On Predicting the Content of Paraffin and Resin-Asphaltene Substances in Oils
Abstract Generated abstract
This paper examines whether a limited set of routine physicochemical indices can be used to predict paraffin and resin-asphaltene contents in crude oils, reducing the need for extensive laboratory characterization during exploration and process design. Using experimental data for oils from eastern regions of the Soviet Union and Tomsk, the authors apply correlation and regression analysis, test normality and correlation reliability, and compare relationships among density, viscosity, sulfur content, pour point, coke residue, molecular weight, and distillation yields. The analysis identifies pour point as the main predictor of paraffin content and coke residue as a practical predictor of resin-asphaltene content, yielding simple linear equations intended for approximate practical estimation.
Full Text
On Predicting the Content of Paraffin and Resin–Asphaltene Substances in Crude Oils
G. G. Blagopoluchnaya, A. F. Fedorov
(Presented by the scientific-methodological seminar of the organic departments of the Faculty of Chemical Technology)
The expansion of the area in which crude oil is used as a feedstock for petrochemistry and for the production of a wide range of fuels and oils leads to an increase in the number of indices used in assessing the quality of crude oils and serving as the basis for designing technological processes. This is associated with carrying out a large number of lengthy, costly investigations. A more complete description of the properties of crude oil is needed already at the stage of exploration of oil fields, as applied to the various directions of its use. However, the introduction into the set of oil analyses of a large number of indices leads to excessive complication and higher cost of exploration and research work, as well as to an increase in the time required to carry it out.
In the present work, an attempt has been made, using methods of mathematical statistics, to determine the presence of linear relationships between certain properties of crude oil and to select a comparatively small number of indices by means of which, with accuracy sufficient for practical purposes, crude oil can be characterized as a feedstock for the principal directions of its use.
The main difficulty in the quantitative investigation of indices of oil properties is the need to take into account the simultaneous influence of a large number of factors. Such influence, from the mathematical point of view, may in some cases be regarded as random and evaluated from the standpoint of probabilistic-statistical theories. The problem posed is solved by means of correlation and regression analysis of experimental data. In the present work the possibilities of predicting the content of paraffin and resin–asphaltene substances (RA) from the principal physicochemical characteristics of crude oils have been determined: specific gravity \((d_4^{20})\), molecular weight \((M)\), viscosity \((\nu_{20})\), coke residue \((K)\), pour point of the oil \((T)\), total sulfur content \((S)\), and the yield of fractions up to 200 and up to \(300^\circ\mathrm{C}\).
The initial data for analyzing the linear relationships between the indicated indices were materials from studies of oils of the eastern regions of the Soviet Union [1, 2] and data on Tomsk oils obtained in the Laboratory for the Comprehensive Utilization of Peat of the Tomsk Polytechnic Institute.
In most cases, with a sufficiently large sample size of the initial data, the results of laboratory analyses are subject to
Table of pair correlation coefficients
| Parameters | $d_4^{20}$ | $\gamma_{20}$ | $S$ | $T$ | $K$ | $\Pi$ | $M$ | $B_{200}$ | $B_{300}$ | CA |
|---|---|---|---|---|---|---|---|---|---|---|
| $d_4^{20}$ | 1 | 0,6 | 0,26 | 0,07 | 0,82 | −0,22 | 0,86 | −0,45 | −0,5 | −0,82 |
| $\gamma_{20}$ | 0,6 | 1 | 0,16 | 0,11 | 0,3 | −0,14 | 0,83 | −0,29 | −0,31 | 0,35 |
| $S$ | 0,26 | 0,16 | 1 | −0,13 | 0,84 | 0,17 | 0,27 | −0,4 | −0,2 | 0,82 |
| $T$ | 0,07 | 0,11 | −0,13 | 1 | 0,3 | 0,6 | 0,13 | −0,02 | −0,04 | −0,1 |
| $K$ | 0,82 | 0,3 | 0,84 | 0,3 | 1 | 0,2 | 0,59 | 0,25 | −0,15 | 0,91 |
| $\Pi$ | −0,22 | −0,14 | 0,17 | 0,6 | 0,2 | 1 | 0,14 | 0,07 | −0,11 | 0,04 |
| $M$ | 0,86 | 0,83 | 0,27 | 0,13 | 0,59 | 0,14 | 1 | −0,55 | −0,45 | 0,26 |
| $B_{200}$ | −0,45 | −0,29 | −0,4 | −0,02 | 0,25 | 0,07 | −0,55 | 1 | 0,37 | 0,076 |
| $B_{300}$ | −0,5 | −0,31 | −0,2 | −0,04 | −0,15 | −0,11 | −0,45 | 0,37 | 1 | 0,142 |
| CA | −0,82 | 0,35 | 0,82 | −0,1 | 0,91 | 0,04 | 0,26 | 0,076 | 0,142 | 1 |
normal law [3]. Checking the data by Pearson’s criterion [4] gave satisfactory results.
The influence of various characteristics on the content of paraffin and resin-asphaltene substances was evaluated by the magnitude and sign of the correlation coefficients [5] presented in the table.
The reliability of the correlation coefficients was checked using the function
\[ z=\frac{1}{2}\ln\frac{1+r}{1-r}, \tag{1} \]
proposed by Fisher [4]. It is known that the pour point of crude oil is determined mainly by the paraffin content [6, 7], and the coefficient of pair correlation indicates this relationship. The pair-correlation coefficients for the other factors are small, and their influence may be neglected.
The equation relating the pour point and the paraffin content was found in the form
\[ \Pi = 4.95 + 0.69 \cdot T, \tag{2} \]
The coefficients of equation (2) were calculated by correlation analysis on the “MIR-1” computer [5] from 130 data points. A close relationship exists between the content of resin-asphaltene substances and specific gravity, total sulfur content, and coking value. Satisfying the condition of independence of the arguments, taking into account the magnitude of the pair-correlation coefficient, we obtain the equation in the form
\[ \mathrm{SA} = 1.16 + 3 \cdot K \tag{3} \]
The coefficients of equation (3) were calculated from 112 data points. Equations (2) and (3) adequately describe the initial data and can be used to predict the content of paraffin and resin-asphaltene substances in crude oils.
REFERENCES
-
New Crude Oils of the Eastern Regions of the USSR. Reference book edited by S. N. Pavlova and Z. V. Driatskaya. Khimiya, Moscow, 1967.
-
Crude Oils of the Eastern Regions of the USSR. Reference book edited by S. N. Pavlova and Z. V. Driatskaya. Khimiya, Moscow, 1962.
-
S. A. Aivazyan. Statistical Studies of Dependencies. Metallurgiya, Moscow, 1968.
-
N. A. Plokhinsky. Biometrics. Moscow, Moscow State University Press, 1970.
-
V. A. Silich, M. M. Shepelina. Collection “Cybernetics and the Higher School,” 4, 164, 1971.
-
B. M. Rybak. Analysis of Petroleum and Petroleum Products. Gostoptekhizdat, Moscow, 1962.
-
S. S. Nametkin. Petroleum Chemistry. Publishing House of the USSR Academy of Sciences, Moscow, 1955.
-
C. R. Rao. Linear Statistical Methods and Their Applications. Nauka, Moscow, 1968.