4 and 6.
4. The test age of abrasion resistance should be mentioned in the second section. The 91-day test age and 365-day test age shown in figure 5 are very abrupt, which should be explained earlier.
5. The correlation of the curves in figure 5 is very poor, and there is no need to fit, so why not use a bar chart?
6. The connection between figure 5 and context is almost nonexistent, so why does it exist?
7. The discussion of the experimental results should be compared with the relevant literature in order to explain the experimental results.
8. It is obviously that granite powder and fly ash have no change on the abrasion resistance of mortar.
9. The author completely failed to explain the source of the 88 cases that make up the database.
10. Why only seven samples are listed in Table 3?
11. The author should provide the ANN structure model and its parameters, such as whether there is a hidden layer, activation function, weight. We think that the author does not grasp the ANN.
12. It is suggested that the conclusion should be written separately.
Author Response
The Authors would like to thank the Reviewer for they effort in revising Our Manuscript. We believed that the novel version would satisfied the Reviewer and the Editor. Please see the attached file.
Sincerely
Authors.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
1) Lines 115. Authors stated that “Any plasticizer was not used in the mixing process.” But, Figure 5 shows the results as a function of superplasticizer.
2) Authors did not explain test results in Figure 5 at all.
3) Lines 150-158.
- In lines 151, It is very confusing. For example, we do not compare directly test results in Figures 4a (as a function of depth of wear) with test reulst in Figure 6a. - As a same matter, in line 152, Figure 6a (as a function of mass of the sample for FA10,20,30) with Figure 4b (as a function of depth of wear for GP10, 20, 30).
- Also, in line 153, Figure 4c (as a function of depth of wear for FA5GP5, FA10GP10, FA20GP10, FA10GP20) with Figure 6a (as a function of mass of the sample for FA10,20,30).
- Again, in line 155, Figure 5b (No figure 5b) with Figure 6b (as a function of mass of the sample for GP10, 20, 30).
Again, in line 157, Figure 5a (No figure 5a) with Figure 6b (as a function of mass of the sample for GP10, 20, 30).
4) Prediction. Please clarify the number of input variable for the machine learning. In Table 5 and 6, authors count 6 input variables, but it is 5. Depth of wear is output variable.
5) Authors explained influencing parameters to predict abrasion resistance of the concrete surface as hardness of the ingredients and curing conditions. Thus in authors’ ML simulation, at least hardness of the mix ingredients and curing conditions must be included.
6) It is rather reasonable that mass of the sample in Figure 6 could be represented by the average decrease in specimen mass as shown in Figure 7.
7) In Figure 4c, FAGP5 can be denoted by FA5GP5.
8) Line 103-104. Table 1. Caption title must be corrected.
9) Please provide specific gravity of cement, F/A and Granite powder.
Author Response
The Authors would like to thank the Reviewer for they effort in revising Our Manuscript. We believed that the novel version would satisfied the Reviewer and the Editor. Please see the attached file.
Sincerely
Authors.
Author Response File: Author Response.pdf
Reviewer 3 Report
1) I think a review of the structure of the work would be good. For example, shouldn't the methodology used in section 4 be mentioned in section 2?
2) A rigorous revision of the writing is necessary, for example, "Figure 5b and Figure 6b presents" (L. 155).
3) Please use [x1, x2] representation and not [x1], [x2] (eg lines 61 and 91).
4) Observe what is in the description of Table 1!
5) It is necessary enter a space between the numerical value of a quantity and its unit (for example, lines 116 and 125).
6) The presentation of the graph in Fig. 4 (and 6) does not facilitate comparison between different sample results.
7) I don't see much point in making a linear fit like the one shown in Fig. 5. Also, are the initial two points correct?
8) L. 155: There is no Fig. 5b at work!
9) Please check the writing of L. 157.
10) L.186: The text needs adjustments.
11) Eliminate unnecessary space on L. 209.
12) L. 214: Figure 10 and not 7.
13) L. 220: "The linear coefficient R2 represents... Is R2 really a linear coefficient?
14) L. 243: "It was proven that the accuracy of the analysis is at
satisfactory level." I believe that the term "accuracy" is not properly used.
Author Response
The Authors would like to thank the Reviewer for they effort in revising Our Manuscript. We believed that the novel version would satisfied the Reviewer and the Editor. Please see the attached file.
Sincerely
Authors.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Now it looks much better.
Author Response
Thank You for Your effort in revising our Manuscript.
Reviewer 3 Report
1) L. 175: "presented in figure 9"... However, figure 9 does not appear in the text...
2) Please use "Figure" or "figure", but not both, e.g. lines 134 and 179.
3) L. 185: "Figure 5 and 6 present"... Figures 5 and 6 present... This same error also appears on line 196.
4) L. 203: Please use "figure 5b" and not "figure 5 b".
5) In figure 10 use "R2" and not "R2".
6) Please avoid building a paragraph with a single sentence (lines 309 e 310).
Author Response
Thank You for Your comments. We revised Manuscrip:
1) L. 175: "presented in figure 9"... However, figure 9 does not appear in the text...
We carefuly checked the figure numbers and revised the whole Manuscript with such mistakes.
2) Please use "Figure" or "figure", but not both, e.g. lines 134 and 179.
We corrected that
3) L. 185: "Figure 5 and 6 present"... Figures 5 and 6 present... This same error also appears on line 196.
We corrected that.
4) L. 203: Please use "figure 5b" and not "figure 5 b".
We did this.
5) In figure 10 use "R2" and not "R2".
It was revised
6) Please avoid building a paragraph with a single sentence (lines 309 e 310).
We have done it.
