Calculation of Crude Oil Processes Using Simplified Model Mixture

Abstract

This paper presents the modeling of the existing crude oil separation process in a system consisting of two rectification columns with side drafts operating at higher pressure. The composition of the crude oil was approximated by a model mixture of hydrocarbons. The installation calculations have been performed for two different model compositions containing 32 and 10 different hydrocarbons. The whole technological process was based on the assumption that the feed stream, containing a model crude oil, was introduced to the first column, and then the other expected products (different petroleum fractions), characterized by their respective boiling points, were collected (side drafts) from the appropriate trays of the distillation columns. The obtained calculation results for both of the model crude oils were compared with the results obtained in the existing petroleum process and discussed from the point of view of their practical applications. The detailed data concerning size, composition, and process parameters for all the streams of the investigated installations, as well as the necessary energy expenditure for each of the columns, have been determined. Moreover, some recommendations are presented for the modeling and optimization of industrial distillation processes of very complex, multi-component systems using simpler model compositions.

1. Introduction

Chemical engineering describes industrial processes involving multi-component mixtures. At the same time, the obtained product should be characterized by high purity (in the case of a single chemical compound) or a precisely defined composition (in the case of a mixture of chemical compounds). Unfortunately, by their nature, many of these mixtures tend to be homogeneous and thus difficult to separate. However, differences in the properties of the constituents allow for the designing of separation processes [1,2]. The separation and purification processes most often used for this purpose, such as distillation, extraction, crystallization, etc., are based on the difference in the composition of the phases that are in equilibrium with each other. To facilitate the separation, so-called “cascade processes” have been developed. The most popular among them is distillation.

In the case of distillation, the most fundamental element of a separation cascade process is the single stage, where the two streams entering the stage are in the vapor and liquid phases. In the model situation, they are mixed on the stage until they approach equilibrium and are mechanically separated. So, the steams leaving the stage are in equilibrium. At a steady state, the total mass and energy entering the stage is equal to that leaving.

In industrial processes, a single equilibrium stage is seldom sufficient to provide the separation needed. Therefore, it is necessary to link several stages together to create a “multistage” separation process called distillation or rectification column.

The modeling of such a multistage separation process requires the solution of a series of equations referred to as the M-E-S-H equations [3]. The M-E-S-H equations are Material balances, Equilibrium relations, Summations of mole fractions, and H (enthalpy) energy balances.

During the modeling of the process, the material balance must be closed around each stage as well as for the entire process. This ensures that the mass of material on any given stage, as well as of the whole column, remains constant.

Equilibrium relations consist of procedures that enable the calculation of the composition of phases in equilibrium at each stage at the temperature and pressure therein. The equilibrium relations, describing the compositions of two phases in contact in a process model, are usually determined by thermodynamic estimation techniques such as equations of state or equations based on the so-called γ–φ approach. Summations of mole fractions have to be performed at every point in the process model and must be equal to 1.0000 in any phase.

H (enthalpy) energy balances at steady-state must close at each stage of the model. This requirement is necessary for the steady-state energy changes at any stage to be zero.

Distillation calculations can be grouped into three categories: design, modeling, and optimization. Distillation design algorithms generate the design specifications of a distillation column based on information about the feed and the product streams. In contrast to design algorithms, distillation modeling calculates the composition profile of a specified column based on its configuration. Optimization is aimed at finding the best solution for the user in terms of the parameters set by them.

Since 1983, when Sorel [4] and Hausbrand [5] published the mathematical model of distillation and its material and energy balances, the mathematical basis of the calculations has not changed much. However, significant progress has been achieved in the field of computational technology. Initially, simple calculation and graphical methods gave way to modern, very complicated numerical techniques.

In the 1970s, large computer systems were commonly used for such calculations. Examples of distillation programs developed for this type of computer have been published in many books [6,7,8,9].

The microcomputer revolution (especially the remarkable progress in the field of computers compatible with the IBM PC standard) made computer programs a very popular work tool. New requirements have arisen for the programmers who create them. Currently, it is not enough to develop a precise and reliable calculation method; it is also necessary to ensure an easy and understandable form of using the program and entering data. That is why specialized computer companies were established, providing very complicated but very good and relatively easy-to-use programs describing individual apparatuses used in the chemical industry, individual separation processes, as well as entire technologies [10,11,12,13]—Chemcad, Aspen Plus, Aspen HYSYS, Unisim i PRO II.

A common industrial process of distilling highly complex multi-component mixtures is the rectification of crude oil, which is a mixture consisting of hundreds of hydrocarbons and also contains water, nitrogen, sulfur compounds, and some metal complexes [14]. The rectification column is the most important unit in the oil refining industry for separating crude oil into simpler and useful mixtures. A correct description of the rectification process of multi-component mixtures is difficult (due to various (two-, three- and multi-body) intermolecular interactions) and time-consuming. Therefore, the possibility of replacing such a system with a much smaller number of “model” components reflecting the physicochemical character of the tested multi-component mixture is very desirable. It is also the most expensive and energy-intensive unit in the refinery, consuming energy equivalent to 1–2% of the processed crude oil [15]. Therefore, there is a need to modernize and optimize crude oil distillation units in order to continuously improve the economic results of refineries, which is also associated with a significant increase in research interest in this field.

The optimization of the crude oil distillation process is widely discussed in the scientific literature [16,17,18,19,20,21,22]. A very diverse approach is used to address this complex problem. For example, already in the year 2000, a mixed-integer nonlinear programming model was developed and used to determine the optimal feed location and operating conditions of the crude oil distillation unit by Seo et al. [16]. Then, in 2002, Basak et al. [23] developed a nonlinear, stationary crude oil distillation unit to maximize net profit. In 2015, exergy analysis and mathematical programming were combined for the energy optimization of multistage crude oil distillation units by Gu et al. [17].

However, crude oil distillation requires simultaneously addressing multiple criteria; therefore, Inamdar et al. [17] applied elitist nondominated sorting genetic algorithms to use the multi-objective optimization, and Huang et al. [24] used a multi-objective optimization model to study the trade-off between economic benefit, energy consumption, and carbon dioxide (CO₂) emissions. Some process simulation packages (Aspen HYSYS, Aspen Plus, Unisim i PRO II) capable of designing an operational model for a given process are discussed in [25]. However, More et al. (2010) [26] suggest that process simulation using Aspen Plus is not a valid approach for optimizing continuous and binary variables (e.g., feed location and side stream location) in a crude oil distillation unit. Meanwhile, Ibrahim et al. [27] applied Aspen HYSYS and a simulation-optimization approach to design a single-stage crude oil distillation system and heat recovery network. The complexity of the problem of simulation and optimization of crude oil distillation columns leads to the search for methods to facilitate the solution of such problems. One of the solutions is surrogate modeling based on statistical regression of input variables and output responses [28,29,30].

It is necessary to mention that all the popular software packages [11,12,13] enable accurate modeling of the most common apparatuses, units, and multi-stage processes, as well as entire technological lines that are used in the chemical industry. Modeling is based on the most commonly used and best methods of thermodynamic description, the most effective numerical methods, and the largest data banks. One of the greatest advantages of these packages is the easy (WINDOWS type) possibility of constructing a process (model manager), all components of which are then calculated automatically, using data and appropriate methods of their description contained in the data banks of these programs. These packages also provide the possibility of using other data and other methods than those existing in data banks.

The program user decides which method and what data should be used. A package of programs prepared in this way facilitates the designer’s work, but, of course, it is not able to influence his decisions regarding the conditions and principles of the modeled process. Users of this type of software package must, therefore, now have much more knowledge about design, thermodynamics, quality of thermodynamic data, methods of their description, etc., in order to take full advantage of all its possibilities.

It should also be remembered that economic effects are a very important aspect when designing technological processes. The economic effects of distillation processes are dependent on many circumstances. Exact economic parameters (cost of energy and installations, cost of operating, etc.) are different in various countries and various periods of time. That is why, in this paper, we will concentrate only on the technological aspects of these processes.

In the case of modeling the crude oil processing process (obtaining appropriate fractions), it is crucial to reproduce the temperature range and the corresponding compositions of individual fractions, i.e., recreate the distillation curve of the tested process.

Hence, from the point of view of the modeling of the crude oil processing process, it is most important to reproduce the distillation curve using the smallest possible number of components of the model mixture.

Reproducing the distillation curve of the tested multi-component mixture subjected to the rectification process using a significantly reduced number of leading model components not only facilitates calculations but also shortens the simulation time of the tested process, which is particularly important in the case of complex technological processes.

That is why the presented method of modeling the crude oil processing process can be treated as a model both for the processing of crude oil from sources other than Azerbaijan as well as for other chemical processes in which rectification columns are used.

The purpose of this paper is to check which factor, the number of components taken into account or the correct representation of the distillation curve, has the most important influence on the simulated crude oil processing. All the calculations have been performed using the Chemcad software package in version 8.1.2.18002 for Windows 10.

2. Modelling

The calculations were carried out for a typical crude oil processing process. The technological line consisted of two rectification columns working under increased pressure. Before starting the calculations, the person using the computer program must pay close attention to the following specifications:

• A poorly specified feed will result in instabilities in the distillation model. A feed that is very hot will greatly increase the vapor flows on trays above the feed point. On the other hand, a cool feed will increase liquid flows on lower trays. In extreme cases, these conditions can make convergence impossible.
• The poor selection of components for a feed can also cause problems. If components with large differences in volatilities are present (in high concentrations), the distillation model may not give realistic results. The thermodynamic models have difficulty predicting equilibrium conditions under these conditions.

The specification of the column generally comprises the following:

• Fixing the parameters of the reboiler and the condenser (vapor pressures and liquid distillate, efficiencies, reflux ratio, liquid subcooling, bottom’s product rate);
• An increase/decrease in the number of trays;
• Specification of the stream flow and its composition for a given tray;
• Specification of the parameters of each of the trays (pressure, vapor product flow rate, liquid product flow rate, interheater duty, Murphree vapor efficiency).

After setting the parameters of the distillation process, one can start the computation and then see the following results:

• For the whole column: tray number, temperature, liquid and vapor flow rate, liquid and vapor enthalpy, reboiler and condenser duty (that is, the value of heat absorbed or desorbed in a period of time);
• For each of the trays: temperature, pressure, liquid and vapor composition, flow rate and enthalpy of internal liquid and vapor, optional feed stream, optional side-stream of liquid and vapor (that is, the flow rate of liquid and vapor products derived from a given tray).

The precision of calculations by the use of the program depends strongly on the quality of thermodynamic data comprised in the adequate data bank. The data-accompanying program should always be treated as an example, and the responsibility for its preparation belongs to the user (it is impossible to generate a general data set for any problem).

The precision of binary parameters of the selected γ–φ model plays a very important role in achieving the precision of computation of the program. The model parameter data file should always be created by the user on his responsibility. The NRTL equation [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29] has been selected for the description of the liquid–vapor equilibrium in all binary subsystems formed, specified by us, to describe the composition of crude oil. This equation for the Gibbs excess energy (G^E) of a two-component system takes the following form:

$${\displaystyle \frac{G^E}{RT}}=x_i{x}_j({\displaystyle \frac{{\tau }_{ji}{G}_{ji}}{(x_i+x_j{G_{ji})}^2}}+{\displaystyle \frac{{\tau }_{ji}{G}_{ji}}{(x_j+x_i{G_{ij})}^2}}),$$

(1)

where

$$G_{ij}=-\alpha {\tau }_{ij},$$

(2)

$${\tau }_{ij}=(g_{ij}-g_{jj})/(RT),$$

(3)

${(g}_{ij}-g_{jj}),(g_{ji}-g_{ii})$—adjustable parameters.

A general diagram of the entire crude oil processing installation is shown in Figure 1.

The installation consists of two distillation columns operating under increased pressure, with the operating pressure of the first column approximately twice as high as the operating pressure of the second column. The first column is fed with crude oil, in the form of a mixture of liquid and vapor, heated in a heat exchanger to the temperature determined by the process conditions (increased pressure), and two streams are collected from it: gasoline and naphtha. Gasoline is collected at the top of the column, and naphtha is removed by side draft by means of a stripper. The stream leaving the bottom of the first column is heated to the appropriate (process conditions) temperature in the heat exchanger and, in the form of a mixture of liquid and vapor, feeds the second column, from which four streams are collected: naphtha, kerosene, diesel oil, and topped crude. The heavy naphtha stream is received at the top of the column, and the topped crude stream is at the bottom of the column. Kerosene and diesel oil streams are extracted from this column as side drafts using strippers (as in the case of the first column). Both the columns and the strippers are heated by compressed steam (they do not have a boiler) supplied to them from the bottom (for the clarity of the drawings, steam, and water flows are not shown in Figure 1 and other figures in the text). Moreover, each of the columns has a fully condensing dephlegmator with decantation (on the top), one circulation pump with a heat exchanger, and one extra heat exchanger.

All three strippers used for side drafts collection are double-tray distillation columns that are steam-heated and are fed with a stream of liquid from a tray with a temperature characteristic for a given fraction. From the second stripper tray (bottom of the stripper), a liquid stream of a given fraction is collected, while the vapor stream from the first stripper tray (top of the stripper) is delivered to the main column to the shelf above the shelf from which the side draft is received. The strippers also play the role of compensating for the flow of liquid and vapor streams in the columns caused by the side removal of the relevant fractions. The first column has one stripper on the naphtha side draft, and the second column has two strippers on the kerosene and diesel side drafts.

In addition, in order to fully compensate for liquid and vapor flows on the tray in the middle of each column, from which the side draft is received, and, at the same time, to improve the purity of this stream, a pump with a heat exchanger was used. This pump takes the liquid from the tray from which the side draft is taken, and then it is cooled and delivered two trays above the tray from which it was taken. The cooled stream of the liquid supplied above the side draft receiving point additionally improves the fractionation in this part of the column and positively affects the purity of the stream of a given fraction received below. Such a pump with a heat exchanger has been installed on a tray from which naphtha is collected in the first column and diesel in the second column.

Since the proposed installation columns and strippers are heated with steam (no boilers), in practice, we are dealing with steam distillation. All the steam delivered to the columns and strippers is taken off at the top of the columns in fully-condensing dephlegmators with decantation, in which the process of separating the obtained appropriate fraction of crude oil (product) from water takes place.

Moreover, to ensure adequate heat flow in the columns, a side heat exchanger was installed in each of the columns on the feed tray.

Calculations of a complex installation defined in this way require many assumptions. The main ones concern defining the composition and stream of crude oil, determining the number of trays in both columns, the parameters (pressure and temperature) of dephlegmators operation, the selection of feed plates for both columns, the parameters of the heating steam used to heat both the columns and the strippers (on which mainly depend the flows in the columns), the determination of the trays fed by streams of liquid and steam coming out of the strippers, and the working conditions of the heat exchangers.

3. Results

We performed two types of calculations, in which the composition of crude oil was approximated by the following models: 32- and 10-component hydrocarbon mixtures. The model mixture compositions are given in

Table 1. The model mixture compositions [31].

No.	32 Component Mixture			10 Component Mixture
	Component Name	Mass Fraction	Boiling Point	Component Name	Mass Fraction	Boiling Point
1	Isobutane	0.004682	−11.7	Isobutane	0.004682	−11.72
2	Cyclopentane	0.008477	49.25
3	Benzene	0.009462	80.1
4	Cyclohexane	0.024641	80.72	Cyclohexane	0.057759	80.7
5	1,1-Dimethylcyclopentane	0.024641	87.85
6	2,2-Dimethylhexane	0.024641	106.89
7	Toluene	0.024641	110.63	Toluene	0.034103	110.63
8	n-Octane	0.024641	125.7	n-Octane	0.119611	125.68
9	Ethylcyclohexane	0.024641	131.8
10	m-Xylene	0.024641	139.12
11	Styrene	0.024641	145.16
12	n-Nonane	0.024641	150.82
13	n-Propylbenzene	0.024641	159.24
14	1-Ethyl-3-Methylbenzene	0.030520	161.33
15	1-Ethyl-4-Methylbenzene	0.020668	162.01	1-Ethyl-4-Methylbenzene	0.121138	162.01
16	1,2,4-Trimethylbenzene	0.020668	169.38
17	1-Decene	0.021047	170.6	1-Decene	0.021047	170.6
18	n-Decane	0.021047	174.15
19	Naphthalene	0.054674	217.99	Naphthalene	0.079315	217.99
20	n-Pentadecane	0.056662	270.75
21	n-Hexadecane	0.057657	286.9
22	n-Heptadecane	0.056662	302.15	n-Heptadecane	0.221346	302.15
23	n-Octadecane	0.030092	317.0
24	n-Nonadecane	0.020274	330.0
25	Eicosane	0.020274	343.0
26	n-Tetracosane	0.045817	391.0
27	n-Pentacosane	0.045817	401.9
28	n-Hexacosane	0.045818	412.2
29	n-Heptacosane	0.045818	422.1	n-Heptacosane	0.295181	422.1
30	n-Octacosane	0.045818	435.6
31	n-Nonacosane	0.045818	440.9
32	Triacontane	0.045818	449.7	Triacontane	0.045818	449.7

.

Based on the assumed composition of the model liquid, distillation curves (Figure 2) were developed at a pressure equal to 1.000 bar and a temperature of 25 °C.

Then, all devices included in the production line, i.e., dephlegmators, heaters, pumps, and rectification columns, were defined, and the conditions of the column’s operating parameters, flows, and expected product reception temperature were imposed on the tested system.

The basic assumptions were to assume constant parameters (temperature and pressure) of steam heating (supplied at the bottom) for both the columns and the three strippers and the fact that all water vapor supplied to the columns and strippers is collected in decanting dephlegmators placed at the top of columns 1 and 2.

Calculations of the crude oil distillation process require assumptions regarding the apparatus included in the process line. All the assumptions made (the operating parameters, some data accepted for simulation, and the assumed flows of selected streams) are given in

Table 2. Column 1—data accepted for simulations.

	Configuration	Process Parameters
Column-1	Number of trays (including a dephlegmator—tray no. 1)—13	Pressure at the top of the column 3.65 bar, pressure drop in the column 0.14 bar
	Feed tray no. 11	Feed stream (29.7% vapor, 70.3% liquid): total flow 121,171 kg/h (813 kmol/h), temperature before heater 20.0 °C, temperature after heater 204.4 °C, pressure 3999 bar
	Tray no. 10	Flow of liquid 2.65 m3/h
	Tray no. 13	Flow of liquid 11.3 m3/h
	Column bottom fed with steam	Steam: total flow 1362.32 kg/h (75.6 kmol/h), temperature 168.3 °C, pressure 7.9 bar
Dephlegmator	Fully condensing with decantation	Cooling to temperature 37.8 °C Pressure 3.3 bar
Stripper-1	Number of trays 2
	Powered (liquid) from tray no. 8; Returned vapors to tray no. 7	Flow of exhausted liquid 35.9 m3/h
	Stripper bottom fed with steam	Steam: total flow 408.69 kg/h (22.7 kmol/h), temperature 168.3 °C, pressure 7.9 bar
Pump with the Heat Exchanger-1	Liquid withdrawn from tray no. 8, Liquid returned to tray no. 6,	Flow 34.3 m3/h
	Heat duty (heat flux received)	−5275.3 MJ/h

and Figure 3 and Figure 4 for the first column and

Table 3. Column 2—data accepted for simulations.

	Configuration	Process Parameters
Column-2	Number of trays (including a dephlegmator—tray no. 1)–16	Pressure at the top of the column 1.59 bar, pressure drop in the column 0.14 bar
	Feed tray no. 14	Feed stream (34.8% vapor, 65.2% liquid): total flow 90,401.66 kg/h (421 kmol/h), temperature before heater 205.7 °C, temperature after heater 315.6 °C, pressure 1.59 bar
	Tray no. 13	Flow of liquid 48.6 m3/h
	Column bottom fed with steam	Steam: total flow 735.64 kg/h (40.8 kmol/h), temperature 168.3 °C, pressure 7.9 bar
Dephlegmator	Fully condensing with decantation	Cooling to temperature 37.8 °C Pressure 1.38 bar
Upper Stripper	Number of trays 2
	Powered (liquid) from tray no. 8 Returned vapors to tray no. 7	Flow of exhausted liquid 25.04 m3/h
	Stripper bottom fed with steam	Steam: total flow 272.46 kg/h (15.1 kmol/h), temperature 168.3 °C, pressure 7.9 bar
Lower Stripper	Number of trays 2
	Powered (liquid) from tray no. 12; Returned vapors to tray no.11	Flow of exhausted liquid 24.09 m3/h
	Stripper bottom fed with steam	Steam: total flow 326.95 kg/h (18.1 kmol/h), temperature 168.3 °C, pressure 7.9 bar
Pump with the Heat Exchanger-2	Liquid withdrawn from tray no. 12; Liquid returned to tray no. 10	Flow 31.1 m3/h
	Heat duty (heat flux received)	−7712.5 MJ/h

and Figure 3 and Figure 5 for the second column. Moreover, Figure 3 shows the process flowsheet generated in the Chemcad application, which was used in simulation calculations.

After making all the assumptions, crude oil distillation calculations were performed on the proposed installation. A detailed diagram of the proposed crude oil processing installation is shown in Figure 3.

The obtained results (including water) included the column distillation profiles, compositions, flows, and process parameters (temperature and pressure) of all obtained products—appropriate fractions of crude oil, defined by their boiling points, as well as the calculated amount of heat supplied in the exchangers installed in the production line are given in

Table 4. Column 1—distillation profile (data for stripper trays are listed as tray number 14 for 1 tray of the stripper and tray number 15 for 2 trays of the stripper).

Stg	Temp	Pres	Liquid	Vapor	Feeds	Product	Duties
	°C	bar	kg/h	kg/h	kg/h	kg/h	MJ/h
1	37.8	3.31	22,974.77			1963.75	−15,330
						1730.71 (1)
2	86.3	3.65	26,320.90	26,668.78
3	104.8	3.67	28,311.44	30,014.92
4	113.5	3.68	29,203.78	32,005.46
5	120.6	3.69	29,060.35	32,897.75
6	131.8	3.70	69,313.23	32,754.35			−5275
	67.5				27,584.07 (2)
7	140.2	3.72	70,300.00	45,423.29	4728.94 (3)
8	153.7	3.73	7690.67	41,681.16		27,584.07 (4)
						33,166.14 (5)
9	186.4	3.74	5490.96	39,821.95
10	201.0	3.75	2179.25	37,622.23
11	217.2	3.77	101,358.00	34,310.51	121,171.00 (6)		8709
12	212.0	3.78	96,027.38	12,318.19
13	205.7	3.75		6987.65	1362.32 (7)	90,401.66
Stripper-1
14	146.0	3.73	31,165.39		33,166.14 (5)	4728.94 (3)
15	136. 9	3.73		2728.13	408.69 (8)	28,845.83

⁽¹⁾—decant. (water), ⁽²⁾—return stream from the heat exchanger, ⁽³⁾—return stream from the stripper, ⁽⁴⁾—side draw to the heat exchanger, ⁽⁵⁾—side draw to the stripper, ⁽⁶⁾—feed stream, ⁽⁷⁾—steam, ⁽⁸⁾—steam (to the stripper).

,

Table 5. Column 1—distillation profile for 10 components (data for stripper trays are listed as tray number 14 for 1 tray of the stripper and tray number 15 for 2 trays of the stripper).

Stg	Temp	Pres	Liquid	Vapor	Feeds	Product	Duties
	°C	bar	kg/h	kg/h	kg/h	kg/h	MJ/h
1	37.8	3.31	20,863.00	2245.03		1736.67 (1)	−15,190
2	104.8	3.65	26,865.29	24,844.24
3	119.2	3.67	28,848.87	30,846.51
4	122.7	3.68	28,891.03	32,830.09
5	127.0	3.69	28,266.12	32,872.25
6	136.4	3.70	67,901.73	32,247.34	27,116.26 (2)		−5275
7	144.6	3.72	69,452.30	44,766.87	4804.15 (3)
8	156.5	3.73	8061.24	41,513.29		27,116.26 (4)
						32,733.00 (5)
9	186.6	3.74	5832.61	39,971.43
10	201.2	3.75	2231.77	37,742.80
11	218.8	3.77	101,364.61	34,141.96	121,000.00 (6)		9280
12	213.5	3.78	95,970.49	12,274.79
13	207.2	3.79	6880.67	90,451.76	1362.32 (7)
Stripper-1
14	148.5	3.73	30,690.70		32,733.00 (5)	4804.15 (3)
15	139.2	3.73		2761.72	408.69 (8)	28,337.55

⁽¹⁾—decant. (water), ⁽²⁾—return stream from the heat exchanger, ⁽³⁾—return stream from the stripper, ⁽⁴⁾—side draw to the heat exchanger, ⁽⁵⁾—side draw to the stripper, ⁽⁶⁾—feed stream, ⁽⁷⁾—steam, ⁽⁸⁾—steam (to the stripper).

,

Table 6. Column 2—distillation profile (data for stripper trays are listed for stripper 1 as tray number 17 for 1 tray of the stripper and tray number 18 for 2 trays of the stripper; for stripper 2, they are listed as tray number 19 for 1 tray of the stripper and tray number 20 for 2 trays of the stripper).

Stg	Temp	Pres	Liquid	Vapor	Feeds	Product	Duties
	°C	bar	kg/h	kg/h	kg/h	kg/h	M J/h
1	37.8	1.38	35,162.28			10,248.66	−29,230
						1345.15 (1)
2	153.5	1.59	58,822.68	46,755.74
3	164.2	1.60	61,566.38	70,416.13
4	169.0	1.61	62,408.21	73,159.84
5	172.1	1.62	62,402.53	74,001.67
6	175.3	1.63	60,882.52	73,996.00
7	180.9	1.64	54,422.75	72,476.00	4069.95 (2)
8	199.2	1.64	13,387.77	61,946.29		24,739.74 (3)
9	245.2	1.65	13,208.92	50,651.14
10	266.4	1.66	59,971.93	50,472.29	24,513.83 (4)		−7712
11	280.5	1.67	64,499.00	67,716.34	4722.86 (5)
12	292.3	1.68	6995.73	67,520.58		24,513.83 (6)
						24,013.42 (7)
13	325.4	1.69	4494.16	58,549.20
14	334.2	1.70	45,812.93	56,047.64	90,401.66 (8)		10,570
15	330.1	1.71	42,775.86	6964.74
16	323.7	1.72		3927.67	735.64 (9)	39,583.62
Upper Stripper
17	190.0	1.64	22,779.93		24,739.74 (3)	4069.95 (2)
18	180.5	1.64		2110.03	272.46 (11)	20,942.28 (10)
Lower Stripper
19	235.8	1.68	21,920.12		24,013.42 (7)	4722.86 (5)
20	277.1	1.68		2630.02	326.95 (13)	19,616.96 (12)

⁽¹⁾—decant. (water), ⁽²⁾—return stream from the upper stripper, ⁽³⁾—side draw to the upper stripper, ⁽⁴⁾—return stream from the heat exchanger, ⁽⁵⁾—return stream from the lower stripper, ⁽⁶⁾—side draw to the heat exchanger, ⁽⁷⁾—side draw to the lower stripper, ⁽⁸⁾—feed stream, ⁽⁹⁾—steam, ⁽¹⁰⁾—kerosine, ⁽¹¹⁾—steam, ⁽¹²⁾—diesel, ⁽¹³⁾—steam.

,

Table 7. Column 2—distillation profile for 10 components (data for stripper trays are listed for stripper 1 as tray number 17 for 1 tray of the stripper and tray number 18 for 2 trays of the stripper; for stripper 2, they are listed as tray number 19 for 1 tray of the stripper and tray number 20 for 2 trays of the stripper).

Stg	Temp	Pres	Liquid	Vapor	Feeds	Product	Duties
	°C	bar	kg/h	kg/h	kg/h	kg/h	M J/h
1	37.8	1.38	35,228.52			10,007.69	−29,200
						1344.91 (1)
2	153.6	1.59	58,325.00	46,580.76
3	166.5	1.60	61,388.17	69,677.28
4	171.6	1.61	62,317.86	72,740.41
5	174.5	1.62	61,825.65	73,670.11
6	177.9	1.63	59,331.65	73,177.90
7	185.4	1.64	52,836.81	70,683.89	3888.88 (2)
8	206.8	1.64	18,595.05	60,300.18		25,230.26 (3)
9	259.3	1.65	19,984.38	51,288.73
10	278.8	1.66	64,602.48	52,678.07	24,510.45 (4)		−7712
11	292.2	1.67	69,126.58	72,785.78	4950.83 (5)
12	301.9	1.63	6227.98	72,359.00		24,510.45 (6)
						24,210.37 (7)
13	337.6	1.69	4469.91	58,181.18
14	344.8	1.70	45,398.63	56,423.11	90,451.76 (8)		14,570
15	340.4	1.71	42,399.59	6900.08
16	333.6	1.72		3901.03	735.64 (9)	39,234.00
Upper Stripper
17	197.5	1.64	23,364.50		25,230.26 (3)	3888.88 (2)
18	187.8	1.64		2023.06	272.46 (11)	21,613.82 (10)
Lower Stripper
19	294.7	1.68	22,047.92		24,210.37 (7)	4950.83 (5)
20	235.3	1.68		2788.40	326.95 (13)	19,586.38 (12)

⁽¹⁾—decant. (water), ⁽²⁾—return stream from the upper stripper, ⁽³⁾—side draw to the upper stripper, ⁽⁴⁾—return stream from the heat exchanger, ⁽⁵⁾—return stream from the lower stripper, ⁽⁶⁾—side draw to the heat exchanger, ⁽⁷⁾—side draw to the lower stripper, ⁽⁸⁾—feed stream, ⁽⁹⁾—steam, ⁽¹⁰⁾—kerosine, ⁽¹¹⁾—steam, ⁽¹²⁾—diesel, ⁽¹³⁾—steam.

and

Table 8. Main heaters data.

Equip. No.		1	3
Name		Heater-1	Heater-2
Temperature out	°C	204.44	315.56
Heat absorbed	MJ/h	55,339.07	35,896.19
Fueal usage (SCF)		77,705.35	50,404.28

(a for the 32 components and b for 10 components model crude oil) and Figure 6 and Figure 7.

The obtained results of calculations carried out on a model crude oil composition in the crude oil processing installation proposed by us were compared to the results of the distillation of Azerbaijani crude oil and presented in Figure 8.

As can be easily seen (Figure 8), the quality of the results obtained is not dependent on the number of components in the feed liquid. Both in the case of a model liquid containing 32 components and 10 components, the obtained results are fully consistent with the actual results for the processing of Azerbaijani crude oil. The calculated temperatures of individual fractions are within the real temperature range of the fractions. There are two exceptions. First, in the case of a smaller number of components, the temperature of the gasoline receiving stream (Column 1) is slightly overestimated. The second exception, occurring both in the case of the 32- and 10-component model crude oils, is the temperature of the last fraction (topped oil), which, although not much different from the actual temperature, is slightly lower than it. This difference, caused by not taking into account the heaviest crude oil components in the calculations, is so small (Table 8) that it does not change the correctness of the calculations of the entire crude oil processing installation. Similarly to the temperatures of the received fractions, the calculated top and bottom temperatures of both columns, as well as the compositions of the received fractions, are very close to the real values (Table 4, Table 5, Table 6 and Table 7). Also, a point that should be particularly emphasized is that the energy balance of the entire installation is very reliable (Table 4, Table 5, Table 6 and Table 7).

4. Conclusions

The selection of the components of the model mixture is carried out in such a way as to best represent the distillation curve, and then the boiling points of the fractions calculated in the simulation of the column operation will correspond to the actual boiling points of these fractions in the real process. The composition of model mixtures can be freely selected as long as the distillation curve is correctly reproduced. Therefore, the discussion of the influence of individual components on the course of the simulation is not carried out.

Looking at the obtained results, one thing that seems to be crucial for a successful simulation of the crude oil processing process should be noted, namely, the distillation curves of the model crude oil (Figure 2). In the case of simulating the crude oil processing process, the number of model crude oil components is much less important than the correct representation of the distillation curve of the simulated crude oil, which is the key issue of accurate process modeling.

Therefore, when selecting model components (number of components and their mass fractions) representing the simulated real petrochemical process, it is necessary to remember that their mixture reflects the distillation curve of the process.

The ability to model and optimize the crude oil rectification process based on its reduced model composition is important because commercial software (Chemcad, Aspen Plus, etc.) need not have, in its database, all the real components of crude oil. However, it is certainly possible to select a group of components that adequately simulate the distillation curve.

Summing up, the results obtained for both model crude oils fully confirm the possibility of simulating the actual crude oil processing process using the Chemcad Software 8 based on the model composition of crude oil.

However, it should be remembered that, for the installation design to be correct, users of this type of software package must have extensive knowledge of design, thermodynamics, quality of thermodynamic data, and methods of their description. The compatibility of the program data on the phase equilibria in multi-component systems with data available in the literature and databases should always be verified. The obtained results allow for the formulation of a general conclusion regarding the modeling of chemical processes in which rectification columns are used. In these processes, we usually deal with multi-component systems in which physicochemical data are not always included in the database of commercial simulation programs.

In such situations, as our research shows, the multi-component real mixture can be replaced with a model system consisting of a much smaller number of components (the physicochemical data that are available in the databases of these programs), provided that they reflect the actual distillation curve. Such a curve can be determined experimentally for any real multi-component system.