Development of New Algorithm for Aniline Point Estimation of Petroleum Fraction

Round 1

Reviewer 1 Report

Title: Development of new algorithm for aniline point estimation of petroleum fraction

The paper is related to the estimation of the Aniline Point using other oil parameters/measured values.

The paper is interesting, but it needs of several improvements. The English is poor and it needs a revision from a native English speaker.

Some references are missing and some parts of the text must be moved in the appropriate paragraphs (see my detailed comments).

The data of the 127 samples that the authors used to validate the model and of the 142 points used to calibrate the model have to be reported in the paper (may be in the appendix), so readers and researcher can use these values for repeat/validate the experiments and they can be used as future references.

The author have to verify the parameters obtained using the 142 data. What happen if a part of these values are removed and then estimate if considered unknown (cross-validation)? What is the parameters stability if we add or remove some data? To validate the parameter obtained a sensitivity analysis will validate the proposed method.

Detailed comments:

Line 9: the aniline point is not devoted to “characterising the solubility of petroleum fractions”.

Line 12 characteristic parameters of what?

Line 13: the iteratively modified algorithm of what?

Line 15: the sentence is not correct, need of an English revision.

Line 23: oil? What kind of oil?

Line 25: reference about AP?

Line 34: national standard of Cina?

Line 48: new ideas? Please clarify

Line 54-55 add references.

Line 55-56: too poor affirmation. Please justify and explain because it is insufficient. What can be consider sufficient from you point of view?

Line 56-57: the sentence “For example, the number of substances to be assessed is excessively small, and the evaluation is too simple” need to be carefully revise and you have to justify your affirmation.

 Line 57-58: add reference or give more information. Scholars?

General comments about all the presented methods: each variables of the presented formulas have to be explained.

Line 71:74: the explanation of the methods need improvements, from a technical and English style point of view.

Line 114 measured value: please explain if you made the measure or if you just all the data from the producer/literature.

Table 1 reports ARE and AAE. What is AAE?

Line 155 ARE is ARD? Pay attention, the text use ARE and ARD along the paper, but I think it is a refuse.

Line 124: respectively of what?

Line 125: the causes of the deviations were analysed. Where?

Line 126 to 136. The discussion about the cause of the deviation is very poor. Just for example, you cannot justify the deviation by using: “the method is too simple.”

Paragraph 3.1: the description of the new proposed method is confused. Line 145 to line 152 can be moved in the introduction, as here the paragraph must be focused on the description of the proposed method and not about previous methods.

Line 153: the formula is wrong. You use the same function for the 3 different parameters?

Line 164-165: the functions must be renamed

Line 166: which article?

Line 202: the paragraph title must be “data source evaluation”

Line 230-231: the information is repeated at line 218-219

Figure 5:plotting the AP versus the oil number not have any sense. It is confusing the reader by giving a “virtual” trend to the oil number, that in principle is random. I guess that the trend depend on other parameters (like density or other parameter of the considered oil).

Conclusion: what about the MRD of the proposed method?

Symbol description AAE I think must be AAD

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper under review considers the issue of development of new algorithm for aniline point
estimation of petroleum fraction

In the reviewer’s opinion, in general, the paper is quite interesting.
However, there are several important aspects that require authors comments
or possibly improvements:

1) Equation (1)-(2) shold have the same font and fromat like equations (3), (4) etc.
2) Line 72: "gravity [11].According" shold be "gravity [11]. According"
3) Figure 1. x-axis is not clear. Please format.
3) Equation (3) should be formatted. This equation is divided into two parts and these parts should be aligned. The equation number should be in the middle of these two parts.
4) In equation (8) the brackets are too small
5) Line 104 - Move to the next page.
6) It is proposed to place a description of each element in the equation under all
equations or enter an explanation of abbreviations at the beginning of the article
7) Table 1, row 4 should be removed.
8) Line 218-220. Please format.
9) Line 282 - 298. Please format.
10) At the beginning of the article, the authors mention neural networks, genetic algorithms, etc. that help solve the problem. Then the authors give analytical models which then compare with their own method. Why don't the authors make a comparison with neural networks and / or genetic algorithms? Even if they are better, you can show such results. Because when reading the article, such a question is obvious.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The author fully replied to the answers/suggestions of the reviewers.

Detailed comment:

Line 78-79: “the model of Shoudeqing-toward positive method is too simple, although the original data is more but the model extrapolation is not strong;” please avoid the use of is “too simple” and the “original data is more”.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The answers to the questions are satisfactory, however, the article must be corrected from an editing point of view for example, each equation should be in the middle and the equation number on the right (at the right margin). Authors should format the entire article carefully before resubmitting. The introduction of a response to the reviewer's comments without a comprehensive formatting of the article does not show that the authors put themselves to work. If we look at the equation numbers, they should all be in line.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

The answers to the questions are satisfactory, however, if authors make editorial corrections they should look at the entire article and not at specific
comments, e.g. the equations are correct but e.g. figure 1 is in bad quality!

In addition, in Table 2, 3 you can write e.g. K1_1 as K1_1 where the second digit is in the subscript. Then all the numbers in the  row will be on one line.