Prediction of Viscosity Values of Nanofluids at Different pH Values by Alternating Decision Tree and Multilayer Perceptron Methods

Round 1

Reviewer 1 Report

The authors present the prediction of viscosity of nanofluds using computer intelligence approach. Especially two methods, ADT and MPM wer compared.

The manuscript must be substantially re-written focusing on discussion and comparison of the results from two methods. Most of the results on the nanofluid are validation whether particles were properly synthesized. Also most of the analysis methods are described in the level of textbook, which are unneccessary for this journal readers.

Give more literature survey on the nanofluid property prediction using computer intelligence. I have reviewed another paper on the similar topic.

Focus more on the results from computer intelligence. reduce the nanofluid analysis part.

Figure 12 and 13 are not properly presented. Horizontal axis must be one of the variables. The current presentation does not give any helpful information. Additionally, if one method is better than the other, more discussion on the reason must be given.

there more more to things. but I will discuss when I receive the revised version.

Author Response

Thank you very much for your suggestions. We tried to fix the errors you specified in the corrections given below. Our respect.

We send the corrections we made in the article as word files.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript "Prediction of Viscosity Values of a Nanofluids at Different pH Values by Alternating Decision Tree and Multilayer Perceptron Methods" by Das et al. describes the measurement of nanofluid viscosities. It can be published following a major revision, addressing my following comments:

The way the paper is presented in very confusing. Figure, 6 to 10, which represent the characterization of produced particles should be shown in the supplementary, or at least grouped together.

The process of calculating viscosity is described in Figures 4 and 11 in different ways. If this is the main point of the paper, this should be improved.

You need to add a simple yet informative schematic figure to present all variables you need to calculate the viscosity. For example, I am not sure why do you need to measure the average velocity of the nanofluid, as you have not made it clear.

You need to compare your results with the literature. How do you know about the accuracy of your model?

Author Response

Thank you very much for your suggestions. We tried to fix the errors you specified in the corrections given below. Our respect.

We send the corrections we made in the article as word files.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript of Prediction of Viscosity Values of a Nanofluids at Different pH Values by Alternating Decision Tree and Multilayer Perceptron Methods is interesting article.

I commented as follows;

1.(major)

Nanofluids are very important for many applications.

Moreover, nanofluids exhibited non-Newtonian viscosity owing to their suspension rheology.

The author should be non-Newtonian viscosity of nanofluids (viscosity plotted against strain rate).

2.(major)

Reynolds number was defined as equation (1).

However, the equation should be used for Newtonian viscosity.

3.(minor)

The many symbols and letters were used.

The author should summarize them as nomenclatures.

Author Response

Thank you very much for your suggestions. We tried to fix the errors you specified in the corrections given below. Our respect.

We send the corrections we made in the article as word files.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

A lot of improvements from the previous version can be seen.

However, there are still more things to be clarified and improved.

Some wordings and expressions seem awkward. For example, line10, "enhance heat transfer development"...I think removing "development" better. Line11: viscosity determines the rheological behaviors? viscosity itself is a rheological property. That sentence sounds awkward to me.

A great improvement in the literature survey in the introduction! However, this literature survey must provide why you choose this particular system and particular approach in this work. For example, Ref [13] also performed MLD prediction on nanoparticle solutions. What is new in your work compared to Ref 13? Why did you choose to compare MLD and ADTree, particularly?

Fig 2 is not needed. This is only necessary if you made a custom set up. Just report which model of which brand of this viscometer you used. 

Fig 3. Diagram would be more helpful to the reader. Fig3 can be in Appendix or Supplement materials.

Line 192~199 are not needed. This is not a fluid textbook.

Line222 and 223 yi should be y_i. please check the styles.

What are LM1, LM2, LM3... in this figure?

The description of Fig 6 should be "Experimental measurement of viscosity of CuO solution as a function of T with various pH values and wt% values"

Please include Viscosity vs Re or Viscosity vs average velocity. Whether this system shows non-Newtonian behavior or not is important.

Why did you choose the input variables Re, average velocity, and density? As you indicated, Re is a function of velocity and density... Shouldn't the input variables be independent each other as much as possible? 

I still don't understand the presentation of fig 7 and fig 8. The vertical axis must give some information. For example, there are a certain set of data numbers which seem to be measured in a certain variable changes. Such as data set 1 to 9...this particular set is where the most evident discrepancy between ADTree and MLD. Which variables were tested in that set? That will give more valuable information such as ADTree is not in predicting in such condition ...   

Author Response

First of all, thank you very much for your corrections and suggestions.

Corrections and answers are presented in addition.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript can be accepted in its current form.

Author Response

Thank you very much for your interest and suggestions.

Reviewer 3 Report

The revisions are satisfied as my comments.

Author Response

Thank you very much for your interest and suggestions.