Mamdouh Article 2 PDF

Energy xxx (2015) 1e16
Contents lists available at ScienceDirect
Energy
journal homepage: www.elsevier.com/locate/energy
A new graphical method for Pinch Analysis applications: Heat

exchanger network retrofit and energy integration
Mamdouh A. Gadalla*
Dept. of Chemical Engineering, The British University in Egypt (BUE), El Sherouk City, Misr-Ismalia Road, 11837, Egypt
a r t i c l e i n f o a b s t r a c t
Article history: Energy integration is a key solution in chemical process and crude refining industries to minimise
Received 9 July 2014 external fuel consumption and to face the impact of growing energy crises. Typical energy integration
Received in revised form projects can reach a reduction of heating fuels and cold utilities by up to 40% compared with original
18 November 2014
designs or existing installations. Pinch Analysis is a leading tool and regarded as an efficient method to
Accepted 5 December 2014
increase energy efficiency and minimise fuel flow consumptions. It is valid for both natures of design,
Available online xxx
grassroots and retrofit situations. It can practically be applied to synthesise a HEN (heat exchanger
network) or modify an existing preheat train for minimum energy consumption. Heat recovery systems
Keywords:
Pinch Analysis
or HENs are networks for exchanging heat between hot and cold process sources. All heat transferred
Graphical technique from hot process sources into cold process sinks represent the scope for energy integration. On the other
Energy saving hand, energies required beyond this integrated amount are to be satisfied by external utilities. Graphical
Retrofit representations of Pinch Analysis, such as Composite and Grand Composite Curves are very useful for
Energy integration grassroots designs. Nevertheless, in retrofit situation the analysis is not adequate and besides it is
Crude oil graphically tedious to represent existing exchangers on such graphs.
This research proposes a new graphical method for the analysis of heat recovery systems, applicable to
HEN retrofit. The new graphical method is based on plotting temperatures of process hot streams versus
temperatures of process cold streams. A new graph is constructed for representing existing HENs. For a
given network, each existing exchanger is represented by a straight line, whose slope is proportional to
the ratio of heat capacities and flows. Further, the length of each exchanger line is related to the heat flow
transferred across this exchanger. This new graphical representation can easily identify exchangers
across the pinch, Network Pinch, pinching matches and improper placement of fuel consumption.
Furthermore, such a graph can recognise promising modifications to improve the energy performance
and hence less fuel and cooling water requirement. Graphs developed in this work can be used to analyse
the energy performance of existing networks with respect to energy targets. They can also be used in
junction with the background process to modify basic designs or existing network for better energy
integration opportunities and minimum fuel demands. The application of the new graphical method to a
case study showed savings of approximately 17% in energy demands and fuel consumption.
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction optimise existing operating conditions for minimum energy de-

mands. More general, Smith in his textbook [2] devoted complete
Chemical plants, including crude oil up-stream processing chapters to introduce the fundamentals of energy consumption in
units are energy-intensive process industries. Among these unit distillation industry, distillation sequencing, retrofit of distillation
operations and process equipment, distillation is found to be the units, optimisation of superstructure for distillation sequencing,
most energy consuming unit. As an example, Gadalla et al. [1] heat integration in distillation, etc. The textbook also considered
reviewed the energy problems of refinery crude distillation normal distillation as well as azeotropic mixtures, and presented
plants as major energy intensive processes, and focused to in detail all principles and design guidelines of Pinch Analysis. On
the other hand, El-Halwagi [3] considered distillation industries in
many of his applications for energy integration, optimisation and
* Tel.: þ20 22 689 0000x1471; fax: þ20 22 630 0010. sustainability. This book additionally introduced the conceptual
E-mail address: mamdouh.gadalla@bue.edu.eg.
http://dx.doi.org/10.1016/j.energy.2014.12.011
0360-5442/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Gadalla MA, A new graphical method for Pinch Analysis applications: Heat exchanger network retrofit and
energy integration, Energy (2015), http://dx.doi.org/10.1016/j.energy.2014.12.011
2 M.A. Gadalla / Energy xxx (2015) 1e16
design of distillation-based bio-refineries as alternative sustain- Literature is rich in methods and approaches for energy inte-
able plants. Significant amounts of energy are essential for heating gration; Pinch Analysis or Pinch Technology is one very important
and cooling purposes within the process or to provide necessary method among others for the efficient use of energy and other raw
heat of reactions. Typically in crude refining distillation, large materials, such as water, hydrogen, etc. Most previous approaches
amounts of fuel oil, natural gas, or in some cases part of the crude are applied to increase energy efficiency of preheat trains used in
oil processed are burned in the fired heater to provide the energy chemical and petrochemical industries. In one of these ap-
required for crude fractionation. This applies to a wide range of proaches, the existing distillation unit was optimised simulta-
industries, such as petrochemicals, chemicals separation, distil- neously with the associated exchanger network to reduce energy
lation industries, etc. Also, in some applications cooling is consumption and for more capacity to be processed [5]. Shortcut
required to remove heat of reactions. Typical external utilities models were developed within the work to model the retrofit of
used for process industries are steam, hot oil, petroleum fuels, flue distillation columns. Also, in this method, the details of the
gases, cooling water, low-temperature cooling liquids, air, etc. existing exchanger network were considered in the optimisation
Steam for heating or stripping is produced through the combus- algorithm through an empirical equation obtained by an extensive
tion of natural gas or crude oil fuel products. In typical process separate retrofit study. Other researchers focused on the distilla-
design of chemical plants, the design normally starts with the tion column as stand-alone equipment and increased its energy
reactor, the heart of the process. Then the design proceeds to the efficiency through an internal heat integration scheme [6]. The
separation task to recover un-reacted materials, leaving as much proposed scheme exploited the energy of the hot vapours after
as possible pure product streams. Un-reacted materials are recy- being compressed in heating the cold liquid of rectifying sections
cled to the reactor for more reaction opportunities, while the in an external medium, known as heat panels. These new columns
product is further purified to reach required market specifications. are configured in a specific manner commonly called internal
Next, the design of heat exchanger network comes into stage. In HIDiCs or internal heat integrated distillation columns. Asante
this stage, all heating and cooling requirements are provided by and Zhu in their research works [7,8] developed a breakthrough
either heat integration or through external utilities. The last phase approach known as Network Pinch method for retrofitting an
of design is related to the utility system, in which appropriate existing exchanger networks. In this approach, the bottlenecks
utilities are selected to supply the external heating and cooling limiting heat recovery were first identified and then topological
demands (see Fig. 1). This hierarchical procedure is called the modifications were suggested to increase the energy recovery.
‘onion diagram’ and is given in detail by Smith [2]. Given that Bottlenecked exchanger units in this method were termed as
distillation processes like other operations require substantial Pinched Matches. This last work used mathematical programming
amounts of energy as hot or cold, many research efforts were put techniques in solving the retrofit problem. Based on the work of
into action to identify opportunities for minimising energy con- Asante and Zhu [7,8], a group of researchers extended the
sumption. Energy integration in processes is meant by recovering Network Pinch approach to revamp industrial heat exchanger
heat from those process streams and units that need cooling and networks by considering the change in the thermal properties of
thus through heating other streams or units which require heat- process streams [9]. In the latter work, structural modifications
ing. Hot streams or units are heat sources, while cold streams or and cost optimisation were combined to minimise topology
units are known as heat sinks. This process is commonly known as modifications. Also, Bakhtiari and Bedard [10] modified the
heat integration or energy recovery. Such integration takes place standard Network Pinch approach by Asante and Zhu to include
in a HEN (heat exchanger network) or a preheat train. Heat is practical features of stream segmentation and splitting. This last
recovered from process hot sources to process heat sinks as much work considered the advantages of varying the supply and target
as possible such that energy consumption is reduced. The total temperatures of streams which increased the flexibility of the
energy consumption of a process is the amount of energy supplied work. On the other hand, Zhang et al. [11] used more recent
by or the flows of external utilities in heaters or coolers. The more modelling techniques and Process Integration technologies to
external utility required by the process, the more utility cost improve the environmental performance of some existing crude
necessary for the process. Petroleum refining distillations are very distillation units. In this work, the researchers adopted the opti-
typical industries with preheat trains for the processing of crude misation of the operating conditions to reduce the energy con-
oil streams [4]. sumption of an industrial facility by some 8%. In a very recent
work by Lluvia et al. [12], both the distillation process and the heat
recovery system were optimised simultaneously to reach best
results. This recent work combined models for describing distil-
lation columns with models for the details of the existing
exchanger networks to perform potential changes to each stage.
Previous shortcut models [5] were extended in the work of Lluvia
et al. [12] which resulted in semi-rigorous models based on stage-
by-stage calculations.
Jiang et al. [13] considered retrofitting existing heat exchanger
network through the selective use of heat transfer enhancement.
In this work, the existing structure of the network was fixed while
improving the network energy performance. A good advantage of
this work was that the selective enhancement avoided the addi-
tion of new equipment or piping and engineering costs. In the
same line of intensification in heat transfer, Wang and co-workers
[14,15] addressed a very common problem in retrofit solutions
that the conventional retrofit projects normally require significant
topology modifications with more engineering complexities. This
work thus developed mathematical models to solve the retrofit
Fig. 1. Onion diagram for chemical process design [2]. problem based on heat transfer enhancement. Heat transfer
M.A. Gadalla / Energy xxx (2015) 1e16 3
coefficients and pressured drop using enhancement were pre- modifying the existing networks by adding extra areas, relocating
dicted by these models. In the work of Gu et al. [16], material and units, new exchangers, stream splitting, heat transfer enhance-
energy performances of existing crude distillation units were ment, etc. The success of retrofit approaches greatly depends on
considered, focussing on vacuum towers to recover more energy. whether the energy inefficiencies and potential for improvement of
This work used rigorous simulations to compare and analyse be- existing networks can be identified or not. This could mean (a)
tween different processes' opportunities to increase the furnace determining the relative performance of the existing network with
outlet temperatures. Away from the details of HENs in retrofit, respect to energy targets, (b) quantifying the deviation of the
some researchers focused on the very details of the exchanger existing network from the Pinch Analysis principles, and (c)
equipment in improving the energy recovery of the networks [17]. locating Network Pinch that limits the maximum energy recovery.
In this work, common details of shell-and-tube exchangers were Existing HENs are conventionally represented using grid dia-
included, such as shell passes, tube passes, logarithmic mean grams, such as that shown in Fig. 2. These diagrams are kind of
temperature differences, heat transfer intensifications, etc. sketches and can describe existing networks with details of con-
Considering such details contributed in debottlenecking existing nections but cannot interpret the principles of Pinch Analysis
HENs for case studies. Many other researchers directed their qualitatively nor quantify the energy inefficiencies particularly for
works, focussing on increasing the energy recovery of specific very large HENs. In addition, the location of Network Pinch and
applications, such as refinery HENs considering different crude oil Pinched Matches are not easy to detect using HEN grid diagrams,
origins [18], heat integration in hot oil networks [19], energy and particularly for very complex networks, such as in refining pro-
distance-related costs in heat integration [20], Iranian refineries cesses. At the time that Composite Curves can describe fully new
revamping with topology modifications [21,22], and refinery HEN designs, existing exchangers cannot be represented using
operation optimisation [23]. such curves. In fact, it is very tedious to super-impose one
It is obvious from the above discussion that there have been exchanger match on these Composite Curves, i.e. inconvenient for
many approaches and methods that addressed the problem of heat a very complex HEN. It is valuable then if the existing network can
integration in many applications. Shortcut models, mathematical be represented fully with its exact details of stream matches,
programming, decomposition techniques and process simulators temperatures and duties for better understanding to improve
were all employed in the presented methods. In some of these energy performances. It is also significant to graphically identify
works [e.g. [18,21,22]], the presence of graphical methods would be the locations of Network Pinch and limiting exchanger matches.
significant since many diagrams for HENs modifications were Having represented an existing HEN with all associated details
evolved as result of retrofit. New graphical representations would using new graphs, the Pinch Analysis principles and guidelines
be helpful in analysing and better understanding the performance can be interpreted easier and thus energy improvement are ob-
of existing HENs, and also would assist in evaluating retrofit solu- tained. Such graphs would be a key in performing revamping
tions. Simplicity and ease of construction are important charac- studies. The current research aims to tackle these previous ob-
teristics to be available in such new graphs. servations and is to develop a new graphical tool featured with a
new visualisation capability that will address such issues in
2. Motive of the research work existing methods.
Pinch Analysis or technology is an outstanding methodology for

energy saving in processes where heating and cooling are charac- 3. A new graphical representation for energy analysis
terising the processing operations. The concept of Pinch Analysis is
originally discovered and further developed in late 70s by Hohman As noted above, when a heat exchanger network is to be
[24], Huang and Elshout [25], Linnhoff et al. [26,27] and Umeda designed, Composite Curves and Grand Composite Curves are of
et al. [28]. Yet it is evolving along the years and its applicability is great relevance to determine the energy targets, i.e. minimum hot
broadening, mainly for saving material flows [3] and energy sup- and cold utility consumptions. These curves serve just perfect to
plies. Among other methods of mathematical programming for represent the new exchanger network performance and achieve
energy savings, Pinch Analysis is found very popular and successful the design. For revamping, the situation is different that exchanger
given the fact that it is conceptually simple and with impressive matches are known with known terminal temperatures for cold
results, i.e. 10e35% in energy savings [29]. The Pinch Analysis and hot streams; the physical topology of the network is also
principles are set of rules, established using graphical representa- constrained. The conventional method for representing such
tions such as Composite Curves and Grand Composite curves, or by existing networks is super-imposing exchanger matches on the
calculation-based methods known as the ‘Problem Table Algorithm’ previous mentioned curves, i.e. Composite Curves. This graphical
[30]. Smith [2] and others [3,26e28] provide a systematic proce-
dure on how to construct Composite Curves for a given process, and
outline the main features of such graphs. Maximum heat recovery
is achieved within a given process, by complying with the princi-
ples of Pinch Analysis in the design for a given DTmin. Also in retrofit
situation, the energy efficiency can be improved by removing vio-
lations to these principles.
Composite Curves have been used widely to determine the en-
ergy targets and guide the design of new HENs. Also the Pinch
design method was applied in many energy applications to syn-
thesise new HENs. At the time that new exchanger networks are
commonly designed for many chemical/refining processes, existing
HENs are rather important to be reused or revamped. The reuse/
revamping is necessary for many objectives, such as energy saving,
capacity increase, debottlenecking, or changing feed or product Fig. 2. A grid diagram for a complete HEN design (numbers below exchangers refer to
specifications. These objectives are achieved mostly through heat duties in MW, H: heater, C: cooler).
method seems tedious to obtain with the difficulty of matching between hot streams below the pinch with cold streams
existing temperatures with those target values of the curves. above the pinch.
Instead, a new graphical representation is proposed that is easier to
achieve and is of significance for energy analysis of new designs and The new graphical representation shown in Fig. 3 features the
in evaluating the performance of the existing HEN. following:
The new graphical representation is obtained by plotting the
temperatures of hot streams entering and leaving an exchanger (1) Any point located inside the graph is defined by two tem-
unit versus those corresponding temperatures for cold streams. perature coordinates, i.e. (Tc, Th).
Thus, the y-axis is to represent the temperatures of all hot streams (2) The minimum temperature approach difference appears
around exchangers whereas the x-axis is representing the corre- twice, first with the vertical cold pinch temperature and then
sponding cold streams temperatures, as demonstrated in Fig. 3. Any with the horizontal hot pinch temperature.
process or utility hot stream can be represented by two horizontal (3) The diagonal line in the graph represents the case when hot
lines on the diagram of Fig. 3. The upper horizontal line will stream temperatures approach cold stream temperatures or
represent the supply temperature of the stream, whereas the lower Th ¼Tc. Accordingly the diagonal divides the graph into two
horizontal line will represent the target temperature. Similarly, cold regions of heat integration, feasible energy integration above
process or utility streams will be represented through two vertical the diagonal, and infeasible energy integration below the
lines, one for the supply temperatures (to left side) and another for diagonal. The infeasibility region below the diagonal is due to
the target temperatures (to right side). The pinch temperatures of the fact that within this region any possible heat sources are
the process are plotted on the graph as a vertical line for the cold colder than heat sinks. Therefore heat integration is ther-
pinch temperature and a horizontal line for the hot pinch tem- modynamically impossible. This implies that all exchangers
perature. These two lines divide the graph into four distinct regions, integrating heat must lie above the diagonal.
left upper, right upper, left lower, right lower regions. Each region (4) The ‘Right Upper Region’ is in energy balance with external
reflects the location of a group of exchanger matches within the hot utilities. Thus it is expected in this region that only
heat exchanger network, as follows: heaters may appear.
(5) The ‘Left Lower Region’ is similarly in energy balance with
(1) LUR (left upper region): exchangers located in this region external cold utilities. Thus it is only allowed for energy
represent the matches exchanging or integrating heat be- efficient networks to have coolers within this region.
tween hot streams above the pinch with cold streams below (6) Exchanger matches located in the ‘Left Upper Region’ would
the pinch. transfer heat across the pinch. The heat integration of this
(2) RUR (right upper region): exchangers located in this region region is still feasible, although it violates Pinch Analysis
represent the matches exchanging or integrating heat be- principles.
tween hot streams above the pinch with cold streams above (7) Energy integration or heat recovery for maximum energy
the pinch. efficiency will only take place through the exchangers in the
(3) LLR (left lower region): exchangers located in this region two regions ‘Right Upper’ and ‘Left Lower’. Although the
represent the matches exchanging or integrating heat be- situation in many real plants of refining units or preheat
tween hot streams below the pinch with cold streams below trains is different that several exchanger matches are
the pinch. exchanging heat and are located within the region ‘Left Up-
(4) RLR (right lower region): exchangers located in this region per Region’. This last situation of poor design leads to inef-
represent the matches exchanging or integrating heat ficient heat integration or recovery in existing refineries and
preheat trains. Thus considerable amounts of fuels and
cooling utilities are consumed as a result of such a poor
design. Therefore, these exchanger matches need particular
consideration if the energy efficiency is to be maximised.
(8) Coolers located in the ‘Left Upper Region’ violate the Pinch
Analysis principles. For this, the identification of such units is
useful and would help in increasing the energy efficiency of
an existing networks, i.e. heat loads on these coolers can be
exploited for more energy integration. This feature may help
in creating paths for heat integration. A path is linking a hot
process stream and cold process stream with an exchanger
unit at the presence of a relevant cooler and heater for these
streams [2].
(9) Exchanger matches touching the pinch temperatures from
one point signify that these matches start or end heat inte-
gration between streams at the corresponding pinch tem-
peratures, i.e. energy integration takes place at the pinch
temperatures.
It is obvious then that the new graphical representation can help

examine the principles of Pinch Analysis presented in the previous
section. For example, if exchangers exist in the ‘Left Upper Region’,
these exchangers will violate the principles of energy integration
because they would transfer heat across the pinch. It can also
determine whether the external hot and cold utilities are improp-
Fig. 3. A new graphical representation for energy analysis. erly placed within the existing networks.
3.1. A new graphical representation for existing HENs
As shown in Fig. 4, consider a heat exchanger hx, transferring

heat between a hot stream H and a cold stream C with a heat
integration load of Q. The inlet temperature of hot and cold streams
are Tin in
h and Tc respectively, while these streams leave the exchanger
unit at Tout
h for the hot stream and Tcout for the cold stream. The hot
stream has a mass flow rate of mh and a specific heat of Cph. On the
other hand, the cold stream has a mass flow rate of mc and a heat
capacity Cpc. Performing an energy balance on the given exchanger
match results in:

Q ¼ mh Cph Thin Thout ¼ mh Cph DTh (1)

Q ¼ mc Cpc Tcout Tcin ¼ mc Cpc DTc (2)
If the heat capacities of hot and cold streams are considered

constant across the range of temperatures, then Eqs. (1) and (2) can
be written as: Fig. 5. Representation of an exchanger match on the new graphical plot.

Q ¼ CP h Thin Thout ¼ CP h DTh (3)
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u !2 !2
u
u Q Q
Q ¼ CP c Tcout Tcin ¼ CP c DTc (4) L¼t þ (7)
CPh CPc
where, CP h is the mean heat capacity flow for the hot stream, and
CP c is the mean heat capacity flow for the cold stream. Dividing Eq.
u !2 !2
u
(3) by Eq. (4): u 1 1
L¼Q t þ (8)
. . CPh CPc
CP c CP h ¼ DTh DTc (5)
Thus, Eq. (8) tells that the length of the line representing an
If the exchanger match is plotted on the new graphical repre- exchanger match is directly proportional to the heat duty trans-
sentation as a straight line (with an arrow), then the line should ferred across this exchanger (Fig. 5). Although, it is still possible to
start at the point (Tcout, Tin in out
h ) and end at (Tc , Th ); see Fig. 5. The shift all temperatures values for hot and cold streams by a certain
exchanger line is bounded by the temperatures of both the hot parameter in order to obtain an exchanger match whose length is
stream and cold stream. The hot temperatures could be the supply exactly equal to the heat duty transferred Q. This can be easily
and target temperatures of the hot stream, or some of its inter- proved as shown below:
mediate temperatures. Likewise, the cold temperatures could Put both Eqs. (6) and (7) equal:
represent the supply and target temperatures of the cold stream or
its intermediate temperatures. Given the assumption of constant u !2 !2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u
specific heat within the range of temperatures, exchangers are u 1 1
Qt þ ¼ ðDTh Þ2 þ ðDTc Þ2 (9)
represented by straight lines. Still, variable thermal properties can CPh CPc
be taken into account in the plot. The slope of the line representing
exchangers is given by Eq. (5), as the ratio between the heat ca- Then, divide both sides of Eq. (9) by the
pacity flows of cold stream to hot stream or the temperature dif- vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!2 !2
u
u
ference of hot stream divided by that of the cold stream. Thus the term t 1 þ 1 , resulting in:
vertical distance of the line is equivalent to DTh, while the hori- CPh CPc
zontal distance is equivalent to DTc. The length of the straight line L

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
can be calculated as follows: 1
Q¼ ðDTh Þ2 þ ðDTc Þ2 (10)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi J
L ¼ ðDTh Þ2 þ ðDTc Þ2 (6)
where, j is a term related to the heat capacity flows of the hot and
cold streams of the exchanger match and can be calculated by:
!2 !2
1 1
J2 ¼ þ (11)
CPh CPc
Then, Eq. (10) can be written as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

DTh 2 DTc 2
Q¼ þ (12)
J J
Fig. 4. Variables for an existing heat exchanger match.
Eq. (12) is simplified through Eqs. (13)e(15) to be related to new pinching matches whenever heat recovery is maximised and a
temperature values divided by the term j calculated by Eq. (11); Network Pinch takes place. Consequently, the new graphical rep-
these new temperature values are called shifted temperatures. resentation is useful in identifying the pinching matches for an
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi existing network and locating the Network Pinch. Therefore, to
u ! ! increase heat recovery beyond the maximum for the given condi-
u T in T out 2 T out in 2
t h h c Tc
Q¼ þ (13) tion, modifications must be made to overcome the Network Pinch.
J J When the diagonal is taken as Th ¼ Tc as the reference for Network
Pinch, the temperature approach limit assumed is zero. Then again,
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi any positive temperature limit may be used (DTmin). In this latter
u ! !2
u T in T out 2 T out T in case, Th ¼ Tc þ DTmin.
Q¼ t h
h
þ c
c
(14)
J J J J Now to describe an existing HEN with full details, all exchanger
units are plotted through lines on the graphical representation, as
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi described in Fig. 5. Exchangers are represented knowing either
2 2 terminal temperatures, i.e. hot end and cold end (Tc, Th) or one
Q¼ T in*
h Th
out*
þ T out* c T in* c (15)
terminal temperature and a slope (ratio of heat capacity flows).
The values of other variables can then be determined. Exchanger
where, Thin* is the shifted temperature of inlet hot stream to the units between the same cold and hot streams will have the same
exchanger unit and can be calculated by diving the real tempera- slope. This facilitates the execution of the HEN graph because the
ture by the term j, as: exchanger lines can be drawn as parallel. For successive ex-
changers, their representation is done such that one exchanger
Thin follows the other, starting at the intermediate temperature. A
Thin* ¼ (16)
J complete HEN representation is obtained by plotting all the
Similarly, all other shifted temperatures can be calculated from various exchanger units within the network. In the new HEN
similar relationships with real temperatures. graphical representation, a conventional loop of two exchangers
appears as two lines, each for each exchanger; the two lines
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi plotted between the same hot stream temperatures, on the y-axis,
2 2
Q¼ DT *h þ DT *c (17) connecting the same cold stream on the x-axis. On another hand, a
utility-path appears as three lines of exchangers (or more), con-
Eq. (17) indicates that when the exchanger match is represented necting a process stream (or streams), and cold utility and hot
such that the temperatures of inlet and outlet streams within the utility. The hot utility line will be located on the top right region of
exchanger are divided by the term j, then the length of the Fig. 5, whilst the cold utility line will lie within the bottom left
exchanger match in the graph will exactly equal Q. This is mathe- region. Heat is shifted within the loop graphically by fixing the
matically correct since the length of the line is the square root of the exchanger lines' slopes and extending these lines for more tem-
summation of both the squares of temperature differences on the perature difference in the case of added heat and for less tem-
shifted-scale (Eq. (17)). The new graph will be obtained simply by perature difference in the case of removed heat. The extended
dividing each temperature for hot/cold stream by the term j for the exchanger lines will then determine the new intermediate tem-
specific exchanger match (Eq. (11)). Then, the new values are peratures. Note that while shifting heat within loops, the
plotted on the new graph of shifted temperatures. Note that each boundary temperatures for both hot and cold streams remain un-
vertical difference for an exchanger line refers to shifted hot tem- changed. A similar graphical procedure is followed when utility-
perature difference ðDTh Þ. Similarly, the horizontal difference for an paths are exploited to reduce utilities consumption. In this case,
exchanger indicates the shifted temperature difference for the cold the amount of utility heat to be saved is shifted from the utilities'
stream ðDTc Þ. Therefore, the graphical representation shown in exchangers and added to the exchanger. When heat is added or
Fig. 5 can be redrawn by shifting all temperatures of hot and cold removed, slopes of exchangers' lines are fixed, and only the
streams for each exchanger match by the term j for the particular temperatures are changing. The following illustrative example
exchanger streams calculated by Eq. (11). Hence, the heat duties explains the application of the new graphical method on an
across each exchanger match can be read directly from the shifted existing HEN. It highlights the procedure of the new technique,
graph as their numerical values of the each line length according to and points out to some significant observations, findings and the
Eq. (12). This graphical nature is of importance as one can compare advantages of the approach. Cases of paths and loops will also be
between any exchanger matches or modifications then according to referred to.
their exact lengths, i.e. heat duties.
A significant feature observed in Fig. 5 is that the vertical dis- 3.2. An illustrative example e an existing HEN
tance between the straight line and the diagonal gives the tem-
perature driving force profile across each exchanger (DT). The closet An existing heat exchanger network presented in previous work
the exchanger line to the diagonal, the smallest the temperature by Asante and Zhu [7] is considered for the application of the new
driving force the exchanger has. Then if the line is parallel to the graphical representation. The case represents a real network of five
diagonal, this implies a constant temperature driving force along exchanger units, as shown in Fig. 6. The existing network has four
the exchanger match. An exchanger match becomes a limiting for hot streams and one cold stream. Streams data, including supply
heat recovery as the line approaches the diagonal, and hence a and target temperatures, and required heat loads for cooling and
decreased temperature driving force. This feature would help in heating are given in Table 1. The existing network consumes hot
identifying limiting exchanger matches for heat recovery by utility and cold utility at the rate of 160 MW and 50 MW respec-
locating those matches with the smallest vertical distances to the tively. The heat integration is achieved through some three
diagonal, and hence temperature driving forces. exchanger matches between hot and cold process streams as
The maximum heat recovery for an existing network is identi- shown in the grid diagram of Fig. 6. On the other hand, external
fied when exchanger matches tend to touch the diagonal in the utilities are applied for the existing exchanger network via one
graphical representation, where Th ¼ Tc. Thus, these matches are heater and two coolers.
Fig. 6. An existing heat exchanger network (numbers below exchangers refer to heat
duties in MW).
The existing heat exchanger network is described using the new

graphical method, and is demonstrated in Fig. 7. A minimum
temperature approach difference of 10 C is used in the new rep-
resentation. On the same plot, also the cold stream (c1) and the hot
streams (h1, h4) are displayed as dashed-lines (horizontal and
vertical) for some later and further explanations. Cold stream c1 is
displayed by the vertical line at 100 C for the supply temperature, Fig. 7. A new graphical representation for an existing HEN (small arrows refer to
and the vertical line at 190 C for the target temperature. In the streams h1, h4 and c1).
same way, hot stream h1 is represented through two horizontal
lines at 160 C and 120 C for the supply and target temperatures
The existing heat exchanger network shown in Fig. 7 can then
respectively. Hot stream h4 is represented between the supply
easily be evaluated with respect to Pinch Analysis principles and
temperature of 200 C and target temperature of 130 C. For this
energy targets, as follows:
DTmin, the pinch temperatures for the process streams are 150 and
140 C for the hot and cold respectively; these pinch temperatures
(1) The heater E104 is appropriately placed in the region ‘Right
are easily obtained by using any traditional Pinch Analysis method
Upper’.
such as given by Smith [2]. The energy targets for this process at
(2) The two coolers E100 and E106 are partly inappropriately
DTmin ¼ 10 C are 125 MW and 15 MW for hot energy and cold
placed in the region ‘Left Upper’, where energy integration is
energy respectively. It is obvious that the existing network per-
to take place across the pinch.
forms away from the energy targets. External hot utility is
(3) The exchanger matches E103 and E101 are integrating heat
consumed by 28% more than energy target. This means that the real
consistently with the Pinch design method. E103 is inte-
process excessively consumes a fuel oil or natural gas to provide the
grating energy from hot stream above the pinch to a cold
process heating. Therefore, a potential of fuel savings is available at
stream above the pinch. Likewise, E101 is transferring the
the rate of 28%.
heat from hot stream below the pinch to a cold stream below
The new graphical representation describes the existing HEN
with some important features, including:
▪ All exchanger matches are shown, connecting on the cold side

the intermediate temperatures for the cold stream (c1) starting
at the supply temperature of 100 C till reaching the target
temperature of 190 C. Intermediate temperatures for the cold
stream (c1) are displayed as the dotted-vertical-lines.
▪ Cold stream c1 is heated successively from the supply temper-
ature (100 C) through the exchangers E101, E102, E103 and then
in the fired heater E104 till the target temperature (190 C). This
layout agrees with the grid diagram shown in Fig. 6.
▪ The cooler E106 cools the hot stream h4 leaving the exchanger
E103 at 160 C to its target temperature of 130 C.
▪ The cooler E100 is bringing the hot stream h1 from its supply
temperature (160 C) down till its target temperature (120 C).
Table 1
Stream data for a chemical process given by Asante and Zhu [7].
Stream Supply Target Heat Heat capacity

name temperature ( C) temperature ( C) duty (MW) flow (MW/ C)
H1 160 120 20 0.5

H2 200 160 20 0.5
H3 150 120 140 4.67
Fig. 8. A new graphical representation for an existing HEN with shifted temperatures
H4 200 130 70 1
(solid exchangers refer to shifted temperatures; dotted exchangers refer to original
C1 100 190 360 4
temperatures).
the pinch. This means a Network Pinch may probably take away from the region ‘Left Upper’ to the region ‘Right Upper’. A
place at this exchanger match when energy integration is to closer look at Fig. 9 reveals a new feature that can be of significance
be performed. to retrofit modifications. This feature is the overlap area (yellow-
(4) The exchanger E101 is close to the diagonal, hence it is a shaded) between the hot stream h1 and cold stream c1. The overlap
potential limiting exchanger in energy saving projects. between the two streams implies potential heat integration from
(5) The exchanger E102 is integrating heat across the pinch, i.e. the hot stream to the cold stream. Therefore, the cooler E100 can be
exchanging heat from above the pinch into below the pinch. replaced by an exchanger unit to integrate the heat from h1 to c1.
(6) For the current performance, there appears no pinching Certainly, the amount of heat integration between the two streams
matches as none of the exchanger matches (represented by can be quantified providing the temperature differences and heat
lines) touch the diagonal (DTmin ¼ 0 C) or the line corre- capacity flows of both streams. The proposed HEN modifications
sponding to DTmin ¼ 10 C. On the contrary, there is still a are systematically performed as follows (see Fig. 9 as procedure
positive vertical distance between the exchanger lines and basis):
the diagonal. This implies positive temperature driving
forces and hence more potential for energy integration. (1) Determine the point of intersection between the target
temperature of the hot stream h1 (120 C) and the supply
If the temperatures of the existing HEN shown in Fig. 7 are temperature of the cold stream c1 (100 C). Note that since
shifted using Eqs. (11) and (17), then the exchanger lines will the intersection point lies above the diagonal
represent the heat duties transferred across these exchangers (Th ¼Tc þ DTmin), the driving force at the cold end is more
(Fig. 8). The figure shows three exchangers of the HEN with shifted than the DTmin.
temperatures, compared with the exchanges with original tem- (2) To relocate E100: at the intersection point, draw an
peratures as they are in Fig. 7. Exchangers with shifted tempera- exchanger line with the slope given by Eq. (5) (CPc1/CPh1 ¼ 4/
tures are termed E101*, E102* and E103*, and are shown by solid 0.5) to intersect with the supply temperature of the stream
lines. Now, the lengths of the new exchanger lines are indicators to h1 (160 C) at the new intermediate temperature for the cold
the heat duties of each exchanger. The longer the line is, the more stream c1. The new straight line signifies the new exchanger
heat the exchanger transfers. E100. The intermediate temperature is found to be 105 C and
As shown above, the new graphical methods helps examine the it represents the new temperature of the cold stream leaving
Pinch Analysis principles in an existing HEN. Furthermore, the E100 before entering the heater E101. The length of E100
improper placement of E100 and E106, and together with the heat corresponds to 20 MW that can be recovered from h1 to c1.
integration across the pinch done through E102 explain why the (3) The exchanger line representing E101 is plotted between the
existing network performs poorly with respect to energy targets hot stream h3 and cold stream c1. At the point with the co-
(28% excess in fuel or utility consumption). This suggests that these ordinates of (105 C, 120 C), draw the line for E101 with the
exchanger matches need to be of concern for any energy savings' slope CPc1/CPh3 ¼ 4/4.67.
projects. (4) Extend the exchanger line to meet the diagonal cold pinch
Guided by the conclusions drawn from the graphical method of temperature (140 C). This will ensure recovering all the
Fig. 7, a modified HEN is proposed. The modifications proposed to available heat of stream h3 below the pinch. It appears that
the existing HEN are: modification #1; relocating the cooler E100 to the exchanger line also intersects with the diagonal at the
recover more heat from the hot stream h1 through heating the cold pinch point (140 C, 150 C), where Th ¼ Tc þ DTmin. At the
stream c1 rather than using cooling water, and modification #2; intersection point, the driving force at the hot end of the
shifting the exchanger integrating heat across the pinch (E102) exchanger E101 is exactly 10 C (i.e. DTmin). This entails that
Fig. 9. A graphical procedure for potential modifications for HEN revamping. Fig. 10. A new graphical representation for a modified HEN.
this exchanger unit is pinched and thus causing a Network be still obtained if the same procedure stated for E100 is followed
Pinch. with respect to the cooler E106 which is cooling the hot stream h4.
(5) To shift E102: draw a line starting at the point (140 C, 160 C) In other words, the overlap area between h4 and c1 is identified and
with the slope of CPc1/CPh2 ¼ 4/0.5, and extend the line within then part of the heat load of the exchanger is relocated into more
the region of heat integration above the pinch. The line will heat recovery between the two streams. This means the integrated
intersect with the supply temperature of h2 at 200 C. The heat load of the exchanger E103 will increase and that of the cooler
length of this exchanger line will correspond to the heat load E106 will decrease. This is similar to transferring energy through a
of the stream h2 (20 MW). As shown in Fig. 9, the new path between cold and hot utilities. Of course, there is a maximum
location of the exchanger E102 now becomes completely amount of heat that can be recovered additionally from the hot
within the area of heat integration above the pinch. There- stream h4 to the cold stream c1 within the unit E103. This proposed
fore, this exchanger after modification does not violate the modification is executed with the maximum heat recovery possible
Pinch Analysis principles as it did in the existing HEN. graphically by (refer to Fig. 12):
(6) The intersection of E102 with the horizontal line of 200 C
results in the new temperature of c1 to enter the exchanger (1) Draw a vertical line at the intermediate temperature of the
E103 (Tint). Then, at the point (Tint, 160 C) the exchanger line hot end of E102 (145 C).
representing E103 can be plotted with the slope CPc1/ (2) Determine the point of intersection of the vertical line with
CPh4 ¼ 4/1 between the horizontal temperatures 160 C and the diagonal (Th ¼ Tc þ DTmin; DTmin ¼ 10 C). The intersec-
200 C which are the intermediate and supply temperatures tion with the diagonal implies that the temperature driving
of h4, respectively. A new intermediate temperature for the force of the exchanger line is equal to the minimum tem-
stream c1 is obtained which corresponds to the temperature perature approach difference of 10 C.
leaving E103, i.e. entering heater E104. (3) At the point of intersection, draw an exchanger line with the
(7) The heat load on the heater E104 to heat the cold stream c1 slope as given in Eq. (5) (CPc/CPh ¼ 4/1) to intersect with the
till its target temperature (190 C) can then be graphically supply temperature of the stream h4 (200 C) at the new
determined. intermediate temperature for the cold stream c1. The new
straight line represents the new exchanger E103 with its new
Fig. 10 implements the proposed above modifications. As a heat load. The intermediate temperature is found to be
result of these modifications, more heat is integrated within the 156.3 C and it gives the new temperature of the cold stream
process (20 MW), and thus the heat load on the heater E104 de- leaving E103 before entering the heater E104. The new length
creases to 140 MW, compared with 160 MW for the existing of E103 determines the value of the new heat load that can be
network. Also, the temperature before the heater E104 increases by recovered from h4 which is found to be 45 MW. Thus, the
5 C over the base existing HEN. At the same time, the consumption heat load on the cooler E106 is lowered to 25 MW with an
of cooling water is decreased by a 20 MW equivalence. This cor- associated saving in cooling water. Also, the heat duty of the
responds to a saving of 12.5% in the fuel consumption (or hot utility) heater E104 is decreased as the cold stream c1 will be heated
and 40% for cooling water with the existing performance. Also, the only from 156.3 C till the target temperature of 190 C
temperature distribution across the modified HEN changes ac- (135 MW).
cording to the results of Fig. 10. It is obvious that the new location of (4) The exchanger E103 will be Pinched Match since its tem-
E100 pushed all exchanger units to the right, i.e. more heat inte- perature driving force at cold end is equal to DTmin of 10 C.
gration. This explains why E102 is now properly placed within the Therefore, a Network Pinch takes place at the cold end of this
region of heat integration above the pinch, and no longer violates exchanger unit.
the Pinch Analysis principles (Fig. 10). The modifications proposed
by Fig. 10 can be extracted to result in the grid diagram of Fig. 11.
The new temperature distribution of the cold stream c1 becomes
from supply temperature of 100 till 105 C in E100, 105 till 140 C in
E101, 140 till 145 C in E102, 145 till 155 C in E103, and finally from
155 till the target temperature of 190 C in the heater E104.
It is worth noting that the relative exchanger length of the
heater E104 in Figs. 7 and 10 is exactly equal to 12.5% which is the
percentage savings in energy obtained. Further energy savings can
Fig. 11. A HEN grid diagram with temperature distribution for a modified network
proposal (E101 is pinched). Fig. 12. A pinched HEN for DTmin ¼ 10 C.
crude oil feed, and includes atmospheric crude distillation tower

and crude vacuum unit. The existing refinery is taken from the
available library data of the simulator ‘Aspen Hysys’ [31]. The
existing plant configuration and basic process data are shown in
Fig. 14. Also, Table 2 provides relevant data for process hot streams
and cold streams, including supply temperatures, target tempera-
tures and target heat loads. Temperatures for inlet streams and
outlet streams for each exchanger unit within the existing HEN are
displayed in Table 3; it is observed that the minimum temperature
driving force of all exchanger units is approximately 20 C (i.e.
DTmin ¼ 20 C). The layout of the existing network, focussing on the
preheat train of crude oil feed stream, and including the desalter is
Fig. 13. A HEN grid diagram with temperature distribution for a modified network
depicted in Fig. 15. The diagram shows details of all process hot
proposal (E101 and E103 are pinched). streams exchanging heat with the main process cold stream that is
the crude oil feed. As shown, the temperature of crude oil feed just
prior to the fired heater is shown to be 247 C. The temperature of
The detailed procedure and modified HEN are displayed in the crude oil leaving the fired heater towards the atmospheric
Fig. 12. The heat load on the heater E104 in Fig. 12 is found to reach tower is 343 C.
135 MW with a saving of 15.6% compared with the existing HEN. The existing refinery is producing various crude oil derived
No more energy savings can be achieved for the existing HEN via fuels/fractions, such as:
additional exchanger areas since the modified network is pinched
at E103 and E101 as clearly shown in Fig. 12. Pinching matches can ▪ ovhd, TopV (overhead and top vapours).
easily be identified as E103 and E101 since they touch the line with ▪ Heavy naphtha.
10 C, and thus Network Pinch takes place at both exchangers. This ▪ Kerosene.
means that both exchangers have a minimum temperature ▪ Diesel.
driving force of 10 C. This condition limits the heat recovery for ▪ AGO (atmospheric gas oil).
the existing HEN and defines its maximum energy recovery level. ▪ LVGO, HVGO (light and heavy vacuum gas oils).
Any future projects on energy savings must consider the identified ▪ Slob wax.
pinching matches so that the topology of the network is to be The energetic performance of the existing refinery is summar-
changed [7]. The HEN modifications are shown in the grid diagram ised as follows:
of Fig. 13. It must be noted that the proposed modifications can be ▪ Hot energy requirement for ADU (Distil Heater): 41.6 MW.
also made through simulation by using Aspen Energy Analyser ▪ Hot energy requirement for VDU (Vacuum Heater): 7.2 MW.
[32]. ▪ Trim heating for ADU: 14.4 MW.
The above illustrative example revealed the significance of the ▪ Trim heating for VDU: 16.9 MW.
new graphical representation as it can be used to evaluate the ▪ Total heating requirement: 80.1 MW.
performance of an existing design with respect to the Pinch Anal- ▪ Overall cooling water consumption: 35.2 MW.
ysis principles. In other words, the new graphical representation ▪ Heavy fuel oil is used in fired heaters.
can help determine existing matches transferring energy across the ▪ Cooling water: 20 Ce45 C.
pinch, besides the location of Network Pinch and hence the ▪ Hot flue gases: 100 Ce400 C.
pinching matches. As a result, modifying existing networks guided
by the new representation can lead to an improved heat recovery The objective of the case study is, by using the new graphical
with the networks, leading to substantial savings in fuel method:
consumptions.
(1) Analysing the energy performance of the existing refinery
4. A case study plant.
(2) Improving the energy efficiency of the existing refinery
A real refining plant for producing crude oil fractions and fuel plant.
products is considered to apply the new graphical method. The (3) Minimising the atmospheric emissions of CO2 from the
existing plant processes a capacity of 85 thousand barrels per day of existing refinery unit.
Fig. 14. A real refining plant for crude oil fractionation.
Table 2 64.83 MW as target, resulting in an excess consumption of fuel oil

Stream data for real refinery plant. by approximately 24% (i.e. 15.3 MW extra). This signifies that the
Stream name Supply Target Heat existing preheat train was not designed in an energy efficient
temperature ( C) temperature ( C) duty (MW) manner.
14_To_15 143.8 136.2 0.37 To analyse the energy performance of the existing refinery, the
59_To_Diesel 275.6 214.6 3.04 new graphical method is applied. The existing preheat train (HEN)
65_To_66A 336.7 288.4 4.51 of the refinery plant is represented by the new graphical method
47_To_48 222.2 166.9 10.52
(Fig. 16) for a DTmin of 20 C. Note that the hot pinch and cold pinch
56_To_57 288.4 229.1 10.12
Raw Crude Oil 37.8 145.0 31.08 temperatures determined using Aspen Energy Analyser [32] at this
from Storage_To_10 condition are 114.1 C and 94.1 C respectively. As shown, all
VacRes_To_91 357.2 339.4 5.89 exchanger units are shown by straight lines (heaters: in red colour,
74_To_Feed to 363.8 397.2 7.23
coolers: in blue colours). Several important features can be
Vacuum Tower
66_To_67 288.4 242.9 4.00
observed:
50_To_Kerosene 203.6 143.9 3.04
82_To_84 281.1 191.5 27.18 (1) Exchanger matches conforming to the Pinch Analysis prin-
300_To_16 115.6 124.4 0.37 ciples, such as:
12_To_22A 142.6 343.3 83.14
▪ The exchangers from E-6 to E-12 are integrating heat
LVGO P/A_Draw_To_LVGO 152.5 56.1 9.81
P/A_Return@COL2 above the pinch.
Ovhd_To_32 114.1 38.9 29.71 ▪ The exchanger E-1A is integrating energy below the pinch.
TrimDuty@ADU 351.6 368.7 14.39 ▪ The heaters including Distil Heater and Vacuum Heater are
TrimHeat@VDU 325.7 357.3 16.85 heating process cold streams above the pinch.
▪ The cooler E-2A is cooling a process hot stream below the
pinch.
4.1. Results and discussion (2) Exchanger matches violating the Pinch Analysis principles,
such as:
Thermal efficiency of the fired heater is approximately 90% [33]. ▪ The exchangers E-3, E-4 and E-5 are integrating heat across
If the net calorific value of the fuel used is 39.8 MJ/kg [33], then the the pinch.
flow rate of fuel oil necessary for process heating is (80.1 MW)/ ▪ The cooler lvgo P/A is cooling process hot stream partly
(0.9 39.8 kJ/kg). That is 2.236 kg/s and corresponds to 70,520 above the pinch.
metric tonnes per year. The atmospheric emissions of CO2 due to (3) Exchanger E-7 touches the diagonal line with a minimum
fuel combustion can be calculated as proposed by Gadalla et al. [33], temperature approach difference of 20 C. This implies that
2,796.7 metric tonnes per year of CO2 per each megawatt of process the exchanger E-7 is pinching match and Network Pinch
heating provided by the heavy fuel oil. Therefore, the total emis- takes place.
sions of CO2 from the existing refinery due to the combustion of (4) The exchanger units on the main cold stream (crude oil feed)
heavy fuel oil are equivalent to 2,796.7 80.1 or 224,015 metric are shown as a sequence of exchanger lines, one after
tonnes per year. another. Crude oil stream starts its heating integration pro-
The energy targets (minimum energy) obtained for the existing cess at E-1A, then through a series of exchanger units from E-
refinery unit by using Aspen Energy Analyser [32] are determined, 3 and E-4 (in parallel), to E-5, E-6, E-8, E-9, E-10, E-11, E-12,
and summarised in Table 4 for two minimum temperature and finally through the fired heater the Distil Heater. Inter-
approach differences, 10 C and 20 C. The table shows that the mediate temperatures are also clear between each successive
minimum hot energy consumption for the given process is exchangers. These exchangers are shown as thicker lines.
64.83 MW for 20 C. This shows that the existing refinery plant This layout conforms to the HEN diagram shown in Fig. 15.
performs poor as far as the energy performance is concerned; the Note that E-4 lies above E-3 since they are connected in
actual consumption is 80.1 MW of hot utility compared with parallel with the cold crude oil stream.
Each of the above violations of the Pinch Analysis principles in

Table 3 the existing HEN contributes proportionally to the poor energetic
Heat exchangers data for real refinery plant. performance. It also explains why the network consumes fuel oil for
heating more than necessary by at least 24%. A range of potential
Heat exchanger Tin
c Tout
c Tin
h ( C) Tout
h Load Area
( C) ( C) ( C) (MW) (m2) HEN modifications can be identified out of the above reported
observations of Fig. 16 to improve HEN energy performance and
E-11@ADU 221.6 232.8 336.7 288.4 4.509 126.5
E-12@ ADU 232.8 246.5 357.2 339.4 5.888 146.1 reduce fuel and cooling water consumptions. The potential modi-
E-8A/B@ADU 142.6 173.0 234.5 191.5 12.645 651.6 fications that can be realised within the existing HEN are:
Vacuum Heater@ADU 363.8 397.2 1000.0 400.0 7.233 126.0
TrimHeat@VDU_Exchanger 325.7 357.3 1000.0 400.0 16.847 1597.4 ▪ Modification #1: relocating some or all the exchangers that
E-2A-F@ADU 20.0 45.0 101.3 38.9 25.354 626.8
E-10A/B@ADU 183.5 221.6 281.1 234.5 14.537 759.3
integrate heat across the pinch, i.e. E-3, E-4 and E-5.
LVGO P/A_Cooler@VDU 20.0 45.0 152.5 56.1 9.808 218.1 ▪ Modification #2: replacing part of the heat load of the cooler E-
E–1A/B@ADU 37.8 53.8 114.1 101.3 4.356 327.7 2A by more heat recovery to the crude oil feed stream below the
E-3@ADU 53.8 76.1 203.6 143.9 3.038 98.7 pinch.
E-5@ADU 76.1 112.0 222.2 166.9 10.525 320.2
▪ Modification #3: adding a new exchanger unit to replace the
E-6@ADU 112.0 145.0 288.4 229.1 10.120 209.5
E-9@ADU 173.0 183.5 288.4 242.9 4.000 108.0 cooling duty of the exchanger lvgo P/A on the hot stream ‘LVGO
E-4@ADU 53.8 76.1 275.6 214.6 3.038 58.6 P/A_Draw, 9.81 MW’ and exploit this heat for preheating the
E-7@ADU 115.6 124.4 143.8 136.2 0.370 10.4 crude oil.
Distil Heater@ADU 246.5 343.3 1000.0 400.0 41.566 916.6 ▪ Modification #4: changing the topology of the HEN to remove
TrimDuty@ADU_Exchanger 351.6 368.7 1000.0 400.0 14.390 661.1
the Network Pinch occurring at the exchanger E-7.
Fig. 15. An existing HEN, including side for crude oil feed from well to distillation columns.
In line with the illustrative example presented earlier, per- is drawn from the pinch points (94.1 C, 114.1 C) to intersect
forming the above modifications will expectedly improve the per- with the supply temperature of the crude oil stream (37.8 C).
formance of the existing HEN and reduce the utility consumptions This exchanger line represents the new location of E-1A (E-
substantially. The potential of energy reduction is within the range 1A0 ). The exchanger will satisfy the heating requirement of
of 24%. the cold stream below the pinch, i.e. 16.3 MW. At the same
time, the intersection leads to the intermediate temperature
4.1.1. Implementing HEN modifications (72.8 C) for the hot stream (Ovhd_To_32).
The identified recommendations can be achieved on the exist- (3) At the intersection point (45 C, 72.8 C) the new cooler line
ing network separately or at the same time. First, Modification #2 is (E-2A0 ) is plotted between the hot and cold end to satisfy the
implemented to the existing HEN, following the next procedure cooling requirement of the hot stream, i.e. 13.4 MW (before
(Fig. 17):
(1) The shaded area in Fig. 17 shows an overlap below the pinch
between the crude oil stream and the hot stream
‘Ovhd_To_32’. This overlap indicates the potential heat re-
covery between these streams and also quantifies this po-
tential. The potential amount of heat recovery is determined
by plotting an exchanger line starting at the pinch points
whose slope equals 1; this line intersects with the shaded
area, resulting in the maximum heat recovery of 22.3 MW.
(2) Below the pinch, the crude oil stream only requires a heat of
16.3 MW, implying that the hot stream possesses a heat
surplus. Therefore, an exchanger line with the slope CPc/CPh
Table 4
Energy targets for existing refinery for different DTmin.
DTmin ¼ 10 C DTmin ¼ 20 C
Hot energy (fuel) (MW) 61.77 64.83
Cold energy (water) (MW) 16.89 19.96
Hot pinch temperature ( C) 114.1 114.1
Cold pinch temperature ( C) 104.1 94.1
Fig. 16. A new graphical representation for HEN of case study.
hence large heat transfer area. The same implication of tempera-

ture driving force applies if E-5 is relocated before E-8. In conclu-
sion, relocate E-5 before E-8, and then relocate E-4 after E-8 using
the same procedure as above. Afterwards, all remaining exchangers
E-9, E-10, E-11 and E-12 are drawn on the same figure, knowing the
slopes of exchanger lines and intermediate temperatures. The
presence of the diagonal is very significant to ensure the minimum
temperature driving force. The closer the exchanger line to the
diagonal line, the smaller the temperature driving force
(limitation).
The new configuration of the modified HEN is demonstrated in
Fig. 18. A HEN diagram can be extracted from the data of Fig. 18;
such a diagram is shown in Fig. 19 for the complete HEN retrofit
modifications. Fig. 19 is still useful for making a good analogy with
the new graphical representation for the modified HEN. As shown
in both figures (Figs. 18 and 19), the temperature of crude oil after
HEN modifications becomes 275.7 C, with an increase in temper-
ature of 28.4 C. The corresponding heat load on the furnace ‘Distil
Heater@ADU’ with modifications is decreased from 41.6 MW for the
existing HEN to 28 MW. The total heating requirement for the
Fig. 17. A new graphical procedure for HEN modifications. modified process is then 66.5 MW, compared with 80.1 MW for the
base case. The energy savings accomplished are of approximately
17%. This implies substantial savings in the flow rate of fuel and
25.4 MW). The flow rate of cooling water is then determined thus the operating costs. The corresponding cooling water con-
from the slope of the exchanger line E-2A0 (CPcw ¼ 0.54 MW/ sumption is equivalent to 21.6 MW (load of E-2A and lvgo P/A),

C). compared with 35.2 MW with savings of 39%. Data extracted from
the graphical representation for the modified HEN are summarised
As a result of the above modification (#2), the heat recovery on in Table 5, including terminal temperatures, duties and additional
E-1A0 is increased from 4.4 MW to 16.3 MW. Also the heat load on heat transfer areas for each exchanger unit. Data show that a
the cooler E-2A0 is decreased from 25.4 MW to 13.4 MW, with a number of exchanger units after modifications require additional
corresponding saving in cooling water demand by 47.2%. A simple heat transfer areas. Additional areas can possibly be cut down by
heat balance on the crude oil above the pinch tells that the amount using the exchangers' loops to shift heat between exchangers
of heat required is 14.75 MW. When E-3 is relocated (Modification within the same loops or using the Energy Analyser simulator [32].
#1) to be above the pinch, then the 3 MW of this exchanger will be The fuel oil consumption for the modified refinery is decreased
used in heat integration. Also, E-6 with its heat load of 10.12 MW is as a result of the energy savings. It is calculated as previously shown
already above the pinch. Then heat balance will require an extra as (66.5 MW)/(0.9 39.8 kJ/kg) or 1.857 kg/s (58,561 t/y). On the
1.59 MW to be provided in order to heat the crude oil feed up to the other hand, the atmospheric emissions of CO2 are also reduced to
required temperature before the desalter (145 C). Therefore, a new 185,980 t/y by savings of also 17%. The overall savings and im-
exchanger unit (Modification #3, E-20) is installed to the existing provements of the new modified HEN with respect to the base case
HEN to recover 1.59 MW from the pump-around hot stream ‘LVGO are summarised in Table 6. The table reveals that significant savings
P/A_Draw’. The remaining cooling demand of the ‘LVGO P/A_Draw’ in energy demands, fuel oil consumptions, cooling water flows and
is provided by cooling water in the exchanger hvgo P/A. Following
the same procedure explained for Modification #2, the units E-20
and hvgo P/A are represented on Fig. 17. From the figure, the tem-
perature of crude oil leaving E-20 is determined to be 99.6 C; also
the cooling water demand is estimated (CPcw ¼ 0.329 MW/ C).
Then exchanger E-3 is shifted to the above-the-pinch zone by
plotting an exchanger line with the slope CPc/CPh starting at the exit
temperature of E-20 (new unit). The shifted exchanger is bounded
between the supply and target temperatures of the hot stream
‘50_To_Kerosene’. From the graph, the outlet (intermediate) tem-
perature of crude oil leaving E-3 is thus determined (110.1 C). After
leaving E-3, the crude oil is heated in E-6 to complete the heat
balance and thus reach the required temperature before the
desalter (145 C). In a similar way, E-6 is represented on the same
graph of Fig. 17.
To perform the rest of HEN modifications, E-4 and E-5 are to be
relocated above the pinch as proposed previously. The possible
locations for these exchangers are either before E-8 or after E-8;
also E-4 is either relocated first, or E-5 is moved first. By looking at
Fig 17 for the existing HEN, one can observe that the hot stream of
E-4 is hotter than that of E-5. For this, E-5 is to be relocated first as it
will have large temperature driving force. If E-4 is relocated first
before E-5, the exchanger E-5 in its new location will be closer to Fig. 18. A new graphical representation for modified HEN of case study (E-10 shows
the diagonal. This implies very small temperature driving force and 16 C temperature difference).
Fig. 19. A complete modified HEN for energy savings (relocating exchangers E-3, E-4 and E-5, new exchanger E-20).
atmospheric emissions are achieved by the new modified refinery As proposed in the illustrative example, the graphical technique is
HEN. It is clearly seen that the proposed modifications by the new employed to propose HEN modifications to increase the tempera-
graphical representation improve the heat integration of the ture difference of E-10. The exchanger loop concept explained
existing refinery. In overall, Fig. 18 points out that no exchanger earlier is used for this objective. Fig. 20 depicts the modification
units are located within the area of heat integration across the proposed graphically to increase the temperature driving force of E-
pinch. One may note that Modification #4, which is related to to- 10. The modifications are discussed below graphically:
pology change, is not considered at this stage as it is outside the
scope of the current research work. As observed in Fig. 18, all ex- (1) The exchanger line of E-10 is shifted horizontally to the right
changers within the modified HEN lie above (or just touching) the till it touches the diagonal, i.e. DTmin ¼ 20 C.
diagonal of DTmin ¼ 20 C, except the exchanger unit E-10 which (2) Since E-9 is the adjacent exchanger unit to E-10, then the
shows a lower temperature difference than the minimum (z16 C). exchanger is to be shifted as well to the left side in order to
Table 5
Heat exchangers data for modified HEN-based refinery plant.

Heat exchanger Tin
c ( C) Tout
c ( C) Tin
h ( C) Tout
h ( C) Load (MW) Existing area (m2) Additional areaa (m2)
E-11@ADU 250.6 261.5 336.7 288.4 4.509 127 69

E-12@ ADU 261.5 275.7 357.2 339.4 5.888 146 53
E-8A/B@ADU 167.9 194 227.0 191.5 10.77 652 436
Vacuum Heater@ADU 363.8 397.2 1000.0 400.0 7.233 126
TrimHeat@VDU Exchanger 325.7 357.3 1000.0 400.0 16.847 1597
E-2A-F@ADU 20.0 45.0 72.8 38.9 13.4 627
E-10A/B@ADU 210.9 250.6 281.1 227.0 16.4 759 1336
LVGO P/A_Cooler@VDU 20.0 45.0 136.9 56.1 9.808 218 21
Ed1A/B@ADU 37.8 94.1 114.1 72.8 16.3 328 2503
E-3@ADU 99.6 110.1 203.6 143.9 3.038 99 63
E-5@ADU 142.6 168.0 222.2 166.9 10.525 320 540
E-6@ADU 110.1 145.0 288.4 229.1 10.120 210
E-9@ADU 201.3 211.0 288.4 242.9 4.000 108 53
E-4@ADU 194 201.3 275.6 214.6 3.038 59 193
E-7@ADU 115.6 124.4 143.8 136.2 0.370 10
Distil Heater@ADU 275.7 343.3 1000.0 400.0 28.0 917
TrimDuty@ADU_Exchanger 351.6 368.7 1000.0 400.0 14.390 661
a
Assuming constant overall heat transfer coefficients.
Table 6 certainly imposes some capital investment for the additional areas
Savings for modified HEN. required by heat re-distribution. As illustrated by Fig. 20, note that
Base case Modified case Improvements there are no major changes in the new locations of the exchangers
Hot energy (fuel) (MW) 80.1 66.5 17%
E-12 and Distil Heater.
Cold energy (water) (MW) 35.2 21.3 39% In overall, yet the cooler lvgo P/A is cooling one hot process
Crude oil feed temperature 247 275.7 þ28.7 C stream above the pinch (LVGO P/A_Draw); see Fig. 18. The heat of

before fired heater ( C) this process stream can be exploited for further potential heat re-
Heat integration Yes No
covery. Unfortunately, the part of the heat above the pinch cannot
across the pinch (3 exchangers)
Fuel oil consumption (t/y) 70,520 58,561 11,959 t/y be used to heat process streams. The reason is that no overlap area
CO2 emissions (t/y) 224,015 185,980 38,035 t/y between this hot stream and the other cold process streams away of
the diagonal, i.e. very small temperature driving force. It is impor-
tant to note that the above solution is one of many other possible
complete the energy balance. Thus E-9 can be shifted hori- options. This means that other modifications-based solution
zontally to the left for the same horizontal displacement; this is possible. For instance, the heat content of the stream ‘LVGO
implies shifting all exchangers units accordingly, including E- P/A_Draw’ can be integrated with the crude oil just above the pinch.
4, E-8, E-5 just to the left side of E-9. When the shift of heat Hence the new exchanger E-20 will integrate larger heat load
reaches E-5, the exchanger will then require further cooling (3.72 MW instead of 1.59 MW) and this implies shifting all the
or integration to adjust the energy balance of the hot stream. remaining exchangers to the right side of the pinch point and
As result, this requires extra exchanger unit to integrate the accordingly a different modified HEN scheme is achieved.
excess heat or for cooling purposes. This solution is of course
feasible, although it is expensive. Another option is available 5. Concluding remarks
since E-9 and E-11 form a loop (see Fig. 20). Then instead of
shifting all exchangers starting from E-9, the extra heat load In this work, a new graphical method has been developed to
of E-9 will be shifted through the loop to the exchanger E-11.
describe an existing heat exchanger network, including details of
(3) Then, the cold temperature of the new location of E-10 will exchanger matches, heaters and coolers. Details of exchangers are
intersect with E-9, determining the excess heat to be
plotted as temperatures of hot streams versus temperatures of cold
removed or shifted through the loop. streams. The new graphs can analyse and evaluate the performance
(4) The excess heat of E-9 is shifted to the exchanger E-11 by
of the existing HENs or preheat trains with respect to Pinch Anal-
drawing the line representing E-11 starting at the interme- ysis principles. Existing HEN energy inefficiencies are identified as a
diate temperatures of E-9 and E-10 and with the same slope
result of the performance analysis. Thus, the potential energy re-
of E-11 as before modifications. The exchanger line is covery is estimated through evaluating these inefficiencies quan-
extended towards the hot end to meet the supply tempera-
titatively. The new graphical method is capable of proposing
ture of its hot stream (336.7 C), determining the interme- potential medications for better energy integration. The method
diate of the cold stream at the hot end.
also provides a systematic procedure, graphical-based by which
(5) Similarly, E-12 is shifted by drawing the line starting at the
potential modifications can be implemented. Various modification
intermediate temperature and with the same slope as before
alternatives are also reached as a result of the new approach.
modification. The Distil Heater is also shifted accordingly.
Throughout the energy analysis and modifications implementation,
the presence of the diagonal is key in visualising the temperature
The above modification resembles in physical situation shifting
driving force for existing exchangers. A case study of a CDU
some heat load from E-9, moving it into E-11. The modification including atmospheric and vacuum units has shown significant
results in regard with energy savings and emissions reduction.
Typical case study results showed that 17% in energy savings and
39% in cooling water demands are obtained by HEN modifications.
The principal strengths of the new graphical method are that the
method provides an interactive visualisation of the HEN analysis,
and may replace the use of commercial simulators for energy
analysis (e.g. Aspen Energy Analyser). Further, the new graphical
technique helps better understand the problem of energy analysis
and energy integration practices. It also allows shifting heat
graphically between exchanger loops and reducing utilities' heat
through the use of utility-paths.
References
[1] Gadalla MA, Jobson M, Simth R. Optimization of heat-integrated refinery

distillation systems. Trans IChemE Part A 2003;81:147e52.
[2] Smith R. Chemical process design and integration. 1st ed. West Sussex, En-
gland: John Whiley & Sons Ltd; 2005.
[3] El-Halwagi M. Sustainable design through process integration. 1st ed. Oxford,
UK: Elsevier; 2012.
[4] Kayode Coker A. Ludwig's applied process design for chemical and petro-
chemical plants. 4th ed., vol. 2. Oxford, UK: Elsevier; 2010.
[5] Gadalla M, Jobson M, Smith R. Increase capacity and decrease energy for
existing refinery distillation columns. Chem Eng Prog 2003:44e50.
Fig. 20. A new graphical representation for modified HEN of case study (exchangers [6] Gadalla M, Olujic Z, Sun L, Rijke A de, Jansens PJ. Pinch analysis-based
with dotted-lines represent new locations; new location of E-10 shows 20 C tem- approach to conceptual design of internally heat-integrated distillation col-
perature difference). umns. Chem Eng Res Des (ChERD) 2005;83(A8):987e93.
[7] Asante NDK, Zhu XX. An automated and interactive approach for heat [29] Natural resources Canada. Pinch analysis: for the efficient use of energy, water
exchanger network retrofit. Trans IChemE 1997;75:349. & hydrogen. CANMET; 2003, ISBN 0-662-34964-4. Catalogue # M39-96/
[8] Zhu XX, Asante NDK. Diagnosis and optimization approach for heat exchanger 2003E.
network retrofit. AIChE J 1999;45:1488. [30] Linnhoff B, Flower JR. Synthesis of heat exchanger networks: I. Systematic
[9] Smith R, Jobson M, Chen L. Recent development in the retrofit of heat generation of energy optimal networks. AIChE J 1978;24(4):633e42.
exchanger networks. Appl Therm Eng 2010;30:2281e9. [31] Aspen HYSYS v7.3. USA: Aspen Technology, Inc.; 2011. university program.
[10] Bakhtiari B, Bedard S. Retrofitting heat exchanger networks using a modified [32] Aspen energy analyser v7.3. USA: Aspen Technology, Inc.; 2011. university
network pinch approach. Appl Therm Eng 2013;51:973e9. program.
[11] Zhang N, Smith R, Bulatov I, Klemes JJ. Sustaining high energy efficiency in [33] Gadalla M, Olujic Z, Jobson M, Smith R. Estimation and reduction of CO2
existing processes with advanced process integration technology. Appl Energy emissions from crude oil distillation units. Energy 2006;31:2062e72.
2013;101:26e32. http://dx.doi.org/10.1016/j.apenergy.2012.02.037.
[12] Ochoa-Estopier LM, Jobson M, Smith R. The use of reduced models for design
and optimisation of heat-integrated crude oil distillation systems. Energy
Nomenclature
2014. http://dx.doi.org/10.1016/j.energy.2014.06.043.
[13] Jiang N, Shelley JD, Doyle S, Smith R. Heat exchanger network retrofit with a Cpc : specific heat of cold stream (kJ/kg$ C)
fixed network structure. Appl Energy 2014;127:25e33. Cph : specific heat of hot stream (kJ/kg$ C)
[14] Wang Y, Pan M, Bulatov I, Smith R, Kim JK. Application of intensified heat CPc : heat capacity flow for cold stream (kJ/s$ C)
transfer for the retrofit of heat exchanger network. Appl Energy 2012;89: CP c : mean heat capacity flow for cold stream (kJ/s$ C)
45e59. http://dx.doi.org/10.1016/j.apenergy.2011.03.019. CPh : heat capacity flow for hot stream (kJ/s$ C)
[15] Wang Y. Heat exchanger network retrofit through heat transfer enhancement. CP h : mean heat capacity flow for hot stream (kJ/s$ C)
PhD Thesis. Manchester; UK: The University of Manchester; 2012. L: length of exchanger line (m)
[16] Gu W, Huang Y, Wang K, Zhang B, Chen Q, Hui CW. Comparative analysis and mc : mass flow rate of cold stream (kg/s)
evaluation of three crude oil vacuum distillation processes for process se- mh : mass flow rate of hot stream (kg/s)
lection. Energy 2014. http://dx.doi.org/10.1016/j.energy.2014.08.053. P/A: pump-around
[17] Pan M, Smith R, Bulatov I. A novel optimization approach of improving energy Q: heat duty or flow (kW)
recovery in retrofitting heat exchanger network with exchanger details. En- Tc: temperature of process cold stream ( C)
ergy 2013;57:188e200. http://dx.doi.org/10.1016/j.energy.2012.10.056. Th: temperature of process hot stream ( C)
[18] Pejpichestakul W, Siemanond K. Retrofit of refinery heat exchanger network Thin : temperature of inlet hot stream to exchangers ( C)
under different kinds of crude oil by pinch design method using mathematical Thin* : shifted temperature of inlet hot stream to exchangers ( C)
programming. Chem Eng Trans 2013;32:1411e6. Thout : temperature of outlet hot stream from exchangers ( C)
[19] Ataei A, Tahouni N, Haji Seyedi SM, Hashemian SM, Yoo CK, Panjeshahi MH. Thout* : shifted temperature of outlet hot stream from exchangers ( C)
A novel approach to hot oil system design for energy conservation. Appl Tcin : temperature of inlet cold stream to exchangers ( C)
Therm Eng 2014;66:423e34. Tcin* : shifted temperature of inlet cold stream to exchangers ( C)
[20] Wang Y, Feng F, Chu KH. Trade-off between energy and distance related costs Tcout : temperature of outlet cold stream from exchangers ( C)
for different connection patterns in heat integration across plants. Appl Therm Tcout* : shifted temperature of outlet cold stream from exchangers ( C)
Eng 2014;70:857e66. Thin : temperature of inlet hot stream to exchangers ( C)
[21] Rezaei H, Zarei MM, Fazlollahi F, Sarkari M, Shahraki F, Baxter LL. Energy DT: temperature driving force of exchanger ( C, K)
saving in a crude distillation unit by a retrofit design of heat exchanger net- DTmin: minimum temperature approach difference ( C, K)
works. Indian J Chem Technol 2013;20:290e3. DTc : temperature difference of cold stream ( C, K)
[22] Chegini S, Dargahi R, Mahdavi A. Modifications of preheating heat exchanger DTc* : shifted temperature difference of cold stream ( C, K)
network in crude distillation unit of Arak refinery based on pinch technology. DTh : temperature difference of hot stream ( C, K)
In: Proceeding of the world congress on engineering and computer science DTh* : shifted temperature difference of hot stream ( C, K)
WCECS. San Francisco, USA; 2008.
[23] Kamel D, Gadalla M, Ashour F. New retrofit approach for optimisation and Abbreviation
modification for a crude oil distillation system. Chem Eng Trans 2013;35:1363e8.
[24] Hohman EC. Optimum networks of heat exchange. PhD Thesis. CA, USA:
University of Southern California; 1971. c: cold stream
[25] Huang F, Elshout RV. Optimizing the heat recovery of crude units. Chem Eng cw: cooling water
Prog 1976;72:68. h: hot stream
[26] Linnhoff B, Mason DR, Wardle I. Understanding heat exchanger networks. HEN(s): heat exchanger network(s)
Comp Chem Eng 1979;3:295. hx: exchanger unit
[27] Linnhoff B, Townsend DW, Boland D, Hewitt GF, Thomas BEA, Guy AR, et al. LLR: left lower region
A user guide on process integration for the efficient use of energy. 1st ed. UK: LUR: left upper region
IChemE; 1982. RLR: right lower region
[28] Umeda T, Itoh J, Shiroko K. Heat exchange system synthesis. Chem Eng Prog RUR: right upper region
1978;74:70.

Mamdouh Article 2 PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Mamdouh Article 2 PDF

Hochgeladen von

Copyright:

Verfügbare Formate

Energy xxx (2015) 1e16

Contents lists available at ScienceDirect

A new graphical method for Pinch Analysis applications: Heat

1. Introduction optimise existing operating conditions for minimum energy de-

Pinch Analysis or technology is an outstanding methodology for

It is obvious then that the new graphical representation can help

3.1. A new graphical representation for existing HENs

As shown in Fig. 4, consider a heat exchanger hx, transferring

If the heat capacities of hot and cold streams are considered

zontal distance is equivalent to DTc. The length of the straight line L

Then, Eq. (10) can be written as:

The existing heat exchanger network is described using the new

▪ All exchanger matches are shown, connecting on the cold side

Stream Supply Target Heat Heat capacity

H1 160 120 20 0.5

crude oil feed, and includes atmospheric crude distillation tower

Fig. 14. A real reﬁning plant for crude oil fractionation.

Table 2 64.83 MW as target, resulting in an excess consumption of fuel oil

Each of the above violations of the Pinch Analysis principles in

hence large heat transfer area. The same implication of tempera-

E-11@ADU 250.6 261.5 336.7 288.4 4.509 127 69

[1] Gadalla MA, Jobson M, Simth R. Optimization of heat-integrated reﬁnery

Das könnte Ihnen auch gefallen