See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305384668

Simulation and optimization of an industrial gas condensate stabilization unit to

modify LPG and NGL Production with minimizing CO2 emission to the
environment

Article  in  Chinese Journal of Chemical Engineering · July 2016

DOI: 10.1016/j.cjche.2016.07.006

CITATIONS READS

6 1,092

3 authors, including:

Mahmoud Bahmani
Shiraz University
17 PUBLICATIONS   203 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

I'm working on treatment of salty wastewater using mixed methods based advanced oxidation process+ nano and biotechnology View project

Synthesis of novel photocatalysts based on metal organic framework and their use in wastewater treatment of ammonia and urea units in a flat plate photo reactor View
project

All content following this page was uploaded by Mahmoud Bahmani on 17 November 2017.

The user has requested enhancement of the downloaded file.

  
 
Simulation and optimization of an industrial gas condensate stabilization unit
to modify LPG and NGL Production with minimizing CO 2 emission to the
environment

M. Bahmani, J. Shariati, A. Nemati Rouzbahani, S. Maghsoodloo Babakhani

PII: S1004-9541(16)30091-X
DOI: doi: 10.1016/j.cjche.2016.07.006
Reference: CJCHE 623

To appear in:

Received date: 12 February 2016

Revised date: 4 June 2016
Accepted date: 6 July 2016

Please cite this article as: M. Bahmani, J. Shariati, A. Nemati Rouzbahani, S. Maghsood-
loo Babakhani, Simulation and optimization of an industrial gas condensate stabilization
unit to modify LPG and NGL Production with minimizing CO2 emission to the envi-
ronment, (2016), doi: 10.1016/j.cjche.2016.07.006

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Simulation and optimization of an industrial gas condensate stabilization unit

to modify LPG and NGL Production with minimizing CO2 emission to the

PT
environment

RI
M. Bahmanib, J. Shariatia,b*, A. Nemati Rouzbahanib, S. Maghsoodloo Babakhanib

SC
a
Department of Chemical Engineering, Darab Branch, Islamic Azad University, Darab, Iran
b
Young Researchers and Elite Club, Darab Branch, Islamic Azad University, Darab, Iran

Abstract
NU
MA
In the present study, a great effort was made to improve the performance of an industrial LPG and NGL

production unit in one of the major gas refinery located at Pars special economic zone in Iran. To
D
TE

demonstrate and obtain the optimal condition, the unit was simulated by using a steady-state flowsheet

simulator, i.e. Aspen Plus, under different operational conditions. According to the simulation results,
P
CE

the unit was not operational under its optimal conditions due to some defects in the cooling system at top

stage of the debutanizer tower (DBT) during hot and humid seasons. Additionally, the vapor pressure of
AC

produced LPG and accordingly the amount of its flaring were decreased by reducing the temperature of

debutanizer tower at top stages. In the optimization section, the DBT condenser and reboiler heat duty,

temperature, and pressure were regulated as adjustable parameters. The simulation results demonstrated

that by applying the optimum suggestion in the hot months, the reflux stream temperature was reached

about 55˚C which caused an efficient increment in LPG production (about 4%) with adjusting the

propane component in LPG, based on the standard range as the plant criteria. Moreover, after applying

modifications, about 750 tons of LPG product were saved from flaring during five hot months of the

year, which resulted in 360,000$ extra annual income for the company. Finally, from environmental
 ACCEPTED MANUSCRIPT

point of view, this optimization caused to reduce 81 tons of CO2 emission to the environment. Therefore,

the current investigation must be introduced as a friendly environmentally process.

PT
Keywords: LPG, NGL, Simulation, Optimization, CO2 emission

RI
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

1. Introduction

1.1. Natural Gas products

Natural gas (NG) is one of the world’s favorite, promising, and cheap fuels with a wide variety of

PT
applications. However, its transportation in the gas state encounters various difficulties. Therefore, in

RI
order to smooth the difficulties raised by its gaseous nature, the process of converting NG into

intermediate liquid streams has been recommended and applied for years [1]. Commonly, two main

SC
products namely liquefied petroleum gas (LPG) and natural gas liquids (NGL) are sourced from NG

NU
stabilization units of all industrial gas refineries. As a matter of fact, LPG is a super-pressurized gas

stored in a liquid form in tanks which is known as a valuable type of gas and remarkable demand in the
MA
economical world. This colorless and odorless compound is as twice as heavy as air and as half as heavy

as water due to its constituents. LPG is inherently a mixture of two flammable nontoxic gases namely,
D
TE

propane (C3H8) and butane (C4H10) that each of them has a variable portion in different seasons. In

winter, it is mainly composed of propane; while, the heavier component (i.e. butane) occupies most of
P

its volume during summer. LPG is produced by refining petroleum or "wet" natural gas, and it is almost
CE

entirely derived from fuel sources, being manufactured during the refining of petroleum (crude oil), or
AC

extracted from petroleum or natural gas streams as they emerge from the ground [2-5]. Moreover, NGLs

are commonly determined as heavier hydrocarbons of natural gas which tend to be in liquid state at

atmosphere condition. They are mainly composed of methane, ethane, propane, butane, pentane, hexane,

and heptane. In a typical natural gas refining plant, various components of NGLs are separated one by

one from the natural gas stream by applying a series of fractionation columns namely demethanizer,

deethanizer, depropanizer, debutanizer, etc.[6].

 ACCEPTED MANUSCRIPT

1. 2. Literature review

Chemical process industries face a lot of operational problems in monitoring and controlling the

processes. Up to now, the majority of investigations on problems of gas processing plant focused their

PT
attention on different kinds of controlling systems, methods of preventing product loss with special

RI
attention to the environmental aspects of the issue and energy consumption of the plant. For instant, in

1998, a comprehensive study covering a diversity of controlling systems was done by Ansari et al. [7].

SC
Nonlinear and PID control systems were applied to control the percentage of i-C5 in the top product

NU
stream of the plant. Analyzing the results clearly represented that by implementing the nonlinear

controlling systems, a better performance can be achieved with respect to PID controlling strategies [8].
MA
Environmental pollution especially, CO2 emission from all natural gas refinery flare stacks is one of the

most serious problems that should be considered. In this regards, in 2014, Davoudi et al. tried to perform
D
TE

a great research on the assessment of flare networks in South Pars Gas processing plant. Actually by

modeling the data and calculating the maximum flaring load in five gas refineries, it was revealed that
P

different flare sizing load caused different values of CO2 emission which can be attributed in different
CE

fire zone arrangements [9]. In the same year, NematiRoozbahani et al. investigated the simulation,
AC

optimization, and sensitivity analysis of a natural gas plant. Their study was focused on significant

parameters playing a central role in the dehydration unit and accordingly, the best operating conditions

were determined [10]. Afterwards, in 2015, Amidpour et al. presented the sensitivity analysis, economic

optimization, and configuration design of mixed refrigerant cycles by NLP techniques. They showed the

one-stage cascade of mixed refrigerant refrigeration cycle (MRRC) as the best option to replace pure

ethylene cycle in the olefin plant of Tabriz Petrochemical Complex [11].

Moreover, a steady state simulation was performed by Al-Sobhi and Elkamel for LNG (Liquefied

Natural Gas), GTL (Gas to Liquid), and methanol to determine mass and energy balances, operating
 ACCEPTED MANUSCRIPT

conditions, and equipment specification. They also investigated the other relevant preprocess units of

these utilization processes. Finally, they proposed an optimization model for their simulation [12].

To the best of authors’ knowledge, simulation, optimization, and economic analysis of different

PT
industrial gas processing such as waste gasification, product loss, and greenhouse gas emission through

RI
flaring played an important role in their studies [13-16].

In the current study, simulation and modification of an industrial gas condensate stabilization unit for

SC
more LPG & NGL production were investigated under significant operational conditions. Some

NU
operational problems of the stabilization unit such as gas flaring and products wasting were verified.

Moreover, CO2 emission to atmosphere during more LPG production process as well as economic
MA
analysis of the process optimization and modification were studied completely.
D
TE

2. Process description

The understudied gas refining plant mainly consists of three distinct units including “a pre-separation
P

and dehydration unit”, “a stabilization unit”, and “an emergency stabilization package”.
CE

Natural gas, which is collected from 8 different Gas-Wells, enters to the pre-separation unit at nearly 70
AC

bar operating pressure. Firstly, it runs into six parallel two-phase separators. Then, after the separation

process, the outlet gas is collected into one main stream and sent to the dehydration unit for further water

content removal and treatment. As a result, its dew point can be decreased to less than -20 °C which has

a vital role in process design for produced dry gas. Also, this will prevent forming any kind of hydrate

while transporting via pipeline in long distances. The cooling process in the hydration unit is performed

using propane as the coolant and Diethylene Glycol (DEG) as wet dehydration agent [10]. During this

process, a mixture of water, heavy hydrocarbons, and DEG known as wet hydrocarbons, (i.e. stream 8)

shown in Fig. 1, is produced. This stream will be treated further in another three-phase separator (S-404)
 ACCEPTED MANUSCRIPT

to avoid the waste of precious liquid compounds. The outlet liquids from the six two-phase separators

consist of a significant amount of valuable hydrocarbons. Hence they are collected as the natural gas

liquids, (i.e. stream 1) shown in Fig. 1, and sent to a three-phase separator known as S-205 where the

PT
separation of water content and heavy hydrocarbons from the feedstock proceeds (Fig. 1). In this step,

RI
the heavy hydrocarbons are separated from water and sent to the stabilization unit along with the outlet

hydrocarbons stream from S-404 to produce LPG and NGL. The information about the material

SC
specifications of main streams and operational parameters in Fig. 1 can be seen in Table 1.

NU
Fig. 1

Table 1
MA
Hydrocarbon fractionation was considered as an essential part of all gas processing plants generally

performed by applying a series of distillation columns [17, 18]. Fig. 2 illustrates a schematic diagram of
D
TE

the current stabilization unit (Unit 700). As it is shown, an expansion drum (S-701), two sieve tray

distillation towers, i.e. a de-ethanizer (T-701) and a de-butanizer (T-702), and their relevant condensers
P

and reboilers along with a two-phase separator (S-702) for controlling the pressure of de-butanizer tower
CE

are the key components constructing the main structure of the whole unit.
AC

The S-701 is inherently a three-phase separator (S-701) which is fed by the output of the upstream unit

(pre-separation and dehydration units). In this drum, methane is partially separated from the entering

feed stream and directed from the top of the vessel to the fuel gas system. Moreover, the heavier

hydrocarbons are recovered from the middle section of S-701 to be further treated in the subsequent

fractionation columns.

From there, the middle phase stream runs into the de-ethanizer where ethane and lighter components are

distilled at 18 bar operating pressure by application of heat. The overhead product then flows into the

fuel gas system to be applied in the power generation section of the plant. The C3+ rich product which
 ACCEPTED MANUSCRIPT

leaves the bottom of the column is directed to the 20th tray of the de-butanizer tower with the ultimate

goal of producing Liquefied Petroleum Gas (LPG) as its overhead product. The lighter stream leaving

the top of the tower is totally condensed by passing through the coils of the air coolers (E-701 A/B) that

PT
are incorporated in downstream of the still column. After that, the condensate which continues flowing

RI
into the receiver vessel (S-702) plays the role of a reflux drum. LPG then flows out of the bottom of the

vessel and continues its path to the storage tanks by applying a pumping system.

SC
If, for any unfavorable reason, a failure happens in the transition of LPG to the storage tanks, the

NU
gaseous mixture will by-pass the air coolers and it will be sent directly to the receiver vessel. In the next

step, it released from the vessel to the low pressure flare stack to be burnt out in order to protect the
MA
entire system from facing over pressure problem. This failure can be stemmed mainly from an increase

in tower pressure or temperature which will be discussed later in the following sections. The bottom
D
TE

outlet liquid from T-702 which is one of the main product of this unit, is stored at Natural Gas Liquid

(NGL) storage tanks after it has been cooled down by the downstream air cooler (E-702).
P

A brief information about the operating conditions and chemical composition of the main streams and
CE

the design criteria for some of the main equipment in the stabilization unit are tabulated in Table 2 and
AC

Table 3, respectively.

Fig. 2

Table 2

Table 3

As it has been demonstrated in Fig. 3, emergency stabilization package consists of a small gaseous

furnace (H-1) and some two-phase separators which will be used only, when the severe malfunction

which happens at the upstream units causing the whole stabilization unit to be shut down. The main

purpose of the unit is to prevent sending both the valuable heavy hydrocarbons to the flare by separating
 ACCEPTED MANUSCRIPT

the lighter hydrocarbons and the rest to the NGL tanks. After entering the gas liquids into this package,

due to heating the gaseous furnace, lighter hydrocarbons are separated from liquid phase and consumed

at the fuel gas system or sent to the flare according to the plant conditions. Consequently, heavier

PT
hydrocarbons exit this package and next they will be cooled down and stored as the NGL product. This

RI
procedure is demonstrated in Fig. 3. Additionally, the information of its related streams is provided in

Table 4.

SC
Fig. 3

NU
Table 4
MA
2. 1. Describing the problem

In the studied natural gas refinery plant, during hot seasons of the year, the ambient temperature may
D

exceed over 50˚C at the warmest period of the day. Therefore, in the stabilization unit the efficiency of
TE

heat exchangers which use ambient air as the coolant fluid, i.e. E-701 and E-702, will reduce severely.
P

Although, this issue does not incur any major problem for the outlet fluid from E-702, since it heads
CE

directly into the NGL storage tank (Fig. 2). The main problem occurs when outlet stream temperature
AC

from E-701 increases rapidly because of the mentioned issue, resulting in an abnormal temperature

profile at top stages of T-702. It is needless to say that any temperature increment in recycled LPG will

increase the vapor pressure of the outlet stream from top stage of T-702. The normal operating pressure

of the feed entering T-702 is set on the value of 10.5 bar. However, in designing of any engineering

equipment, the operational condition is under the influence of many factors. Obviously, temperature as

an effective parameter has a leading role in the pressure adjustment of many gas associating equipment.

Therefore, in the present research, the malfunction of the air-cooler condenser (E-701) and its impact on

the temperature and consequently, the pressure of the tower have been chosen as a case study. It is worth
 ACCEPTED MANUSCRIPT

mentioning that the air cooler heat duty cannot be easily changed and there is a limit on the maximum

effective duty that can be exchanged with the upper stage vaporous mixture depending on the ambient

temperature and the driving power of the air cooler electro motor. For that reason, a secondary air cooler

PT
is needed for consuming extra duty and cooling the reflux stream to the proper temperature. Due to the

RI
fact that the temperature increment will pose a rise in the volumetric flow rate inside the tower and

consequently, a pressure climb at upper stages of the column. In order to overcome the problem, an

SC
adjustable control valve, i.e. PCV-714 shown in Fig. 2, with the set pressure of 11.5 bar has been

NU
embedded just in downstream of the receiver drum. Whenever the vessel pressure reaches 11.5 bar, the

control valve will be opened quickly and as a result, a part of the produced LPG will be sent out to the
MA
flare. Moreover, a large volume of greenhouse gases (mainly CO2) will be produced and emitted to the

atmosphere during the flaring.

D
TE

With considering the maximum allowable working pressure (MAWP) of both the receiver vessel and the

de-butanizer tower (Table 3), in a case that their operating pressure exceed the criterion, the emergency
P

inter-lock control system of the stabilization unit will be activated by preventing the feed stream to this
CE

unit, shutting down the reboilers (H-701/702), and consequently shutting down the whole unit. Hence
AC

the feed will be bypassed directly to the emergency stabilization unit (Unit 710) for some simple

separation and then injected to the NGL storage tanks. Therefore, the LPG production of the plant will

be stopped completely at this condition.

The main purpose of the current study is to perform a reliable simulation of the stabilization unit using a

steady state flow sheet simulator, Aspen Plus, in order to be able to use a sensitivity analysis tool on the

most important parameters in this unit, such as LPG and NGL production rate and produced LPG

composition. So the unknown relationship between these parameters and other operational variables like

condenser heat duty of the debutanizer tower and etc can be discovered. It is worth mentioning that,
 ACCEPTED MANUSCRIPT

seeking a way to reduce the flaring of LPG and decreasing the emission of CO2 to the environment in

consequence of high pressure at warm seasons, with keeping the LPG composition at its standard range

should be counted as another objective of this article.

PT
RI
3. Simulation and sensitivity analysis

The whole gas condensate stabilization unit was simulated using a steady-state flow sheet simulator. In

SC
order to obtain the reliable and accurate results from the simulator, a proper property package must be

NU
selected [19]. By taking a closer look at the reported operating data of Table 2, due to low molar flow

rate of water entering to the stabilization unit, a Peng-Robinson equation of state (EOS) was selected to
MA
apply in the current simulation. Also, another factor which motivated us to use this EOS was, its

unerring accuracy for systems containing hydrocarbons at moderate operating pressures and
D
TE

temperatures. Nevertheless, in order to make sure that the selected EOS has the best consistency with

this case study, a comparison between other proper EOSs is necessary to be performed. Since the case
P

study was hydrocarbon system and by considering the acceptable range of its pressure and temperature,
CE

three equations of Peng-Robinson (PR), Lee-Kesler-Plocker (LK-Plock), and Sova-Redlich-Kwong

AC

(SRK) were recommended by simulator [20]. In order to choose the best equation, the simulation was

performed for each EOS separately. A brief result of calculated values by each equation for outlet LPG

stream from the debutanizer tower along with the field design data has been tabulated in Table 5. In

order to compare the results, an error index namely absolute average relative deviation percentage

(AARD) was utilized to investigate the accuracy of each model, individually [21]. Eq. (1) shows the

relevant relationship for calculating AARD:

1  N op op 
AARD(%)  
N i
 yi  ycal
i / y i   100

(1)
 ACCEPTED MANUSCRIPT

Where yop is the actual data, ycal is the calculated result by models, and N is the number of data points.

Table 6 indicates the calculated AARD for each model. According to the table, the lowest offset from

actual data can be achieved by PR EOS with the value of 5% error; while, LK-Plock calculation has the

PT
weakest correspondence with almost 38% offset. It is needless to say that, due to its low generated error,

RI
PR property package was selected for the final simulation and sensitivity analysis.

Table 5

SC
Table 6

NU
Since the main purpose of the this study is to minimize the LPG flaring from the unit and optimize the

performance of its debutanizer tower (T-702), after running the simulation process, a sensitivity analysis
MA
should be performed to inspect the effective thermo-dynamic parameters on the debutanizer tower

performance. In fact sensitivity analysis is a try and error calculating technique used to determine how
D
TE

sensitive an output of a mathematical model or simulation is to any change in the value of a particular

input variable while keeping other independent parameters constant [22]. Sensitivity analysis allows
P

modelers to determine what level of uncertainty is acceptable for a parameter to make the simulation
CE

reasonably valid. To this end, the main independent and dependent variables must be specified
AC

beforehand. Tower pressure, as it has been mentioned before, is the most critical parameter in

controlling the operation which even a slight change on it can pose severe problems and failures on

performance of the whole unit. Tower pressure can be controlled by adjusting the feed pressure through

its relevant control valve and can vary from 9.5 to 12.5 bar (maximum design pressure of the tower). It

should be mentioned that because of the design criteria at current situation, the maximum operational

pressure of the tower has been set to 11.5 bar. Other important dependent and independent variables are

presented in Table 7. In this study, the sensitivity analysis was performed using the Aspen Plus

sensitivity and optimization tools based on the approved simulation of the stabilization unit.
 ACCEPTED MANUSCRIPT

Table 7

4. Results and discussion

PT
4. 1. Condenser temperature and duty

RI
As previously mentioned, the debutanizer tower is designed in a way to operate with maximum

operating pressure of 11.5 bar. Accordingly, the excess pressure imposed on the column must be

SC
modified by flaring a part of LPG that is not an economy-environmental friendly approach. In order to

NU
solve this problem, the engineering decisions must be adopted to keep the tower pressure below 11.5

bar. This will be feasible only, if the temperature profile of the tower remains sufficiently low. Fig. 4 (a,
MA
b) shows the temperature profile of debutanizer tower at all stages of the tower and top stages,

respectively.
D

Fig. 4(a, b)
TE

Throughout the entire article, the simulation was performed by considering a constant reboiler thermal
P

loading of 3315 kW. This constant heat duty can be easily applied using a fuel gas burner around the
CE

reboiler to produce a constant heat duty by adjusting the flow rate of the required fuel gas and ambient
AC

air to be burnt. As it can be observed from Fig. 4, the initial trays of tower have lower temperatures at

lower pressures. This can be explained by decreasing the tower pressure which leads to enhance the

vapor pressure of lighter components at top trays as well as reduction of their condensation temperature.

Consequently, the tower air cooler as a total condenser should cool vapors to lower temperature in order

to produce a saturated liquid stream.

In the current study, attempts were made to determine effective parameters during tower pressure

alternation and finally, pressure reduction of tower by considering operational limitations. Due to the

limitations exerted on the system, the condenser may not be able to react fast enough to cool the vapor

and set the temperature for total condensation of the overhead stream. Thereby, the malfunction of the
 ACCEPTED MANUSCRIPT

tube bundled air cooler causes the mentioned equipment to operate as a partial condenser. Thereafter,

the outlet stream from the air cooler continues its path to the downstream receiver vessel of S-702. In

this vessel the vapor portion of the feed is first separated and after that when the pressure exceeds 11.5

PT
bar, it will be sent to the flare through the PCV-114 as a waste product.

RI
Fig.5

To overcome the problem, the heat exchanger outlet fluid which is a saturated liquid should be cooled

SC
by increasing the heat duty of the condenser through adding a secondary air cooler in conjunction with

NU
E-701, and then the pressure should be reduced. The degrees of temperature reduction must be

proportional to pressure reduction. As it can be estimated from Fig. 4 (b), 1 bar pressure reduction is
MA
attained by 4.5˚C temperature reduction. Fig. 5 indicates the amounts of condenser energy consumption

in different ranges of temperature and pressure at hot seasons of the year (considering the average
D

temperature of 45˚C for the ambient during day). As it would be expected, at higher pressures, the
TE

amount of consumed energy for fluid cooling is more than the lower ones. This phenomenon can be
P

attributed to the reduction of heavier components of the vaporous stream resulting in greater
CE

condensation temperature of the mixture. At three mentioned pressures (9.5, 10.5, and 11.5 bar) in Fig.
AC

4 (a, b), the condensation temperature values for standard LPG stream (stage 1 temperature) according

to the composition reported in Table 2 are 52.1, 56.0, and 60.5, respectively. As it is obvious from Fig.

5, at any pressure value, the energy consumption of the condenser varies with steeper slope at lower

temperatures; while, the slope becomes gentler at greater condensation temperatures. As one can see

from Fig. 5, for 1 bar pressure drop in tower for example 11.5 to 10.5, it is necessary to cool the reflux

stream by about 4.7 ˚C which requires an increase in heat exchanger power from 3275 kW to 3516 kW.

By implementing these alternations and attaining the equilibrium conditions, tower pressure can be

reduced about 1 bar without flaring any LPG product. Without any doubt, the thermal loading of air
 ACCEPTED MANUSCRIPT

cooler should be decreased again (in this stage to 3283kW) by pressure reduction, in order to prevent the

excessive and unnecessary energy consumption. Moreover, according to Table 3, the maximum

available heat duty for air cooler condense at operational pressure of 10.5 bar, is 2950 kW. However, by

PT
looking closely at Fig. 5 it can be seen that the required heat duty for condensing the whole inlet stream

to the condenser, at saturated fluid temperature of 56˚C, is 3265 kW. It means that in hot seasons, an

RI
extra 315 kW heat duty should be provided in such conditions. Therefore the unit can operate normally.

SC
NU
4. 2. LPG and NGL flow rate and compositions

Fig. 6 (a, b) displays the main products flow rate of stabilization unit (unit 700) at different pressures for
MA
various condenser outlet temperature. As it can be observed, the amount of produced LPG decreases due

to further reduction of reflux temperature. At constant pressure, further thermal loading which leads to
D
TE

convert its fluid to sub-cool state, can change the tower temperature profile. As a result, heavy and some

of light components tend to remain in the liquid phase and leave the tower from its bottom. Clearly, at
P

lower pressures heavier hydrocarbon components will tend to move to the gas phase. Consequently, the
CE

production rate for LPG as the main product of the stabilization unit will increase; while, at the same
AC

time the flow rate of produced NGL will drop. It should be noticed that decrement of tower pressure and

temperature must alter in such a way that mass fraction of i-C5 in produced LPG stream stay lower than

2%. This is an important criteria for designing the plant to produce LPG in standard range and with

considering the need of customers. Moreover, the standard range for propane mass fraction in LPG

product is determined between 25% and 37% of outlet stream total mass fraction. Fig. 7 (a, b) indicates

variations of propane and iso-pentane versus condenser temperature at three distinct pressures. This

figure shows that by reducing pressure while the temperature is held in constant value at top stage of

T-702, a higher portion of heavier compound such as i-C5 will move to the vapor phase which leads to
 ACCEPTED MANUSCRIPT

increase the heavy hydrocarbon composition in LPG final product. Based on the figure, since the air

cooler cannot set its inlet stream temperature to fluid dew point, some portions of the produced two

phase flow will be sent to flare and waste. Furthermore, this causes the mass fraction of iso-pentane (i-

PT
C5) in the produced LPG to be out of mentioned standard range. For instance, at pressure and

temperature about 10.5 bar and 57 ˚C of reflux stream, the i-C5 mass fraction is more than 7% which is

RI
against design criteria. However, if the air cooler cools down the reflux stream to 55 ˚C, i-C5 mass

SC
fraction will drop to 2% which is acceptable. Simultaneously, the C3 mass fraction in those temperatures

NU
are around 25% and 37% respectively which both of them are in the standard range. Accordingly, the

approach of fluid sub-cooling in condenser for 1 to 2 ˚C is an effective method to attain LPG at its
MA
standard range. However, a sudden pressure drop happens in the T-702 feed inlet pressure.

Fig. 6 (a, b)
D
TE

Fig. 7 (a, b)
P

4. 3. Optimization section
CE

In order to maintain the operational pressure (10.5 bar) at top stage of DBT in hot seasons, an extra heat
AC

exchanger for further cooling of the reflux stream is suggested (preferably an air cooler due to its low

power consumption). Therefore, an optimized temperature for the reflux stream should be chosen so the

stabilization unit can operate at its optimum conditions. Fig. 8 indicates the composition of C3 versus i-

C5 for the LPG stream at different temperatures. By taking into consideration the previously mentioned

criteria for the LPG compositions (C3 between 25% and 37%, and i-C5 lower than 2%), the temperature

range at the operational pressure may not exceed more than 55 ˚C and surely will not be lower than 51

˚C. To find the optimum temperature in this range, the production rate of LPG versus NGL is plotted in

Fig. 9. According to this figure, by considering the reverse relationship in NGL and LPG production, as
 ACCEPTED MANUSCRIPT

the temperature goes down the production of LPG decreases mildly; while, this decrement slightly

increases the production of NGL. For instance, 1˚C drop in the temperature, (from 56 ˚C to 55 ˚C) will

reduce the overall LPG production by 2.5%; whereas, NGL molar flow will increase less than 0.3%. It is

PT
needless to say that the higher production of LPG as one of the main products and obtaining the standard

RI
range for LPG component as the plant criteria, are the most important factors to optimize the

stabilization unit performance. Therefore, by considering the acquired results from Fig. 8, the optimum

SC
temperature will be 55 ˚C at the selected operational pressure. The overall production of LPG at this

NU
temperature is about 78 ton/day plus an average value of 5 ton/day LPG flaring which has been

prevented by lowering the upstream pressure. Consequently, by saving 5 ton/day LPG from flaring (750
MA
ton/five hot months in overall) and loosing 2 ton/day due to temperature reduction in condenser outlet

temperature, a net increase of 450 tons in LPG production rate was obtained during five hot months of
D
TE

the year (assuming average of 150 hot days per every year). In addition to the financial loss imposed on

the system, LPG flaring has an irreparable effect on environment by emitting about 0.54 ton CO2 to the
P

atmosphere per 1 ton of LPG [23]. Hence, 81 tons of CO2 are sent to the atmosphere during five hot
CE

months of the year. By comparing the obtained results from the simulation with actual field data
AC

presented in Table 2, a slight rise around 4% in LPG final production is detectable.

Fig. 8

Fig. 9

4. 4. LPG flaring rate and economical estimation

Fig. 10 reveals the amount of LPG flaring rate from stabilization unit at abnormal conditions. As

mentioned earlier, due to the reduction in air cooler efficiency, the gradual enhancement in condenser

outlet temperature of the debutanizer tower can lead to increase the pressure parameter at top stages of
 ACCEPTED MANUSCRIPT

the tower. At operating pressure of 11.5 bar, the control systems of the column are activated to control

the tower pressure by LPG flaring. In this industrial stabilization unit, when the tower pressure increases

to 12.5 bar, then total failing occurs in the cooling system, the whole unit becomes out of service, and

PT
LPG production stops. In this situation, NGL continues its path to the emergency gas condensate

RI
stabilization unit which is clear in Fig. 10. It should be considered that both the flaring and LPG flow

rates reported in Fig. 10 are momentarily. In order to perform an accurate statistics for LPG flaring

SC
specially, during five hot months of the year in south of Iran, flaring data values from PCV-114 were

NU
measured from May to October months by considering 3 hours daily effective LPG flaring. The data are

plotted in Fig. 11. This figure shows a perfect consistency between the simulation results and the actual
MA
data of plant. Consequently, the economical estimation can be performed. It must be noticed that the

operating values drew in Fig. 11 were recorded daily and at the maximum pressure of the column.
D
TE

According to the Gold Persian Fob criterion, the accepted price for a ton of LPG is 800$ in 2014. From

an economical point of view and based on the mentioned criterion, about 2400$ equivalent LPG product
P

is sent to flare daily. According to the LPG price and the amount of CO2 emission due to LPG flaring,
CE

360,000$ of LPG product burns in flare system and 81 tons of CO2 are sent to the atmosphere during
AC

five hot months of the year. As a result, the improvement of cooling system of DBT and reducing the

fluid outlet temperature of the condenser as a definitive and profitable method should be considered

carefully. Therefore, in the present study, after applying modifications, about 750 tons of LPG product

were saved from flaring during five hot months of the year, which resulted in 360,000$ extra annual

income for the company. Also, from environmental point of view, this optimization caused to reduce 81

ton per years of CO2 emission (the most effective Greenhouse gas) to the atmosphere that is benefit for

the society.

Fig. 10
 ACCEPTED MANUSCRIPT

Fig. 11

5. Conclusion

PT
The main purpose of the present study was to perform a reliable simulation of a stabilization unit in

RI
order to carry out a sensitivity analysis on the most important parameters of the unit including LPG and

SC
NGL production rate and the composition of the produced LPG. Hence, the existing relationship

between these parameters and other operational variable like condenser heat duty of the debutanizer

NU
tower and etc, were discovered. According to the simulation results, the unit was not operational under

its optimum conditions during the summer. By reduction of DBT temperature, its pressure was
MA
decreased and the amount of LPG flaring through the top of tower decreased respectively. Besides, the

simulation results ascertained that by applying the new scheme in hot seasons, the reflux stream
D
TE

temperature approached the value of approximately 55˚C which caused an efficient increment of about

4% in LPG production. Moreover, the produced NGL slightly decreased and the consumed amount of
P
CE

fuel gas in the tower reboiler decreased as well. In addition, after some modifications, an increase of

about 450 tons in LPG production rate was obtained during five hot months of the year. Hence the
AC

company seemed to be successful in earning 360,000$ extra annual income.

Interestingly, in present investigation, by applying an accurate modification and optimization, the CO2

emission to the atmosphere decreased almost 81 tons. Therefore, from an environmental perspective the

current study revealed some positive points which unquestionably, must be considered by decision

makers particularly, those who care about ecosystem and healthy life.
 ACCEPTED MANUSCRIPT

References

[1] Campbell JM., 2014. Gas conditioning and processing, the basic principles. 9th ed. Norman.
[2] Mokhatab S, Poe AW, Speight JG., 2006. Hand book of natural gas transmission and processing.

PT
Gulf professional publishing.
[3] Himmelblau DM, Edgar TF, Ladson LS., 2001. Optimization of chemical processes. 2nd ed.

RI
McGraw- Hill Inc.

SC
[4] Salehi K, Jokar SM, Shariati J, Bahmani M, Sedghamiz MA, Rahimpour MR., 2014. Enhancement
of CO conversion in a novel slurry bubble column reactor for methanol synthesis. J. Nat. Gas Sci. Eng.

NU
21, 170-183.
[5] Kister HZ., 2002. What Causes Malfunctions in Refinery Towers. Asia: Hydrocarbon.
MA
[6] Canete JF, Gonzalez-Perez S, Saz-Orosco P., 2008. Artificial neural network identification and
control of a lab-scale distillation column using LABVIEW, World Academy of Science, Engineering
and Technology. 2, 11-28.
D

[7] Ansari, R.M., Tade, M.O., 1998. Nonlinear model based multivariable control of a debutanizer,
TE

J. Process. Cont. 8, 279-286.

[8] Fernandez de Canete J, Garcia-Cerezo A, Garcia-Moral I, Del Saz P, Ocho E., 2013. Object-oriented
P

approach applied to ANFIS modeling and control of a distillation column, Expert Systems with
CE

Applications. 40, 5648-5660.

[9] Davoudi M, Aleghafouri A, Safadoost A., 2014. Flaring networks assessment in South Pars Gas
AC

processing plant. J. Nat. Gas Sci. Eng. 21, 221-229.

[10] Nemati Rouzbahani A, Bahmani M, Shariati J, Tohidian T, Rahimpour MR., 2014. Simulation,
optimization, and sensitivity analysis of a natural gas dehydration unit. J. Nat. Gas Sci. Eng. 21, 159-
169.
[11] Amidpour, M., Hamedi, M.H., Mafi, M., Ghorbani, B., Shirmohammadi, R., Salimi, M., 2015.
Sensitivity analysis, economic optimization, and configuration design of mixed refrigerant cycles by
NLP techniques, J. Nat. Gas Sci. Eng. 24, 144-155.
[12] Al-Sobhi SA, Elkamel A., 2015. Simulation and optimization of natural gas processing and
production network consisting of LNG, GTL, and methanol facilities. J. Nat. Gas Sci. Eng. 23, 500-508.
 ACCEPTED MANUSCRIPT

[13] Davoudi M, Rahimpour MR, Jokar SM, Nikbakht F, Abbasfard H., 2013. The major sources of gas
flaring and air contamination in the natural gas processing plants: A case study. J. Nat. Gas Sci. Eng. 13,
7-19.
[14] Abdulrahman, R.K., Sebastine, I.M., 2013. Natural gas sweetening process simulation and

PT
optimization: A case study of Khurmala field in Iraqi Kurdistan region, J. Nat. Gas Sci. Eng. 14, 116-
120.

RI
[15] Bassyouni M, ul Hasan SW, Abdel-Aziz MH, Abdel-hamid SM-S, Naveed SH, Nasir Ani F., 2014.

SC
Date palm waste gasification in down draft gasifier and simulation using ASPEN HYSYS. Energ.
Convers. Manage. 88, 693-699.

NU
[16] Shariq Khan M, Donald Chaniago Y, Getu M, Lee M., 2014. Energy saving opportunities in
integrated NGL/LNG schemes exploiting: Thermal-coupling common-utilities and process knowledge.
MA
Chem. Eng. Process. Intens. 82, 54-64.
[17] Kister HZ., 1990. Distillation operation. New York; McGraw- Hill Inc.
D

[18] Wauquier JP., 2000. Petroleum refining. Separation processes, Editions Technip.
[19] Ludwig EE., 1995. Applied process design for chemical and petrochemical plants. 3rd ed. Gulf
TE

Publishing.
P

[20] Xu P, Xu S, Yin H., 2006. Application of self-organizing competitive neural network in fault
CE

diagnosis of suck rod pumping system. J. Petrol. Sci. Eng. 58, 43-48.
[21] Babakhani, S.M., Bahmani, M., Shariati, J., Badr, K., Balouchi, Y., 2015. Comparing the capability
of artificial neural network (ANN) and CSMHYD program for predicting of hydrate formation pressure
AC

in binary mixtures, J. Petrol. Sci. Eng. 136, 78-87.

[22] Darwish NA, Hilal N., 2008. Sensitivity analysis and faults diagnosis using artificial neural
networks in natural gas TEG-dehydration plants. Chem. Eng. J. 137, 189-197.
[23] Shu Y, S N Lam N, Reams M., 2010. A new method for estimating carbon dioxide emissions from
transportation at fine spatial scales. J. Environ. Res. Lett. 5, 9 pp.
 ACCEPTED MANUSCRIPT

List of figures:

Fig 1. Schematic view of pre-separation and dehydration equipment and condensate feed gathering line
[S stands for Separator].

PT
Fig 2. Schematic view of stabilization unit process (Unit 700) [S stands for Separator, E is for
Exchanger, P is for Pump and H is for Heater].

RI
Fig 3. Schematic view of emergency stabilization unit process (Unit 710) [V stands for Vessel, E is for
Exchanger, P is for Pump and H is for Heater].

SC
Fig 4 (a, b). Temperature profile versus the number of the debutanizer tower stages in different
pressures.

NU
Fig 5.The amounts of condenser energy consumption in different ranges of temperature in hot seasons.
Fig 6 (a, b). a) LPG and b) NGL flow rates in different pressures for various condenser outlet
MA
temperature.
Fig 7(a, b). Variations of a) propane and b) iso-pentane versus condenser temperature in different
pressures.
D

Fig 8. Mass composition of C3 and i-C5 versus temperature on LPG stream at operational pressure (10.5
TE

bar).
P

Fig 9. Mass flow rate of LPG and NGL versus temperature at operational pressure (10.5 bar).
CE

Fig 10. The amount of LPG flaring rate from stabilization unit at abnormal conditions.
Fig 11. LPG flaring from S-702 in different ranges of pressure.
AC
 ACCEPTED MANUSCRIPT

To Fuel Gas System

Natural Gas Wet

PT
Liquids Hydrocarbons
1 8
3 5

RI
SC
S-205 S-404
(Pre-Separation Unit) (Dehydration Unit)

NU
To Glycol
Regeneration Unit 7
Water - To
Drain
MA
6
4 Condensate
System
2 Feed
D
TE

Fig. 1
P
CE
AC
 ACCEPTED MANUSCRIPT

To Fuel Gas

To Fuel Gas
8 PCV-

System

System
2 5 E-701 A/B 714 LPG To LP
Flare
S-702 Header

PT
7 P-702
3
1

T-701

T-702

RI
Condensate 9
Feed
LPG to

SC
Storage
S-701
H-701 H-702

NU
4
E-702
6 10
To HC Drain

P-701 P-703 Stabilized

System

MA
NGL to
Storage

NNF (Normally No Flow)

D

Fig. 2
P TE
CE
AC
 ACCEPTED MANUSCRIPT

1 To Flare Header
Condensate
Feed
V-01

PT
RI
2
V-02 V-03 Stabilized NGL
to Storage

SC
3
4 5

Fuel Gas E-1

NU
H-1 P-1

MA
Air

Fig. 3
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

230
P = 9.5 bar
210
10.5 bar

Temperature Profile (ºC)

190 11.5 bar
170

PT
150
130

RI
110
90

SC
70
50
0 10 20 30

NU
Stages
MA
95
90
Temperature Profile (ºC)

85
D

80
75
TE

70
65
P = 9.5 bar
P

60
10.5 bar
55
CE

50 11.5 bar

45
AC

0 1 2 3 4 5 6
Stages

Fig. 4 (a, b)
 ACCEPTED MANUSCRIPT

4000
P = 9.5 bar
3900
10.5 bar
3800

PT
11.5 bar
3700
Qc (kW)

3600

RI
3500
3400

SC
3300
3200
3100

NU
3000
49 50 51 52 53
MA 54 55 56 57 58 59 60 61

Condenser Temperature (ºC)

Fig. 5
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

120
P = 9.5 bar
110
10.5 bar
100 11.5 bar
90

LPG (ton/day)

PT
80

70

RI
60

SC
50

40
50 51 52 53 54 55 56 57 58 59 60

NU
Condenser Temperature (ºC)
MA
1100
P = 9.5 bar
1090 10.5 bar
11.5 bar
1080
D

1070
NGL (ton/day)

TE

1060
1050
1040
P

1030
CE

1020
1010
AC

49 50 51 52 53 54 55 56 57 58 59 60 61
Condenser Temperature (ºC)

Fig. 6 (a, b)
 ACCEPTED MANUSCRIPT

0.50
P = 9.5 bar

0.45 10.5 bar

11.5 bar
0.40

C3 (mass)

PT
0.35

RI
0.30

0.25

SC
0.20
49 51 53 55 57 59 61

NU
Condenser Temperature (ºC)
MA
0.20 P = 9.5 bar
10.5 bar
11.5 bar
D

0.15
i-C5 (mass)

TE

0.10
P

0.05
CE

0.00
AC

49 51 53 55 57 59 61
Condenser Temperature (ºC)

Fig. 7 (a, b)
 ACCEPTED MANUSCRIPT

0.160 0.45

0.140 0.40

PT
0.120 0.35
i-C5 mass fraction

C3 mass fraction
0.30
0.100

RI
0.25
0.080
0.20

SC
0.060
0.15
i-C5 (mass)
0.040 0.10
C3 (mass)

NU
0.020 0.05
0.000 0.00
50 52 54 56 58 60
MA
Temperature (˚C)
D

Fig. 8
P TE
CE
AC
 ACCEPTED MANUSCRIPT

120 1085
1080
100
1075

NGL mass flow (ton/day)

LPG mass flow ( ton/day)
80 1070

PT
1065
60
1060

RI
40 1055
LPG

SC
1050
20 NGL
1045
0 1040

NU
50 51 52 53 54 55 56 57 58 59 60
Temperature (˚C)
MA
Fig. 9
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

60
Flare (kg/min)
55
Mass flow rate (kg/min) LPG
50

PT
45
40

RI
35
30

SC
25
20
15

NU
10
11.4 11.6 11.8 12 12.2 12.4 12.6
Top stages pressure (bar)
MA
Fig. 10
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

10
9 Flaring

Average Mass flow rate of Flaring

8 Actual Field Data
7
6

PT
(ton/day)
5
4

RI
3
2

SC
1
0
11.5 11.7 11.9 12.1 12.3 12.5

NU
Top stages' pressure (bar)
MA
Fig. 11
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

List of Tables:
Table 1. Components flow rate and thermodynamic conditions for main streams in the pre-separation

and dehydration unit (specified in Fig. 1).

PT
Table 2. Components flow rate and thermodynamic conditions for main streams in the stabilization unit

(specified in Fig. 2).

RI
Table 3. Detailed information for designing of major equipment in the stabilization unit.

SC
Table 4. Components flow rate and thermodynamic conditions for main streams in the emergency

stabilization unit (specified in Fig. 3)

NU
Table 5. Calculated values by each equation for outlet LPG stream from the debutanizer tower versus the field
MA
design data.

Table 6. Calculated AARD for each model.

D

Table 7. The important dependent and independent variables for stabilization unit.
P TE
CE
AC
 ACCEPTED MANUSCRIPT

Table 1

Stream Numbers
7 8
1 2 3 4 5 6
Stream Components

PT
C1 kmol/hr 89 0 59.9 29.1 85.3 0.1 20.2 105.5

RI
C2 kmol/hr 12.6 0 4.2 8.4 9.8 0 10.5 20.3

SC
C3 kmol/hr 11.7 0 1.8 9.9 5.3 0 17.6 22.9

NU
IC4 kmol/hr 5 0 0.4 4.6 1.2 0 9.3 10.6

NC4 kmol/hr 9.4 0 MA0.6 8.8 1.9 0 18.9 20.8

IC5 kmol/hr 7.9 0 0.2 7.6 0.7 0 15.4 16.1

NC5 kmol/hr 7.5 0 0.2 7.3 0.5 0 14.9 15.4

D

C6+ kmol/hr 229.2 0 0.3 229.1 0.5 0 76 76.5

TE

N2 kmol/hr 2.5 0 2.1 0.5 1.9 0 0.2 2.1

P

CO2 kmol/hr 1.2 0 0.5 0.6 0.5 0 0.3 0.8

CE

H2O kmol/hr 309 308.1 0.4 0.4 0.1 80.9 0.1 81.1
AC

DEG kmol/hr 0 0 0 0 0 49 0 49

Mass Flow kg/hr 42308 5551 1375 35382 2282 6664 13008 21954

Temperature °C 51.6 51.6 51.6 51.6 30.4 30.4 30.4 30.4

Pressure bar 27 27 27 27 26.99 26.99 26.99 26.99

Vapor Fraction 0.103 1 1 0 1 0 0 0.256

 ACCEPTED MANUSCRIPT

Table 2

Stream Numbers

PT
1 2 3 4 5 6 7 8 9 10
Stream Components

C1 kmol/hr 49.3 12.5 36.8 0 0 0 0 0 0 0

RI
C2 kmol/hr 18.9 1.4 17.5 0 0 0 0 0 0 0

SC
C3 kmol/hr 27.6 0.7 26.8 22.8 254.7 0 231.9 254.7 22.8 0

IC4 kmol/hr 13.9 0.2 13.8 12.9 143.7 0.2 130.8 143.7 12.9 0.1

NU
NC4 kmol/hr 27.8 0.3 27.5 26.3 288.6 1.2 262.7 288.6 25.8 0.4

IC5 kmol/hr 23.1 0.1 23 22.5

MA 14.1 64.4 12.6 14.1 1.3 21.3

NC5 kmol/hr 22.2 0.1 22.2 21.8 0.6 65.9 0.6 0.6 0.1 21.7

C6+ kmol/hr 305.2 0.1 305 304.4 0 922.1 0 0 0 304.4

N2 kmol/hr 0.6 0.3 0.3 0 0 0 0 0 0 0

D

CO2 kmol/hr 0.9 0.1 0.7 0 0 0 0 0 0 0

TE

H2O kmol/hr 0.5 0.1 0.4 0 0 0 0 0 0 0

Mass Flow kg/hr 48391 337 48054 46482 37417.3 130700 34066.4 37417.3 3351 13131
P

Temperature °C 45.1 45.1 44.8 199.2 68.1 219.2 55.4 54.5 58.1 60.5
CE

Pressure bar 21.5 16.5 18 12 10.5 11 10.5 10.3 16.7 2.5

Vapor Fraction 0.032 1 0.019 0.198 1 0 0 0 0 0

AC
 ACCEPTED MANUSCRIPT

Table 3

Equipment Design Details

PT
Type: sieve tray distillation column
De-ethanizer tower (T-701) Number of trays: 23

RI
Design temperature: 311 ˚C
Design pressure: 20.7 bar
Maximum allowable working pressure (MAWP): 22.7 bar

SC
Type: sieve tray distillation column
De-butanizer tower (T-702) Number of trays: 29
Design temperature: 304 ˚C

NU
Design pressure: 11.5 bar
Maximum allowable working pressure (MAWP): 12.5 bar

Type: air cooler condenser

MA Design duty: 2950 kW
E-701 A/B Design temperature: 102 ˚C
Design pressure: 11.0 bar
Maximum allowable working pressure (MAWP): 12.5 bar
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

Table 4

Stream Numbers

PT
1 2 3 4 5
Stream Components

RI
C1 kmol/hr 49.3 37.4 37.4 0 0

SC
C2 kmol/hr 18.9 17.6 17.6 0.2 0.2

NU
C3 kmol/hr 27.6 26.9 26.9 2.2 2.2

IC4 kmol/hr 14 MA 13.8 13.8 3.1 3.1

NC4 kmol/hr 27.8 27.5 27.5 7.9 7.9

IC5 kmol/hr 23.1 23 23 11.6 11.6

D

NC5 kmol/hr 22.2 22.2 22.2 12.4 12.4

TE

C6+ kmol/hr 305.2 305.1 305.1 284.8 284.8

P

N2 kmol/hr 0.6 0.3 0.3 0 0

CE

CO2 kmol/hr 0.9 0.8 0.8 0 0

AC

H2O kmol/hr 0.5 0.4 0.4 0 0

Mass Flow kg/hr 48398 48079 48079 40576 40576

Temperature °C 45.2 45.2 95 81 55

Pressure bar 21.8 21.8 9.12 2.5 2

Vapor Fraction 0.031 0 0.149 0 0

 ACCEPTED MANUSCRIPT

Table 5

Natural gas compositions and process parameters Field Data PR LK SRK

PT
C1 kmol/hr 0 0 0 0

RI
C2 kmol/hr 0 0 0 0

SC
C3 kmol/hr 22.8 21.3 20.2 19.7

NU
IC4 kmol/hr 12.9 12.1 13.3 10.3

NC4 kmol/hr 25.8 26.2 24.1 18.4

MA
IC5 kmol/hr 1.3 1.5 5.2 6.9
D

NC5 kmol/hr 0.1 0.5 3.3 2.8

TE

Mass Flow kg/hr 3351 3310 3678 3142

P
CE

Temperature °C 58.1 57.39 58.23 56.74

Pressure bar 16.7 16.7 16.7 16.7

AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
Table 6

NU
Selected EOS Percent of Variation from operational data
MA
Peng-Robinson 5.00

Lee-Kesler-Plocker 37.66
D

Sova-Redlich-Kwong 25.76
P TE
CE

Table 7
AC

Dependent Variables Independent Variables

Mass flow rate of LPG Debutanizer tower condenser and reboiler heat duty

Mass flow rate of NGL Debutanizer tower condenser temperature

Composition of LPG Reflux ratio

Debutanizer tower pressure

 ACCEPTED MANUSCRIPT

Highlights:

PT
 An industrial stabilization unit was studied and simulated with high accuracy.

RI
 An important operational problem in gas refineries was explained and a particularly way was

SC
suggested.

 The optimum operational conditions and economical analysis were performed.

NU
 A friendly environmentally process as a result of research was also obtained.
MA
D
P TE
CE
AC

View publication stats