Orthogonal Experiments and Neural Networks Analysis of Concrete Performance
Round 1
Reviewer 1 Report
i have attached file to this email
Comments for author File: 
Comments.pdf
Author Response
Reviewer 1
- An outline of the paper at the end of the introduction section is recommended.
Reply: An outline has been added to the last paragraph of the introduction.
- Did the authors compare the results of their study with the work of other authors?
Reply: In the introduction, I have quoted the latest article, that is, why should we use fly ash as a substitute for cement, and the application of neural network and orthogonal test in concrete design. It is concluded that we can use the advantages and disadvantages of neural network and orthogonal test to study these problems, which are the comparison and summary with previous workers.
- The conclusion is brief and does not cover all aspects of work
Reply: The conclusion is supplemented on the basis of the original conclusion, which has covered all research work and content.
- Some important findings can be presented in Abstract.
Reply: The Abstract has been revised and some important findings and conclusions have been refined.
- The innovation of this paper is not highlighted. Explain clearly what the latest progress and previous work of this paper are based on?
Reply: The introduction puts forward: (1) The traditional method of establishing the strength model of concrete design is not suitable, because the interaction between strength factors is difficult to deal with. In order to solve this problem, we adopt the method of orthogonal experimental design (OED). This method can get all combinations of each factor at its high and low levels. Due to many factors, only a few experiments are needed to estimate the main effects and simple interactions.
- Compared with traditional statistical methods, neural network method is simpler and more direct, especially in establishing nonlinear multivariable relationships. However, there is little research on the strength model of high fly ash and mixed alkali concrete based on neural network and OED.
Based on the advantages of the above technology, this study uses neural network and orthogonal test method to study the influence of fly ash replacing cement on the strength and other properties of concrete.
- The authors need to clarify about Lippmann formula
Reply: Already clarified in brackets，(Lippmann formula is, for the first training cycle, the random weight is allocated to the connections between units for training until the average sum of squares error of all training modes is minimized)
- Each graph must have a separate caption
Reply: Each drawing has been assigned a separate title
Author Response File: 
Author Response.docx
Reviewer 2 Report
The article addresses an important and very interesting topic of the orthogonal experiments and neural networks analysis of concrete performance, which is appreciated. The study includes the experimental research. In this paper studied the potential of using OED and ANN to study the effects of mixed alkali and fly ash replacements from 0 to 30%, from 3 to 28 days, Alkali content, from 3 to 8%, at water-binder ratio ratios of 0.5 on the early (3 days) and late (28 days) compressive strength. The Reviewer has some concerns regarding the introduction, results, conclusions and references. Generally, in this paper the English language should be improved (some sentence should be more clearly - too long). Furthermore, check the text of Native Speaker and please check all paper. In opinion of Reviewer this paper should be subjected to major revision.
Other comments:
1. Introduction
Please explain more detailed the aim of this research. In addition, please check in all paper the passive voice. Please describe the impact of your research on the bridge, civil engineering stricture.
2. Results
Please improve the Figures (Graphs) and Tables with results, because in current version are not clear.
3. Conclusions
In current version are really poor. The Reviewer cannot see the most important conclusions from your paper (please use bullets).
4. References
This part should be improved. Generally, the scientific paper should be based on the literature from all world, thus please check the literature from the best Journals.
Finally, I hope that my comments will be helpful for the authors.
Author Response
Reviewer 2
Point 1: 1、Introduction
Please explain more detailed the aim of this research. In addition, please check in all paper the passive voice. Please describe the impact of your research on the bridge, civil engineering structure.
Response 1: (1) In the last paragraph of the introduction, the purpose of this study has been added, namely: This paper studies the influence of alkali content and fly ash replacement on the compressive strength of concrete through experiments. Based on the existing test results, taking full account of the advantages and disadvantages of orthogonal test and neural network,The potential of OED and artificial neural network are explored to study the influence of fly ash replacement from 0 to 30%, from 3 to 28 days, and alkali content from 3 to 8% on the early (3 days) and late (28 days) compressive strength.
(2) Some passive voices have been checked.
(3) In this paper, the influence of fly ash and alkali content on concrete strength is studied by orthogonal test and neural network：
1) The prediction of strength avoids long-term indoor tests, saves raw materials, and has economic benefits; 2) Through the primary and secondary factor analysis of the orthogonal test, we can know the main influencing factors affecting the concrete strength, and can guide the indoor test; 3) The results show that adding alkali can stimulate the early strength, which is of great significance for accelerating the progress of concrete in civil engineering structures such as bridges. Because it is necessary to use early strength concrete in bridge design, because it must reach a certain strength during construction before the next construction, or because the weather conditions are not suitable for long-term maintenance.
To sum up, it has an important impact on civil engineering structures such as bridges.
Point 2: 2.Results
Please improve the Figures (Graphs) and Tables with results, because in current version are not clear.
Response 2: All graphics and tables have been modified and processed. The graphics are vector graphics, and the font, symbol and other information are modified in the table.
Point 3: 3.Conclusions
In current version are really poor. The Reviewer cannot see the most important conclusions from your paper (please use bullets).
Response 3: All work contents in the conclusion have been supplemented; And added the number of the project; The conclusions of laboratory test, orthogonal test and neural network are sorted out.
Point 4: 4.References
This part should be improved. Generally, the scientific paper should be based on the literature from all world, thus please check the literature from the best Journals.
Response 4: the latest and best journals in various regions have been added and modified.
Round 2
Reviewer 1 Report
The corrections have been made.
Author Response
Reviewer 1
Responses 1： has been corrected
Reply 1：thansk for your suggestion
Author Response File: Author Response.docx
Reviewer 2 Report
Thank you for your improving. The Reviewer have still concernes to the paper. Generally, the paper is very careless (please check the template of this Journal) and the Figures (fig. 1- 4) should be improved. In the figures please correct the scale of the "flexural strength" and visual presentation of results.
Finally, I hope that my comments was helpful to the Authors.
Author Response
Response to Reviewer 2 Comments
Reviewer 2
Point 1: Thank you for your improving. The Reviewer have still concernes to the paper. Generally, the paper is very careless (please check the template of this Journal) and the Figures (fig. 1- 4) should be improved. In the figures please correct the scale of the "flexural strength" and visual presentation of results.
Response 1: (1) It has been modified according to the template.
(2) Figures have been improved (Fig. 1-4). The coordinate axis content is simplified in the figure, and the font of the coordinate axis is unified. Already correct the scale of the "flexural strength" and visual presentation of results.
Author Response File: Author Response.docx
