Version of Record: https://www.sciencedirect.

com/science/article/pii/S0959652619315586
Manuscript_6656dbfb58d6523d83aee041b7ea6385

Exceeding Pinch limits by process configuration of an existing modern

crude oil distillation unit – A case study from refining industry

Omar S. Bayomiea,b,*, Omar Y. Abdelazizc,d, Mamdouh A. Gadallae,f,*

a
Department of Energy and Processes, MINES ParisTech, PSL Research University, 75006
Paris, France
b
Department of Chemistry, The American University in Cairo, New Cairo 11835, Egypt
c
Department of Chemical Engineering, Cairo University, Giza 12613, Egypt
d
Department of Chemical Engineering, Lund University, 22100 Lund, Sweden
e
Department of Chemical Engineering, Port Said University, 42523 Port Said, Egypt
f
Department of Chemical Engineering, The British University in Egypt, Cairo 11837, Egypt

*omar.bayomie@univ-psl.fr
*mamdouh.gadalla@bue.edu.eg

Abstract

Crude Distillation Unit (CDU) represents significant challenge for retrofitting and

energy optimisation as the most energy intensive consumer in a conventional crude oil

refinery. Pinch Technology and its based-methodologies are found primary keys for

decades to energy savings in refining industries for a range of common economic-based

and environmental objectives or applications. Typical benefits in energy savings are

reported within 20 to 40% of original designs. However, such savings are limited and

questioned when modern refiners are dealt with. The current paper addresses the

revamping of a modern refinery exhibiting an existing high energy efficiency (≈93%).

This implies the maximum potential energy savings would only be 7% at current

process conditions. The present research proposes an algorithm that tackles energy

recovery of modern refiners, enabling additional savings beyond the energy targets set

by the existing process. The algorithm starts by process simulation and validation

against real plant data, followed by a network optimisation, e.g. stream splitting, to

reach the energy targets set by Pinch Analysis. The energy targets are then moved to

1
© 2019 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
another lower level by performing potential process modifications to reduce the energy

consumption further. Results showed that the current modern refinery unit could reach

its energy targets by stream splitting modifications with hot energy savings of 2.69

MW. Process modifications resulted in additional energy savings of 31.3% beyond the

current level of the existing plant alongside less than a year of payback period for

estimated capital investment. An environmental assessment is performed, and

comparable reductions were obtained with respect to greenhouse gas, with reduction in

CO2 emissions by 45.1%. The proposed retrofit methodology is applicable to

minimising energy consumptions of refiners including modern units to achieve energy

levels beyond energy targets by new process modifications.

Keywords

Oil refinery; Process modelling; Energy savings; Emission reduction; Pinch Analysis;

Retrofit

1. Introduction

The oil downstream industry is facing big challenges to reduce its CO2 emissions while

balancing its economic positive margins. Toward making renewable energies more

affordable and energy efficiency techniques achieving big steps to CO2 emissions

reduction, the existing processes should reduce their energy consumptions. In reference

to current refinery configurations and oil demand, petroleum refineries are already

operating at their most efficient level ∼ 94% (Abdul-Manan et al., 2017), making the

energy saving and emission reduction activates a difficult task with such challenging

profitability margins. Economic and environmental redesigns of existing crude

distillation units (CDUs) appear to be of significant importance in order to achieve the

2
highest possible technical efficiencies simultaneously with a reasonable total investment

cost (Waheed et al., 2014).

One of the main considerable technologies in the last decade is heat and process

integration in the chemical and refining industries with significant advances in that

domain (Ulyev et al., 2018); many of techniques have been applied to reach more

efficient performance in the case of maximising energy recovery (Mehdizadeh-Fard and

Pourfayaz, 2019). Pinch Analysis is an effective methodology that proved its credibility

in many process industries for retrofitting and designing heat exchanger networks

(HENs) and utilities consumption (Čuček et al., 2019). Within the different tools

developed, the application of such techniques for existing oil refinery case studies has

gained importance (Kamel et al., 2017).

Various research efforts have been performed on the energy management of CDUs

since they are considered the major energy consuming unit operations in refinery

infrastructures. Thorough studies tackled the retrofit of existing CDU using sequential

approaches, i.e. column optimisation then HEN or vice versa (Joe and Rabiu, 2013), and

simultaneous approaches with Pinch Analysis targets (Gu et al., 2014). Retrofit designs

are regarded to be favourable than the grassroots design when it comes to petroleum

refineries (Enríquez et al., 2014), with process integration methods (Pejpichestakul and

Siemanond, 2013). Other solutions involving distillation design modifications and

different revamping procedures were also applied to enhance the crude fractionation

distillation yield and raise the energy efficiency in existing CDUs (Bagajewicz et al.,

2014). More research works were focusing on enhancing the economic viability for

operation by analysing progressive distillation over conventional design (Nalinakshan et

al., 2019), retrofit tracing grid diagram (Nemet et al., 2015), or minimising hot utility

3
demand by optimising operating variables to the preflash units associated with CDUs

(Ledezma-Martínez et al., 2018).

Shortcut models were advanced to account for existing HEN details with respect to the

optimisation of refinery CDUs (Ochoa-Estopier et al., 2014). Such models were applied

to optimise existing CDUs with objectives of energy efficiency enhancement and CO2

emission minimisation. Also other models of CDUs were developed for various design

and optimisation objectives and purposes, including area-based (Ochoa-Estopier et al.,

2018), rigorous simulation (Gadalla et al., 2015) and nonlinear programming models

(López et al., 2013). Improvements were also extended to entail the process itself, and

studies have reported investigating feed blending for desired products maximisation

(Cerdá et al., 2017) and two-column process for energy demands reduction (Kim, 2017).

Therefore, there is a regular concern to generate solutions to enhance the energy

efficiency and minimise emissions within existing refinery CDUs.

It is worth mentioning that refineries and HEN units studied in most reported research

outcomes on retrofit/analysis of HENs were relatively old. This implied large potential

energy savings when these networks were retrofitted and thus results were significant.

On the other hand, results of modern refineries would not be very promising providing

that their actual operated energy efficiencies are relatively high. Retrofit of such

refining units would be very challenging. In the present study, process simulation and

performance evaluation were performed for an existing crude distillation unit present in

a local modern refinery in Egypt with a high energy efficiency. A rigorous process

model was developed using Aspen HYSYS commercial software, and the results were

validated with real data, considering specific crude oil properties. Pinch Analysis

technique was conducted for the existing HEN to define the minimum energy

4
requirements for hot utility and cold utility, with the aid of Aspen Energy Analyser. An

algorithm involving the incorporation of process modifications was proposed. Process

modifications were applied on the case studied, achieving additional savings beyond the

energy targets identified. Environmental assessment was finally carried out to evaluate

CO2 emission reduction levels. The study provides tools addressing the energy aspects

and environmental concerns in CDUs of industrial relevance, which can enhance the

overall sustainability profile of modern refineries.

2. Process description

The crude distillation unit studied typically consists of a topping section, fractionation

section and heat exchange section. This fractionation unit belongs to a modern refinery

located in Egypt with originally a high energy efficiency. The local refinery has four

CDU units, one of which is more recent compared to the other three. Fig. 1 presents a

schematic of the process flow diagram for the existing refinery CDU.

Fig. 1 Comes Here

The crude oil is fed through the first booster pump to the topping section; this section

contains a preflash (or prefractionator) tower C-1 (20 trays – numbering is bottom-up),

air cooler, trim coolers (water coolers), overhead drum (OVHD D-1), topped crude

pumps, light naphtha product and reflux pumps, topped crude recirculation pumps and

topped crude/fuel oil exchangers. The preflash tower C-1 feed is preheated in the heat

exchange section to 213 °C then fed to the tower on the 16th tray. Overhead vapours are

partially condensed in the air cooler and trim coolers from 110 °C to 40 °C. Vapour-

liquid mixture enters OVHD where uncondensed vapours are routed to the vapour

recovery unit. Condensed liquid is pumped by the light-naphtha product and reflux

5
pumps a cold reflux to the tower C-1 and light naphtha to the vapour recovery unit. To

improve separation and strip the preflash bottom at its bubble-point, two heat

exchangers are used to re-boil topped crude by fuel oil product from the fuel oil pump

discharge directly as a heating medium.

The fractionation section contains fired heater F-1, main fractionator C-2 (32 trays), air

cooler, trim coolers (water coolers), (OVHD D-2) and several pumps to discharge outlet

streams for fuel oil, gas oil, bottom pump-around, kerosene, top pump-around, heavy

naphtha and reflux. Topped crude is heated from 230 °C to a temperature not more than

360 °C in the fired heater operating conditions (total hot energy consumption of 37.16

MW in the existing refinery) with air-preheater in operation; the main fractionator C-2

is fed with topped crude at temperature not more than 360 °C from the fired heater

below the 5th tray from the bottom. Overhead vapours are totally condensed in the air

cooler E-121 and trim cooler E-124 from 130 °C to 40 °C. The condensate enters the

drum OVHD D-2 where any uncondensed vapours are either routed to the vapour

recovery unit or vented to flare. Condensed liquid is pumped by the heavy naphtha

product and reflux pumps as a cold reflux pumped to the tower C-2 and heavy naphtha

product pumped to storage tanks. Kerosene is drawn from the 21th and 23rd stages to

feed kerosene side-stripper C-3. Top pump-around reflux is drawn from the kerosene

draw-off to the heat exchange section then returned to the tower above trays 22 and 24.

Gas oil is drawn from the 9th and 11th trays to feed kerosene side-stripper C-4. Bottom

pump-around reflux is drawn from the gas oil draw-off to the heat exchange section

then returned to the tower above stage number 12. Fuel oil is drawn from the bottom to

the heat exchange section.

6
The heat exchange section contains a HEN complex, where the heat content of products

(kerosene, gas oil and fuel oil) is used to preheat the crude oil. Not only the products

heat content, but also that of intermediate circulating refluxes (top pump-around and

bottom pump-around) is used to preheat the crude oil before desalting and preflash. In

addition, the fuel oil heat content is utilised to re-boil the preflash bottom for better

separation.

The heat exchange section is divided into three parts as following and depicted in Fig. 2

for the existing network. Furthermore, Table 1 shows the exchanger details for hot/cold

streams exchanging heat, their temperatures, existing exchanger areas, and heat loads.

The existing HEN includes 24 exchanger units.

1- Preheating of the crude before desalting, where the crude is preheated before

desalting as follows:

a- against kerosene in (E-100).

b- against top pump-around in (E-101b, E-102).

c- against gas oil in (E-103).

d- against fuel oil in (E-105, E-106, E-107).

The crude temperature becomes about 139 °C while kerosene, gas oil and fuel oil

temperatures become 55, 110 and 138 °C, respectively.

• kerosene is cooled to 40 °C in water cooler (E-118).

• gas oil is cooled to 55 °C in water cooler (E-104).

• fuel oil is cooled in water heater (E-117) then water cooler (E-119) to 80 °C.

2- Preheating the crude before preflash tower, such that the crude at temperature of 139

°C is heated to 213 °C against:

a- bottom pump-around in (E-109).

7
b- fuel oil in (E-111, E-112, E-113).

3- Re-boiling of preflash bottom; in here, the bottom of preflash tower (topped crude) is

re-boiled by the fuel oil from the main tower in exchangers (E-114a, E-114b).

Fig. 2 Comes Here

Table 1 Comes Here

3. Research methods

The research approach introduced in the current work has main aspects to focus on.

Firstly, it concentrates on the process analysis through the simulation phase of the CDU

studied. Existing data from the refinery are extracted and validated with the simulation

results with respect to product specifications as hard constraints for the process. Then,

energy targeting is identified using Pinch Analysis technique to ascertain Composite

Curves (CC) and consequently determining minimum energy requirements for the

existing CDU without major modifications, except for HEN structure. The HEN, the

preflash and the main column, are hence simulated while the products specifications,

yields, process conditions, energy consumption, and intermediate temperatures are used

as validation constraints between the simulation model and existing actual data. Process

modifications are finally proposed with the objective of achieving extra savings in terms

of energy demands, beyond targets set by initial energy analysis. Fig. 3 shows the

algorithm used in the present study.

Fig. 3 Comes Here

3.1. Simulation of CDU

Building a rigorous and robust simulation model for a real CDU is a challenging task

and involves many iteration-based calculations, such as recycle and reflux streams,

8
integrated hot/cold streams, pump-arounds (PAs) liquid draws, etc. Aspen HYSYS V8.4

of an academic license is used for the simulation, modification, and validation purposes

of the work. The model can be attributed to include all equipment details; a good

strategy is to initially model the fractionation section to get all products that are

necessary for pre-heat train network including the pump-arounds (side coolers). Further,

the structure of the main column in the real plant should ideally match the simulated

model unit (e.g., pump-arounds draws location, steam flow of the bottom of the column

and side strippers, distillation number of stages, initial and final boiling points (IBP,

FBP) of products, beside cut points and gap/overlap of products’ draws from the main

column). Process parameters employed during the simulation include crude oil feed

composition, temperature and pressure, flow rate and intermediate temperatures

between exchangers/units of flowsheet. As for the fractionation column, process

parameters are the reflux ratio in the main tower, the steam flow and conditions in main

tower and side strippers, the liquid flow rate and the temperature differences across

PAs. These process parameters are adjusted in a way to validate the process simulation

results with the existing refinery data. Product flow rates are, on the other hand,

adjusted as those for the existing refinery data. The product specifications of the crude

oil productions are checked/maintained while the simulation is executed.

To define the crude oil feed stream to the simulation, a crude oil assay is needed. The

design basis of the real plant is based on Shukair blend crude with assay given in Table

2 for true boiling point (TBP) with its products ASTM distillation (Table 3). Aspen

HYSYS represents the assay data given to a set of hypothetical pseudo-components plus

water and light ends (C5-). The thermodynamic property model PR (Peng Robinson) is

9
defined as fluid package for simulation basis which is commonly used for organic

hydrocarbons (Yela, 2009).

Table 2 Comes Here

Table 3 Comes Here

3.2.Energy analysis

Process streams of the existing data are extracted from the real process, and energy

targeting using Pinch Analysis principles is performed. Pinch Analysis of the heat

exchanger network is used to get the optimal network design. The analysis and targeting

are made at the minimum temperature approach difference of the existing HEN. The

optimisation of the network is carried out based on minimising the energy consumption

of the existing process. The energy consumption defined by the energy targets are

achieved by modifying the network through adding extra heat exchangers areas with

stream splitting. Aspen Energy Analyser is used to achieve that task after hand

calculations. The existing heat exchanger network of CDU consists of seven hot streams

and four cold streams in the form required for Pinch Analysis.

The data extraction is the first step in the heat integration task carried out. Process data

streams are extracted for Pinch Analysis from material and heat balances of the

operating CDU and used as input to the Aspen Energy Analyser software platform. The

extracted data of the current process that include heat duty, heat capacity flow rate, and

supply and target temperatures are given in Table 4.

Table 4 Comes Here

3.3. Environmental assessment

10
After applying energy analysis procedure, there is CO2 emissions reduction to be taken

into account with equivalence to fuel burnt in the fired heater. The equation used for

calculating the carbon dioxide emission amount (kg/s) is given below (Gadalla et al.,

2006).

%
CO = ( ) (1)

Where α is the molar masses of CO2 and C (3.67), Qfuel is the amount of fuel burnt in

furnace (kW), NHV (net heating value) with 50% natural gas (51,600 kJ/kg) and 50%

heavy fuel oil (39,771 kJ/kg), and C% is the carbon mass percent in the fuel (75.4% for

natural gas) and (86.5% for heavy fuel oil).

4. Results and discussion

By applying the strategy mentioned in the simulation of CDU section, a convergence of

the simulation model was reached. Fig. 4 displays the complete simulation model

developed for the existing CDU.

Fig. 4 Comes Here

Based on accurate simulation outputs, Table 5 gives the design yields of products for

the real plant based on Shukair blend crude 33.7 API (American Petroleum Institute)

which are used to compare with results obtained in the model validation section.

Table 5 Comes Here

4.1. Model validation and performance assessment

To validate the simulation model developed, a comparison between the simulation

results and the equivalent real data is drawn and presented in Table 6 and Fig. 5 in terms

of products flow rates, furnace duty, condensers duties and pump-arounds temperature

differences. The developed model showed a virtuous agreement with the existing data,

11
as the maximum error obtained between model results and real data was less than 4.5%.

Table 7 compares the gap/overlap points between the real data and simulation results. It

showed good match in terms of model prediction for H.N–kerosene and kerosene–gas

oil. It should be noted that an extra comparison of gap/overlap between L.N/H.N could

have been calculated if L.N was extracted from the main column and not from the

preflash.

Table 6 Comes Here

Fig. 5 Comes Here

Table 7 Comes Here

4.2. Retrofit and energy savings with economic improvements

The minimum temperature approach difference for the existing HEN is extracted from

the data given in Table 1 and found to be approximately 10 °C. The corresponding

Pinch temperatures are 240 and 230 °C for hot and cold Pinches, respectively. The hot

and cold energy targets for the existing HEN are 34.47 and 21.5 MW, respectively.

Despite that, this CDU in particular is regarded to be modern compared with the other

three CDUs in the same refinery, its HEN entails two exchangers across the Pinch. The

existing grid diagram of the CDU process consists of 16 heat exchanger units (process-

to-process), eight cooling utility units (water coolers and air coolers) and one hot utility

unit (furnace).

To improve the energy efficiency of the existing HEN, an optimum exchanger network

was designed with the aid of systematic techniques reported in the field textbooks

(Klemeš, 2013; Smith, 2005) to initially draw the CC and Grand Composite Curves

(GCC), as can be seen in Figs. 6a and 6b.

12
Fig. 6a-6b Comes Here

As reported in Fig. 6a, the hot energy target for the existing refinery is 34.47 MW at

∆Tmin of 10 °C. This results in an energy efficiency of the existing refinery of almost

93% compared with energy targets. Therefore, this reported high efficiency of the

existing plant will make the retrofit task challenging as the potential maximum energy

savings would be limited to 7% in this case. After estimating the minimum heating and

cooling requirements (energy targets) for the process, a certain algorithm should be

respected in terms of designing the HEN diagram that achieves the maximum energy

recovery (Kemp, 2007). During the design of HEN, existing structure is fixed and

streams matching. When the process parameters inside the distillation column are

changed, this change is reflected on the operation of the network so as changing the

temperatures and flow rates of the streams in exchanger units. Also, changing

temperatures and splitting streams would influence the operation of the network leading

to changes in additional areas required, and heat loads. The considered changes in

retrofit are represented by changes in temperatures of streams in HEN in addition to

additional areas and heat loads. An optimum HEN has been successfully designed.

Firstly, by splitting the crude before the desalter to be integrated with the light and

heavy naphtha process streams with respect to their mass heat capacities. As the rule

below the Pinch point should respect the principle CPh≥CPc. Afterwards, by following

the same rule a split stream was integrated with the fuel oil with ratio equals to CP(fuel

oil)/CPcrude and the other split integrated with the kerosene stream. The remainder of the

crude stream was integrated with bottom pump-around (BPA) stream. Secondly, the

crude stream after the desalter was split to four streams with the respect of CPc of top

13
pump-around (TPA), gas oil, fuel oil, and (BPA) respectively. Then, the remainder was

integrated with the Fuel Oil as shown in Fig. 7.

Fig. 7 Comes Here

Above the Pinch region, is for the segment reflecting the considerable economic and

energy savings for heating requirements of the network. An integration with the second

heat exchanger of the existing HEN was applied to exploit the highest temperature of

the fuel oil to get around 9 °C extra to the topped crude before the fired heater. In

addition, a new heat exchanger of bottom pump-around was added to be integrated with

the bottom C-1 stream that acts as re-boiler; with this effective replacement, more

energy savings were reached. A better heat integration was achieved by raising

temperature before the fired heater to 239.4 °C instead of 230 °C. The potential energy

saving includes the heating utility saving of about 7% and the cooling utility saving of

11% over the current design. Consequently, the amount of CO2 emissions for the new

network is estimated to be 52,705 t/y using natural gas as fuel gas to save 23.5%

comparing with the existing network which emits 68,898 t/y using natural gas and

heavy fuel oil as combined source of fuel for the furnace.

4.3. Process improvements

In order to enhance the process performance and minimise energy consumption, a new

process cycle is projected, with ensuring the column and products specifications are

fixed without violation. A preflash pump-around cycle could be installed to use excess

energy in HEN (Errico et al., 2009; Waheed and Oni, 2015). However, this modification

will increase the vaporisation rate in the preflash and diversify its temperature profile,

which lets heavier components to amend the distillate product composition. So, this

14
modification was not implemented due to design limitations of the column, for avoiding

high pressure, entrainment or flooding during operation. However, there are some

solutions in operation to reduce vaporization rate i.e. reducing reboiler duty, increasing

reflux by the air coolers rotor speed or decreasing feed rate to the column. Nevertheless,

feasible operation changes with zero topology modifications, like changing intermediate

temperature of CDU top vapours or increasing temperature difference of CDU pump-

arounds has proven to result into considerably minor savings (Gadalla et al., 2016).

Thus, a novel modification is proposed through exchanging a total of 8 MW of energy

via heater E-101 between (two material streams) topped crude stream and new pump-

around stream with amount of 112 t/h drawn from stage four at 337 °C. In real plant

revamping, this instillation will be more than one heat exchanger due to the large area to

be installed.

Since it is the highest stage temperature of the column profile, the new installation of

pump-around at that stage with temperature difference of 95 °C from return temperature

to stage five acts as 2nd bottom pump-around. E-101 is simulated as a heater and tied

with stream set SET-1 to the new pump-around cooler inside C-2 column environment

with the same energy quantity for both the heater and the cooler, as shown in Fig. 9.

Fig. 9 Comes Here

The outlet steam of the heater stream 17 enters a flash drum V-101 to separate 18.6% of

the vapour fractions resulted from the extra heat gained by the topped crude and

introduced to the main column directly to decrease the head load of the fired heater F-1

by decreasing the total mass flow rate inside the fired heater. The remainder liquid

stream 18 is pumped by P-8 to F-1 at 262 °C instead of 231 °C to achieve a total of

15
24.76 MW hot utility, resulting in 31.3% reduction of fuel consumption of the fired

heater and 45.1% CO2 emissions reduction over the existing design.

4.4. Energy and economic comparisons for retrofit scenarios

A complete comparison of the existing case, first retrofit solution and second retrofit

solution is presented in Table 8, showing a summary of energy savings and emissions

savings with economic consideration for each scenario and their supplementary

additional capital investment.

Table 8 Comes Here

The utility cost for fuel and cooling water are obtained from actual plant as 5

$/MMBTU for fuel gas and 13.44 $/MMBTU for fuel oil. Additional capital investment

was related to the additional area installed for each scenario, providing that the existing

column and the flowsheet structure are fixed. Thus, additional capital costs are obtained

from the economic equations in manual of process economic evaluation (Chauvel et al.,

2003) with additional cost index modifications. For retrofit scenarios and calculations,

existing exchanger areas are fixed, while additional areas for retrofit solution are

estimated for new heat loads. For these solutions, additional capital investments were

inflicted whereas energy and utility savings were obtained. The payback time for first

scenario was almost one year and a half for energy savings equivalent to 670,247 $/y.

Second retrofit had a higher and more substantial energy savings equivalent to

3,089,618 $/y and less than a year as a payback period. CO2 releases were significantly

minimised by 45.1% regarding to 8,699 kg/h of the existing plant releases.

5. Conclusions

16
A process simulation model has been developed for a modern crude distillation unit

with preflash existed in an oil refinery. The model has been validated with real

operation parameters and intermediate details of the CDU, showing a good agreement in

the comparative results. A heat integration retrofit analysis has been carried out to

identify energy targets and potential savings for the existing HEN. Pinch Analysis

Technique was adopted for designing a heat integration network that improves energy

efficiency of the process. New modifications of the process design showed that stream

splitting of the main crude stream before and after the desalter was located below the

Pinch point, which subsequently decreased the cooling utility and electricity

consumption. Above the Pinch, better heat integration has been achieved with raising

the temperature before the fired heater to 239.4 °C instead of 230 °C. The potential

energy saving included heating utility saving of 7.2% and cooling utility saving of

11.1%, with 23.5% CO2 emissions reduction over the current design. Furthermore, the

developed simulation model has been used to propose process modifications, which are

then applied to achieve less energy consumption than defined by the energy targets.

This led to substantial energy savings beyond the Pinch limitations by 31.3% and 45.1%

CO2 emissions reduction.

Acknowledgements

We thank Dr. Ehab Zahra, Deputy CEO EPROM, for sharing ideas and practical

insights throughout this work. Our gratitude is extended to Eng. Mohamed El-Sayed,

Process Engineer at EPROM, his valuable interactions and practical experience are

appreciated.

Nomenclature

17
CDU crude distillation unit

Tmin minimum temperature approach difference

HEN heat exchangers network

L.N Light Naphtha

H.N Heavy Naphtha

OVHD overhead drum

TPA top pump around

BPA bottom pump around

CP heat capacity

CC Composite Curves

GCC Grand Composite Curve

IBP initial boiling point

FBP final boiling point

α molar masses

Qfuel fuel burnt in furnace, MW

NHV net heating value

C carbon mass percent in the fuel

References

Abdul-Manan, A.F.N., Arfaj, A., Babiker, H., 2017. Oil refining in a CO2 constrained

18
 world: Effects of carbon pricing on refineries globally. Energy 121, 264–275.

https://doi.org/10.1016/j.energy.2017.01.046

Bagajewicz, M., Lambeth, A., Valtinson, G., 2014. New Technologies To Enhance the

Distillation Yield of Petroleum Fractionation. Ind. Eng. Chem. Res. 53, 16937–

16947. https://doi.org/10.1021/ie5000455

Cerdá, J., Pautasso, P.C., Cafaro, D.C., 2017. Scheduling Multipipeline Blending

Systems Supplying Feedstocks to Crude Oil Distillation Columns. Ind. Eng. Chem.

Res. 56, 10783–10797. https://doi.org/10.1021/acs.iecr.7b02449

Chauvel, A., Fournier, G., Raimbault, C., 2003. Manual of Process Economic

Evaluation.

Čuček, L., Boldyryev, S., Klemeš, J.J., Kravanja, Z., Krajačić, G., Varbanov, P.S., Duić,

N., 2019. Approaches for retrofitting heat exchanger networks within processes

and Total Sites. J. Clean. Prod. https://doi.org/10.1016/j.jclepro.2018.11.129

Enríquez, A.H., Binns, M., Kim, J.-K., 2014. Systematic retrofit design with Response

Surface Method and process integration techniques: A case study for the retrofit of

a hydrocarbon fractionation plant. Chem. Eng. Res. Des.

https://doi.org/10.1016/j.cherd.2014.02.030

Errico, M., Tola, G., Mascia, M., 2009. Energy saving in a crude distillation unit by a

preflash implementation. Appl. Therm. Eng. 29, 1642–1647.

https://doi.org/10.1016/j.applthermaleng.2008.07.011

Gadalla, M., Olujić, Ž., Jobson, M., Smith, R., 2006. Estimation and reduction of CO2

emissions from crude oil distillation units. Energy 31, 2398–2408.

https://doi.org/10.1016/j.energy.2005.10.030

Gadalla, M.A., Abdelaziz, O.Y., Ashour, F.H., 2016. Conceptual insights to

19
 debottleneck the Network Pinch in heat-integrated crude oil distillation systems

without topology modifications. Energy Convers. Manag. 126, 329–341.

https://doi.org/10.1016/j.enconman.2016.08.011

Gadalla, M.A., Abdelaziz, O.Y., Kamel, D.A., Ashour, F.H., 2015. A rigorous

simulation-based procedure for retrofitting an existing Egyptian refinery

distillation unit. Energy 83, 756–765. https://doi.org/10.1016/j.energy.2015.02.085

Gu, W., Huang, Y., Wang, K., Zhang, B., Chen, Q., Hui, C.-W., 2014. Comparative

analysis and evaluation of three crude oil vacuum distillation processes for process

selection. Energy 76, 559–571. https://doi.org/10.1016/J.ENERGY.2014.08.053

Joe, J.M., Rabiu, A.M., 2013. Retrofit of the Heat Recovery System of a Petroleum

Refinery Using Pinch Analysis. J. Power Energy Eng. 47–52.

Kamel, D.A., Gadalla, M.A., Abdelaziz, O.Y., Labib, M.A., Ashour, F.H., 2017.

Temperature driving force (TDF) curves for heat exchanger network retrofit – A

case study and implications. Energy 123, 283–295.

https://doi.org/10.1016/j.energy.2017.02.013

Kemp, I.C., 2007. Pinch analysis and process integration: A user guide on process

integration for the efficient use of energy. Pinch Anal. Process Integr. 416.

https://doi.org/http://dx.doi.org/10.1016/B978-075068260-2.50003-1

Kim, Y.H., 2017. An Energy-Efficient Crude Distillation Unit with a Prefractionator.

Chem. Eng. Technol. 40, 588–597. https://doi.org/10.1002/ceat.201600387

Klemeš, J.J., 2013. Handbook of Process Integration (PI). Woodhead Publishing Ltd.,

Cambridge CB22 3HJ, UK.

Ledezma-Martínez, M., Jobson, M., Smith, R., 2018. Simulation-Optimization-Based

Design of Crude Oil Distillation Systems with Preflash Units. Ind. Eng. Chem.

20
 Res. https://doi.org/10.1021/acs.iecr.7b05252

López, D.C., Hoyos, L.J., Mahecha, C.A., Arellano-Garcia, H., Wozny, G., 2013.

Optimization model of crude oil distillation units for optimal crude oil blending

and operating conditions. Ind. Eng. Chem. Res. 52, 12993–13005.

https://doi.org/10.1021/ie4000344

Mehdizadeh-Fard, M., Pourfayaz, F., 2019. Advanced exergy analysis of heat

exchanger network in a complex natural gas refinery. J. Clean. Prod. 206, 670–687.

https://doi.org/10.1016/j.jclepro.2018.09.166

Nalinakshan, S., Sivasubramanian, V., Ravi, V., Vasudevan, A., Sankar, M.S.R.,

Arunachalam, K., 2019. Progressive crude oil distillation: An energy-efficient

alternative to conventional distillation process. Fuel.

https://doi.org/10.1016/j.fuel.2018.11.033

Nemet, A., Klemeš, J.J., Varbanov, P.S., Mantelli, V., 2015. Heat Integration retrofit

analysis—an oil refinery case study by Retrofit Tracing Grid Diagram. Front.

Chem. Sci. Eng. 9, 163–182. https://doi.org/10.1007/s11705-015-1520-8

Ochoa-Estopier, L.M., Jobson, M., Chen, L., 2018. Area-based optimization approach

for refinery heat exchanger networks. Appl. Therm. Eng. 129, 606–617.

https://doi.org/10.1016/j.applthermaleng.2017.10.049

Ochoa-Estopier, L.M., Jobson, M., Smith, R., 2014. The use of reduced models for

design and optimisation of heat-integrated crude oil distillation systems. Energy

75, 5–13. https://doi.org/10.1016/j.energy.2014.06.043

Pejpichestakul, W., Siemanond, K., 2013. Retrofit of Refinery Heat Exchanger Network

under Different Kinds of Crude Oil by Pinch Design Method using Mathematical

Programming. Chem. Eng. Trans. 32, 1411–1416.

21
Smith, R., 2005. Chemical Process Design and Integration. John Wiley & Sons Ltd, The

Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England.

Ulyev, L., Vasiliev, M., Boldyryev, S., 2018. Process integration of crude oil distillation

with technological and economic restrictions. J. Environ. Manage.

https://doi.org/10.1016/j.jenvman.2018.05.062

Waheed, M.A., Oni, A.O., 2015. Performance improvement of a crude oil distillation

unit. Appl. Therm. Eng. 75, 315–324.

https://doi.org/10.1016/j.applthermaleng.2014.10.078

Waheed, M.A., Oni, A.O., Adejuyigbe, S.B., Adewumi, B.A., 2014. Thermoeconomic

and environmental assessment of a crude oil distillation unit of a Nigerian refinery.

Appl. Therm. Eng. 66, 191–205.

https://doi.org/10.1016/j.applthermaleng.2014.02.007

Yela, S., 2009. Framework for Operability Assessment of Production Facilities: An

Application to a Primary Unit of a Crude Oil Refinery. Master’s Thesis, Lousiana

State University, US.

22
Figure captions

Fig. 1. Simplified process flow diagram of existing CDU.

Fig. 2. Existing heat exchanger network.

Fig. 3. The approach methodology for simulation and energy optimisation of modern

CDUs.

Fig. 4. The developed simulation model for CDU case study.

Fig. 5. Products flowrates comparison of real data and simulation model values.

Fig. 6. Composite Curves with ∆Tmin 10 °C (a) Hot and Cold CC (b) Grand Composite

Curves GCC.

Fig. 7. The optimum HEN grid diagram.

Fig. 8. Illustration of new process modification into the simulation model.

Table captions

Table 1. Existing heat exchanger network design data and specifications.

Table 2. Crude Assay for TBP distillation of Shukair blend crude oil.

Table 3. ASTM D86 distillation for light naphtha, heavy naphtha, kerosene and gas oil.

Table 4. Design yields for Shukair blend crude oil.

Table 5. Data extraction (process streams).

Table 6. Real data vs simulation model results.

Table 7. ASTM D86 gap and cut points for certain products.

Table 8. Existing vs optimum HEN.

23
Fig. 1. Simplified process flow diagram of existing CDU.
24
Fig. 2. Existing heat exchanger network (design data given in Table1).

25
 Energy
Integration

Fig. 3. The approach methodology for simulation and energy optimisation of modern

CDUs.

26
Fig. 4. The developed simulation model for CDU case study.

27
 180
156.9 159.3
160

140

120

100
Real data
80

Flow rate (t/h)

Simulated
60
46.1 45.7
40.9 41.0
40
22.5 21.9 23.6 22.1
20

0
L.N H.N Kerosene Gas oil Fuel oil

Fig. 5. Products flowrates comparison of real data and simulation model value.

28
Fig. 6. Composite Curves with ∆Tmin 10 °C (a) Hot and Cold CC (b) Grand Composite

Curves GCC.

29
Fig. 7. The optimum HEN grid diagram.

30
Fig. 8. Illustration of new process modification into the simulation model.

31
Table 1. Existing heat exchanger network design data and specifications.

Heat exchanger Cold stream Cold Tin (°C) Cold Tout (°C) Hot stream Hot Tin (°C) Hot Tout (°C) Area (m2) Duty (MW)

E-100 Crude Before 1 25 41 Kerosene 1 146 55 235 2.4

E-102 Crude Before 2 41 59 TPA 1 126.7 82.2 235 3.8

E-101b Crude Before 3 59 81 TPA 2 176 126.7 235 3.6

E-103 Crude Before 4 81 103 Gas oil 226 110 235 3.7

E-108 Crude After 1 139 146 BPA 1 162.2 149 274 1.5

E-109 Crude After 2 146 154 BPA 2 196 162.2 274 1.9

E-110 Crude After 3 154 176.8 BPA 3 254 196 274 4.3

E-105 Crude Before 5 103 117.2 Fuel oil 8 155 137.78 235 2.8

E-106 Crude Before 6 117.2 126.6 Fuel oil 7 175 155 235 1.6

E-107 Crude Before 7 126.6 139 Fuel oil 6 199 175 235 2.3

32
E-111 Crude After 4 176.8 185 Fuel oil 5 215.5 199 274 2.2

E-112 Crude After 5 185 200.5 Fuel oil 4 240.5 215.5 274 3.2

E-113 Crude After 6 200.5 213 Fuel oil 3 265.5 240.5 274 2.8

E-114a Bottom C1 232 244 Fuel oil 2 285 265.5 274 3.0

E-114b Bottom 2 C1 244 260 Fuel oil 1 312 285 274 2.9

F-1 Topped Crude 230 360 Fired Heater 37.2

E-118 Cooling Water 1 30 37 Kerosene 2 54 40 69 0.4

E-104 Cooling Water 2 30 37 Gas oil 2 82 54 162 1.6

E-117 Cooling Water 3 30 37 Fuel oil 9 125 100 150 1.1

E-119 Cooling Water 4 30 37 Fuel oil 10 100 76 143 2.1

E-123 Cooling Water 5 30 40 L.N 60 40 228 1.9

E-124 Cooling Water 6 30 40 H.N 60 40 228 2.7

E-122 Air Cooling 1 33 35 L.N 110 60 4.8

E-121 Air Cooling 2 30 33.3 H.N 130 60 9.5

33
Table 2. Crude Assay for TBP distillation of Shukair blend crude oil.

Liquid mass percent, % Temperature, °C

8 100

12 125

16 150

20 175

24 200

28 225

30 250

35 275

40 300

49 355

57 400

67 455

77 510

34
Table 3. ASTM D86 distillation for light naphtha, heavy naphtha, kerosene and gas oil.

Liquid volume percent, L.N, °C H.N, °C Kerosene, °C Gas oil, °C

IBP 45 99 162 257

10 48 106 169 266

30 57 118 186 281

50 64 126 204 291

70 73 135 220 304

90 89 149 237 323

FBP 104 160 254 345

35
Table 4. Data extraction (process streams).

Stream Type Inlet Outlet Specific Cp Flow rate Enthalpy

name Temp. Temp. (kJ/kg °C) (kg/h) (MW)

(°C) (°C)

Kerosene HOT 146 40 2.303 40950 2.7765

TPA HOT 176.7 82.2 2.461 113578 7.3371

Gas oil HOT 226 60 2.48 46165 5.2788

BPA HOT 254.4 149 2.648 99879 7.7438

FO HOT 312.8 80 2.358 156963 23.930

Crude COLD 25 139 2.174 291667 20.083

Before

Desalter

Crude After COLD 139 213 2.669 291667 16.003

Desalter

Bottom C1 COLD 232 260 2.813 267703 5.8576

Topped COLD 230 360 3.844 267703 37.160

Crude

L.N HOT 110 40 6.627 52690 6.7893

(Reflux +

Product)

H.N HOT 130 40 8.268 59376 12.273

(Reflux +

Product)

36
Table 5. Design yields for Shukair blend crude oil.

Product Yield (wt%)

Fuel gas 0.193

Propane 0.312

LPG 1.836

Stabilised naphtha 5.875

Heavy naphtha 8.10

Kerosene 14.04

Gas oil 15.828

Fuel oil 53.816

37
Table 6. Real data vs simulation model results.

Parameter Real data Simulation value

Feed, t/h 291.7 291.7

Furnace outlet temperature, °C 360.0 360.0

Furnace inlet temperature, °C 230.0 231.1

Furnace duty, MW 37.16 35.54

Top pump-around, t/h 113.6 114.1

Top pump-around draw temp., °C 176.7 173.4

Bottom pump-around, t/h 99.88 99.02

Bottom pump-around draw temp., °C 254.5 257.8

C-1 condenser duty, MW 6.74 6.67

C-2 condenser duty, MW 12.44 12.32

38
Table 7. ASTM D86 gap and cut points for certain products.

5–95 Gap/Overlap Real plant (°C) Simulation (°C)

H.N–kerosene 11 11.54

kerosene–gas oil 16 26.83

39
Table 8. Existing vs retrofitting scenarios.

Factor Heating Cooling Reduction in Energy reduction Additional area (m2) Additional capital Payback
(MW) (MW) CO2 (%) investment ($) (y)
emissions (%)
Existing case 37.16 27.7 – – – – –

1st retrofit 34.47 21.5 23.5 (7.21 Heating 5,850 1,056,937 1.57

(HEN optimisation) +11.08 Cooling)

2nd retrofit 24.76 24.2 45.1 31.3 Heating 16,630 3,004,594 0.97

(novel process

configuration)

40