Process modications to mintegration

Kew Hong Chewa, Jir Jaromr K

T), Facultion e CP10, H-82

Adaptation of the PluseMinus prin-ciple to select benecial process

ntal overlap region.gions using the TSP,modied Problem

new heuristics. Theidual processes. Thecontext can furtherthe advantages of

. All rights reserved.

1. Introduction

Efcient integration and optimisation of the energy require-ment within a site utility system can improve the Total Site energy

efciency. Energy saving as a result of improved energy efciencysubsequently translates into reduced CO2 emissions. Such outcomefrom energy saving ts the European Directives 2010/75/UE whichprioritises on the reduction of industrial emissions at the source [1].

Total Site Heat Integration (TSHI) is a tool used to optimise thesite-wide energy demand. It extends the application of PinchAnalysis from a single process to multiple processes in an industrialsite. The concept of Total Site (TS) was introduced by Dhole and

* Corresponding author. Tel.: 60 07 5535533; fax: 60 07 5536165.

Contents lists availab

Applied Therma

sev

Applied Thermal Engineering 78 (2015) 731e739E-mail address: shasha@cheme.utm.my (S.R. Wan Alwi).Keywords:Total Site Heat IntegrationPinch AnalysisProcess modicationsExtended PluseMinus Principles

Sink and Source Proles, (b) the horizontal overlap region and (c) below the horizoThe proposed methodology identies the options to reduce utility targets in these reSite Utility Composite Curves (SCC), Utility Grand Composite Curve (UGCC),Table Algorithm (PTA), Total Site Problem Table Algorithm (TS-PTA) and someidentied changes on the TSP are then linked to the specic changes at the indivillustrative case study shows that the PluseMinus principle application in the TSHIimprove heat recovery. The proposed spreadsheet-based methodology combinesgraphical visualisation, as well as the numerical precision.

2014 Elsevier LtdAvailable online 28 April 2014process modication options to be selected in order to maximise energy savings in TSHI. The Total SiteProle (TSP) is divided into three regions: (a) the region above the horizontal overlap between the SiteReceived 4 January 2014Accepted 18 April 2014http://dx.doi.org/10.1016/j.applthermaleng.2014.04.041359-4311/ 2014 Elsevier Ltd. All rights reserved.total site heat integration (TSHI). The PluseMinus principle has been adapted to enable the benecialmodication options. Development of heuristic to set pri-ority of streams.

Targeting process modications atselected process sections to improveTSHI heat recovery.

a r t i c l e i n f o

Article history:a b s t r a c t

This paper extends the scope of the Pinch Analysis for process modications of individual processes toZainuddin Abdul Manan a

a Process Systems Engineering Centre (PROSPECbCentre for Process Integration and IntensicatTechnology, University of Pannonia, Egyetem u.

h i g h l i g h t s

Extension of the Pinch Analysis forprocess modications from singleprocesses to TSHI.aximise energy savings in total site heat

lemes b, Sharifah Radah Wan Alwi a,*,

y of Chemical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, MalaysiaI2, Research Institute of Chemical and Process Engineering e M}UKKI, Faculty of Information00 Veszpr _em, Hungary

g r a p h i c a l a b s t r a c tjournal homepage: www.el4le at ScienceDirect

l Engineering

ier .com/locate/apthermeng

al ELinnhoff [2]. Klemes et al. [3] established the Total Site Proles(TSPs) to represent the thermal prole of TS and the Site UtilityComposite Curves (SCC), which consists of Hot and Cold utilityComposite Curves. The Total Site Pinch is where the two utility CCoverlap horizontally. The Utility Grand Composite Curve (UGCC)provides a visual illustration of the external utility requirements.These graphical tools are used to set targets for steam usage andgeneration by the site processes, the steam required to be producedby the boilers, and shaftwork produced by the steam turbines. Heatrecovery depends also on the number of steam levels. Even thoughmore steam levels may result in more heat recovery, it has to bebalanced against the increased capital cost and higher complexityof the utility system [4]. The TSHI has been enhanced to LocallyIntegrated Energy System (LIES) which also integrates residential,business, service and even agriculture areas apart from the indus-trial site [5]. Liew et al. [6] extended their numerical TSHI meth-odology to consider the operational changes within a centralisedutility system planning. A brief review of the development of TSHIover the last 40 years can be found in the recent review paper byKlemes and Kravanja [7].

The energy targets from the TSHI analysis are dependent on theoperating conditions of the existing site. A change in the processoperating conditions can alter the TSP and change the energy tar-gets. The exibility to change the process conditions can beexploited to further improve the heat recovery. In a study to inte-grate new CO2 capture and storage (CCS) plants into existing coal-red power stations, Harkin et al. [8] showed that energy penaltyassociated with CCS can be reduced when TSHI is applied. Theirstudy also identied that the pre-drying of coal could furtherimprove the utility targets. Process modication strategies toimprove the Heat Integration (HI) of single processes by exploitingthe shape of Composite Curves (CCs) and the Grand CompositeCurve (GCC) were developed by Linnhoff and Vredeveld [9]. Therules for process modications using the CC and GCC include thePluseMinus principle, keep hot stream hot (KHSH) and keep coldstream cold (KCSC) as well as the appropriate placement principle.Exploiting and optimising the process soft data, use of the appro-priate minimum approach temperature (DTmin) and suitableapplication of insulations above and below Pinch can be effective inimproving heat recovery. Lee et al. [10] has shown that knowledgeof pinch can be used to optimise the location of pipe insulation toreduce utility targets. Applications of these rules have beendescribed earlier by Smith [11], later by Kemp [12], and recently inextended ways by Klemes et al. [13].

The process modication strategies such as the PluseMinusprinciple can be used together with the TSP to identify the scopefor process modications to improve TSHI. Hackl et al. [14]showed that the gap between the TSP and SCC can be used toidentify the potential utility systems changes that can reduce theoverall site heating and cooling requirements in a Side-WideProcess Integration study of a chemical cluster in Sweden.Replacing the low pressure steam (LPS) heating with hot water(generated from the Site Source) reduced the gap between TSPand SCC, changed the shape of the SCC and shifted the TS Pinch.This increases the overlap of the Site Source and Sink Proles andincreases the heat recovery. Nemet et al. [15] developed thestrategies to plan the extension of an existing site by using thePluseMinus principle on the TS. The PluseMinus principle,together with the Process Utility Matrix (which lists the utilitiesconsumptions of the various processes), were used to evaluate themerits of integrating a new process to an existing TS. Only optionswhich are benecial, i.e. those resulting in improved overall heatrecovery will be selected for integration. In this study, the PluseMinus principle of process modications is used to identify pro-

K.H. Chew et al. / Applied Therm732cess changes that can further reduce the TS utility targets. This isessentially done by manipulating the shape of the Site Source andSink Proles.

2. Application of pinch strategies for process modications ofa single process to TS

Knowledge of Pinch location is crucial when exploring theprocess modication opportunities for single processes. The TotalSite (TS) Pinch location can provide similarly a guide during processmodications to reduce the overall energy consumption of the site.The TS Pinch limits the amount of heat that can be recovered fromthe Site Source and Sink. According to Klemes et al. [3], the TS Pinchis the point where the cold utility CC rst intersects with the SiteSink Prole (SSiP) or when the hot utility CC rst intersects the SiteSource Prole (SSoP). TS Pinch occurs where the horizontal overlapbetween the Utility Composite Curves is maximised. The TS Pinchoccurs at the utility temperatures, and spans between the tem-peratures of two successive utility levels as shown in Fig. 1. Anassessment of the impact of process modications of a single pro-cess on the TS as given in Table 1 shows that Heat Integrationstrategies applied for the single process can be directly extended toTS. These include the exploitation of soft data, the appropriate useof minimum temperature of approach (considering the uids inservice and the type of heat exchanger used) and the suitable use ofinsulation (e.g. apply insulation only on hot streams above pinch,etc.). Adaptations of the PluseMinus Principles to TS are describedin the subsequent sections.

3. The PluseMinus principles for TS

A TS comprising several units (e.g. chemical processing plants,business and commercial units) typically uses steam as theworkinguid. For an existing steam system of a TS, little can be done tooptimise the steam utility levels to improve heat recovery. Theutility targets can be reduced by exploring the potential for processmodications in the TS context. The TSP can be a powerful tool toevaluate the potential for further heat recovery improvement evenfor a TS. The Site Source Prole (SSoP) is analogous to the HotComposite Curve and the Site Sink Prole (SSiP) is analogous to theCold Composite Curve.

The TSP can be divided into three regions: (a) above the SSoP andSSiP horizontal overlap, (b) at the SSoP and SSiP horizontal overlapand, (c) below the SSoP and SSiP horizontal overlap. Note that theoverlap between the SSoP and SSiP spans between the highest (TPH)and lowest (TPL) process pinch temperatures on the site, as shownin Fig. 1. Above and below the SSoP and SSiP horizontal overlapregion, the location of TS Pinch has no implication on TS heat re-covery. Above the horizontal overlap region, the heating require-ment can be reduced by decreasing the duty of the cold streams.Below the horizontal overlap region, the cooling requirement canbe reduced by decreasing the duty of the hot streams.

Within the SSoP and SSiP horizontal overlap region, TS Pinchaffects heat recovery in the same way the Pinch does for a singleprocess. For the SSoP, the TS Pinch can also be taken as the tem-perature equal to the higher steam level of the TS Pinch. Above thistemperature, increasing the duty of the SSoP () reduces the hotutility. Below this temperature, decreasing the duty of SSoP ()reduces the cold utility. For the SSiP, the SP can be taken at thetemperature equal to the lower steam level of the TS Pinch (SP).Above this temperature, decreasing the SSiP () reduces the hotutility. Below this temperature, increasing the duty of SSiP reducesthe cold utility.

The UGCC provides a quick visual impression of the externalutility requirements, and can be used to prioritise the changes on

ngineering 78 (2015) 731e739the TSP segments in order to reduce utilities. Fig. 1 illustrates the

and cold CC to set the guideline on howheat recovery can be further improvedwith respect to pinch location [9].Above Pinch, heating requirement canbe reduced by increasing the duty of hotstreams or decreasing the duty of coldstream. Below Pinch, cooling

Site Pinchhorizontal

al ELPSoverlap region

Belowhorizontal

overlap region

TPLHot utility, QHVHPS

T

HPS

MPSSSoP & SSiP

Abovehorizontal

overlap region TPH

K.H. Chew et al. / Applied Thermapplication of the PluseMinus principle on a TS which uses fourutility levels, i.e. very high pressure steam (VHPS), high pressuresteam (HPS), medium pressure steam (MPS) and low pressuresteam (LPS). The results of the application is summarised in Table 2.

4. Methodology

The suggested algorithm to target process modications forTSHI is described below (see also Fig. 2).

(a) Data extraction: Extract the hot and cold stream data, i.e. thesupply and target temperatures as well as the heat capacityof the streams, from individual processes that would be

(a) Initial TSP and SCC

(b) Plus-Minus Principles

Cold utility, QcH

VHPST

H

HPS

MPS

LPS

Hot utility, QH

SSiP Pinch

SSoP Pinch

Cold utility, Qc

+

+

-

-

SSoP & SSiP horizontal

overlap region

Abovehorizontal

overlap region

Belowhorizontal

overlap region

TPH

TPL

(c) New TSP and SCC

SSoP & SSiPhorizontal

overlap region

Abovehorizontal

overlap region

Belowhorizontal

overlap region

New Site Pinch

Hot utility, QHVHP

T

H

HPS

MP

LP

Cold utility, Qc

TPH

TPL

Fig. 1. Analogy of the PluseMinus to a TS (using steam as the working uid).Table 1Application of Pinch strategies for process modications on TS.

Pinch strategies for singleprocess

Application on Total Site

PluseMinus principle For single process, the PluseMinusprinciple is used together with the hot

ngineering 78 (2015) 731e739 733integrated in the TS. In addition, obtain the existing utilitytemperatures. An example of the data required for theanalysis is given in Table 3.

(b) TS analysis:

i. Prepare the individual process Problem Table Algorithm (PTA)e a modied version of PTA [16] by listing the heat capacity (CP)

requirement can be reduced bydecreasing the duty of hot streams orincreasing the duty of cold streams.This principle can be applied on TS withsome adaptation (see Section 3). TSPand SP as dened by Klemes et al. [3]are used.

KHSH, KCSC This describes how the change instream duty can be effected by thechange in stream mass heat capacity soas to reduce utility targets. This isencompassed within the PluseMinusprinciple.

Appropriate placementprinciple

For single process this is used withprocess GCC to explore the integrationof key equipment for e.g. distillationcolumns, evaporators, heat pumps, etc.in order to reduce the utilityrequirements. Detail descriptions ofthis principle and other examples canbe found in Smith [11], Kemp [12] andKlemes et al. [4]

Exploiting and optimising ofprocess soft data

Such operation exibility has beensuccessfully exploited to reduce utilitytargets for single process [12].Walmsley et al. [18] demonstrated thesuccessful optimisation of soft data inimproving the heat recovery of aindustrial milk powder plant.In the context of TS, application of thisprinciple will be at the individualprocess level. For e.g. once thelocation of the favourable processmodication is targeted using the PluseMinus principle and TSP, the specicprocess modication can be broughtabout by the use of this principle.

Appropriate use of DTmin Reducing DTmin increases the overlapbetween the hot and cold CC hencereduces the cooling and heatingutilities. Innovation in heatenhancement and type of heatexchanger that allow smaller DTmin canfurther improve heat recovery.Application of this can simply beextended to TS. Note that the TSHImethodology has been extended toallow specication of individual DTminfor each process and between processand utilities by Varbarnov et al. [19]

Suitable use of insulation This is used when heat loss to theenvironment is signicant compared tothe process heating duties.Application of this can simply beextended to TS.

contribution of individual streams and total CP contributions ofeach process at various temperature levels. In generating the pro-cess PTA, the temperatures are shifted by half of the DTmin (be-tween process and process) to ensure that the minimumtemperature difference between hot and cold streams are main-tained. The shifted temperature, T*:

For hot stream; T* T DTminprocesseprocess.2 (1)

For cold stream; T* T DTminprocesseprocess.2 (2)

ii. Prepare the TS PTA e an expanded version of TS PTA [17]listing the heat capacity contribution of individual processes forthe site source and the site sink. The utility usage and generationare directly interpolated on the TS-PTA at the utility temperatures.

shifted back to their original values, and then shifted again byDTmin(between process and utility) to ensure that the minimum tem-perature difference between process and utility is maintained. Thedouble shifted temperature, T**:

For TS Sink; T** T DTminprocesseprocess.2

DTminprocesseutility(3)

For TS Source; T** T DTminprocesseprocess.2

DTminprocesseutility(4)

Table 3Stream data and utilities level for analysis.

Process Stream Temperature, C Mass CP Utilities Temperature, CSupply Target kW/C Supply Target

A Hot H1 230 55 200.0 Hot VHPS 300 eH2 155 80 733.3 HPS 260 e

Cold C1 120 270 296.8 MPS 200 eC2 70 150 750.0 LPS 150 e

B Hot H1 240 200 800.0 Cold CW 32 40H2 230 70 187.5H3 150 60 444.4 DTmin between process and

process used is 20 CCold C1 50 210 500.0C2 90 250 312.5

C Hot H1 250 90 275.0 DTmin between process andutilities used is 15 CH2 220 80 428.6

Cold C1 150 260 390.9

Table 2Application of the PluseMinus principles to a Total Site (Using steam as the workinguid).

SSoP SSiP

Above SSoP and SSiP overlap areaY duty () QH Y

SSoP and SSiP overlap areaAbove SSoP Pinch, QH Y, limited

by Site PinchAbove SSiP Pinch, QH Y, limited

by Site Pinch[ duty () Y duty ()Below SSoP Pinch, QC Y, limited

by Site PinchBelow SSiP Pinch, QC Y, limited

by Site PinchY duty () [ duty ()

Below SSoP and SSiP overlap areaY duty () QC Y

K.H. Chew et al. / Applied Thermal Engineering 78 (2015) 731e739734In generating the TS-PTA, the shifted temperatures, T* are rstFig. 2. Algorithm to target prociii. Prepare the TSP, SCC (for hot and cold utilities) and the UGCC.ess modications for TSHI.

(f) Similarly, the contributing streams corresponding to the TSP

(h) Asspinoth

(i) Selreson

5.3%. VHPS consumption reduces by 1.5 MW while MPS con-

al E(j) Reconstruct TSP, SCC and UGCC using the new stream data.(k) Check if the resulting heat recovery from the TSP changes is

limited by Site Pinch as in Table 2. If yes, go back to step (h). Ifno, the option is acceptable.

5. Illustrative example

The TS consists of processes A, B and C. Table 3 gives the streamdata and utilities available on-site. Table 4 shows the extended PTAfor each process listing the CPs of the contributing streams. Themodied TS-PTA for Site Sink and Source is given in Tables 5 and 6.From the TS-PTA (Tables 5 and 6) and TSP (Fig. 3), the corre-sponding Site Source and Site Sink enthalpies (hH1, hH2, hH3, hC1,hC2, hC3 and hC4) at each utilities level (VHPS, HPS, MPS and LPS) canbe directly interpolated and the utility consumptions and genera-tions can be determined as follows:Utilityreduce QH (hot utility), reduce the duty of the coldstreams that lie above the Process Pinch. For cold streamsthat straddle across the Process Pinch, those streams thathave larger proportion of their temperatures above theProcess Pinch should be given the priority.Heuristic #3: Within the horizontal overlap region, toreduce Qc (cold utility), increase the duty of hot streamsthat lie above the Process Pinch. For hot streams thatstraddle across the Process Pinch temperatures, thosestreams that have larger proportion of their temperaturesbelow the Process Pinch should be given the priority.Heuristic #4: Below the horizontal overlap region, hotstreams with temperatures lower than TPL should beprioritised for process modications. For hot streams thatstraddle across TPH, those streams that have larger pro-portion of their temperatures below TPL should be giventhe priority.Heuristic #5: Stream with larger CP should be prioritisedfor process modications.ess the scope of feasible process modications, using thech techniques for single process. Repeat steps (e)e(h) forer utilities.ect only the process modication options that wouldult in a net reduction in hot and cold utilities consumptionthe TS.segment and the specic process can be traced back from theprocess PTA.

(g) The priority of the streams to be investigated for processmodications can be set based on the heuristics:

Heuristic #1: Above the horizontal overlap region, coldstreams which lie above TPH should be prioritised forprocess modications. For cold streams that straddleacross TPH, those streams that have larger proportion oftheir temperatures above TPH should be given the priority.Heuristic #2: Within the horizontal overlap region, to(c) From the UGCC, prioritise the utilities to be targeted.(d) Starting with the rst utility, identify the corresponding

segment on the TSP using the PluseMinus principle.(e) From the TSP segment identied, trace back the contributing

process or processes from the TS-PTA at the correspondingtemperature intervals. This can be done easily as the TS-PTAlists the contribution from each process by the temperatureintervals of TSP. If more than one process is involved, there is achoice of whether to eliminate some or to include all forfurther evaluation. The priority of processes to be targeted canbe set by examining the process heat capacity on the TS PTA.

K.H. Chew et al. / Applied Thermconsumption by the Sink:that stream C2 of Process B has a target temperature reduction of5 C. The targeted utility, HPS consumption reduces by 1.6 MW orVHPS hC1 hC2 (5)

HPS hC2 hC3 (6)

MPS hC3 hC4 (7)

LPS hC4 (8)Utility generation by the Source:

MPS hH1 (9)

LPS hH1 hH2 (10)

CW hH2 hH3 (11)TSP represents the overall proles of the heat sources and sinks

on the TS. The SP location is obtained by plotting the SCC (Fig. 4).The net utility requirements are summarised in the UGCC (Fig. 5)and Table 8. The overlap of the Site Sink and Site Source spansbetween 105 C (TPL) and 205 C (TPH). The SSoP only extendsslightly above the MPS temperature of 200 C with very little MPSgeneration. The 0.33 MWof excess LPS that cannot be used is to berejected to CW.

The UGCC sets the priority for utilities to be targeted forreduction. The larger enthalpy and more costly utility should be setas the priority to be reduced rst. For this example, let x be the baseunit cost per MWof utility. Let the cost of VHPS, HPS, MPS, LPS andCW equals to 2.5x, 2x, 1.5x, x and 0.2x. From Tables 5 and 6, the netutilities requirements are VHPS 14.8 MW, HPS 30.0 MW,MPS 32.5 MW, LPS 0, CW 95.9 MW. The corresponding utilityrequirements in terms of x, i.e. utility requirement (MW)multiplieswith utility unit cost per MW, are VHPS 37.1x, HPS 59.9x,MPS 51.1x, and CW 19.2x. The priorities are (1) HPS, (2) MPS, (3)VHPS and (4) CW.

The application of PluseMinus Principles is shown in Fig. 6 andsummarised in Table 7. Segment A2 above the overlap region (i.e.between 205 C and 260 C) has the rst priority, process modi-cations that reduce the heating requirements would directly reducethe TS HPS usage. Segment B within the overlap region (between150 C and 205 C) has the second priority. Process modicationsreducing the duty of Segment B would reduce MPS usage. SegmentA1 (above 260 C) has the third priority. Process modicationsreducing the duty of this segment would directly reduce the VHPSrequirement. Segment C can be ignored since there is alreadyexcess LPS. Segments D1 and D2 have the lowest priority. Reducingthe duty of these segments could reduce the CW requirement.

For Segment A2: from the Site Sink TS-PTA (Table 5), Process Bhas slightly larger CP and should be targeted rst compared toProcess A [Heuristic #5]. From Process B PTA (Table 4), the onlycontributing stream is C2. The heat duty can be modied bychanging the stream heat capacity (CP) and/or stream temperature.Changing of CP is possible only when the stream mass ow can bechanged since specic capacity is the physical attribute of a stream.When a stream mass ow is not dictated by the productionthroughput, such as that of a recycle stream or a reux, the streamow can be optimised with the aim to maximise Heat Integration.The stream temperature can be changed by exploiting the exibilitythat exists in the operating conditions of the process. For illustra-tion purposes, supposed process modications are feasible such

ngineering 78 (2015) 731e739 735sumption reduces slightly by 0.5 MW. There is no change to the

Table 4Process PTA (listing streams CP).

Streams CP Cascade DH MW 1st pass

T* DT H1 H2 H3 C1 C2PCP DH DH DH

C C kW/C kW/C kW/C kW/C kW/C kWC MW MW MW

PROCESS A280 0 38.58220 60 296.8 296.8 17.81 17.81 20.78160 60 200.0 296.8 96.8 5.81 23.61 14.97145 15 200.0 296.8 750.0 846.8 12.70 36.31 2.27125 20 200.0 733.3 296.8 750.0 113.4 2.27 38.58 0.0080 45 200.0 733.3 750.0 183.3 8.25 30.33 8.2570 10 200.0 733.3 933.3 9.33 21.00 17.5845 25 200.0 200.0 5.00 16.00 22.58

PROCESS B260 0 37.72230 30 312.5 312.5 9.38 9.38 28.35220 10 800.0 312.5 487.5 4.88 4.50 33.22190 30 800.0 187.5 500.0 312.5 175.0 5.25 0.75 38.47140 50 187.5 500.0 312.5 625.0 31.25 30.50 7.22100 40 187.5 444.4 500.0 312.5 180.6 7.22 37.72 0.0060 40 187.5 444.4 500.0 131.9 5.28 32.44 5.2850 10 444.4 444.4 4.44 28.00 9.72

K.H. Chew et al. / Applied Thermal Engineering 78 (2015) 731e739736excess LPS generation and CW requirement. The same procedure

PROCESS C270240 30 390.9210 30 275.0 390.9160 50 275.0 428.6 390.980 80 275.0 428.670 10 428.6may be applied to Process A to further reduce the utilityconsumptions.

At the overlap region, changes can be made on the SSiP or SSoPor both as shown in Fig. 6. The effects of changes are not as explicitas with the regions above or below the overlap due to the in-teractions between the sink and the source in heat integration. Fore.g., to modify Segment D1 (i.e. the SSoP): from site source TS-PTA,the only contributing process is Process C. From Process C PTA, twostreams lie at Segment D1, i.e. H1 and H2. H2 is selected as it lies

Table 5TS-PTA for Sink (listing contributing processes CP).below the Process Pinch [Heuristic #3] and its CP is larger than that

0 15.20390.9 11.73 11.73 3.48115.9 3.48 15.20 0.00312.7 15.63 0.43 15.63703.6 56.29 56.71 71.92428.6 4.29 61.00 76.20of H1 [Heuristic #5]. Supposed process modications allow thestream H2 CP to be reduced by 10%. The change only marginallyreduces the VHPS, HPS and MPS consumptions, between 0.7 and2.7%. This is expected since Segment D1 has the lowest priority (seeTable 7). The excess LPS reduces by about 20% (1.3 MW) as expectedof a targeted utility. Applying the same procedure on Segment B(i.e. the SSiP): supposed the CP of the selected stream C1 (Process B)can be reduced by 5% and its target temperature can be lowered by5 C. The targeted utility consumption, MPS, reduces by 14%

Table 6TS-PTA for Source (listing contributing processes CP).

K.H. Chew et al. / Applied Thermal Engineering 78 (2015) 731e739 737200

250

300

350

SSiP

SSoP

VHPS

T** C(4.5 MW) while the VHPS and HPS consumptions remain un-changed. However there is an 18% increase in excess LPS (1.2 MW)and subsequently, 19% (18.5 MW) increase in CW requirement.

Below the overlap region, at Segment D2: from site source TS-PTA, Process C is the larger CP contributor compared to Process Aand B therefore selected for further investigation. From Process CPTA, streams H1 and H2 are good candidates for exploring potentialchange in stream duty as both streams lie below the Process Pinch

1.6 MW (10.8%) VHPS, 3.9 MW (12.0%) MPS and 7.1 MW (7.4%) CWconsumptions. The HPS consumption increased marginally by

0

50

100

150

-120 -100 -80 -60 -40 -20 0 20 40 60 80 100

HPS

MPS

LPS

CWhC1hC2hC3hC4hH1hH2hH3

Fig. 3. Total Site Prole (TSP).

Fig. 4. Site utilities composite curves: CUCC, HUCC.0.4MW (1.3%). The TSP shows that the lowgrade heat is not neededby the processes on site. The cooling duty can be reduced either bymodifying the existing process design to exploit this low gradeheat, or by exporting it to other users in the neighbouring sites.

6. Conclusion

A systematic TSHI methodology to identify and target the po-tential process modications and further increase energy conser-vation at Total Sites has been developed. The case studydemonstrates that the PluseMinus Principles developed for asingle process can be successfully adopted and applied to a TS. TheTSP, SCC and UGCC can provide useful insights for the plant[Heuristic #4]. H1 is selected since H2 has already been modiedwith changes at Segment D1. Supposed process modication isfeasible and the CP of H1 can be reduced by 5% and its targettemperature can be increased by 5 C. The reduction in CWusage isminor at 1.5 MW or 3.2%. The HPS usage increases marginally by0.1 MW (0.5%). Excess LPS reduces slightly by 4.1 MW (4.3%).

A summary of the results is given in Table 8. The combinedchanges of Segments A2, D1, B and D2 resulted in reduction ofdesigner to identify where, in terms of which temperature in-terval, and which streams within the entire TS to focus the process

Fig. 5. Utility Grand Composite Curve (UGCC).

modication efforts to improve the site HI. The implementation ofthe PluseMinus Principles on TSP identies the potential changesthat would increase the energy conservation. The proposedchanges to the selected streams should be assessed from feasibility,practicality and economic perspectives. This requires the expertiseof the plant designer/manager to explore potential feasible andpractical process modications to achieve the targeted changesidentied. The selected and potentially acceptable process modi-cation options can be conveniently merged with potential retrotproject (e.g. to increase plant capacity) considered for the TotalSite.

The proposed methodology can benet from the visualisationadvantages of the graphical method [4], and from the precision ofthe numerical method [17]. The TSP, SCC and UGCC can be easilygenerated using the chart function e.g. in the Excel spreadsheet setup for the process PTA and TS-PTA analysis. The methodologyprovides an uncomplicated and relatively fast way of assessing theimpacts of proposed process modications on the TS hot and/orcold utilities without the need for a time consuming detailed pro-cess simulation efforts.

Acknowledgements

The authors gratefully acknowledge the nancial supports fromthe Universiti Teknologi Malaysia (UTM) Research University Grantunder Vote No. Q.J130000.2509.07H35 and the EC FP7 projectENER/FP7/296003/EFENIS Efcient Energy Integrated Solutions forManufacturing Industries e EFENIS. The support from the Hun-garian project Trsadalmi Megjuls Operatv Program TMOP -4.2.2.A-11/1/KONV-2012-0072 - Design and optimisation ofmodernisation and efcient operation of energy supply and uti-lisation systems using renewable energy sources and ICTs signi-cantly contributed to the completion of this analysis.

Nomenclature

CP heat capacity owrate, MW/ChC corresponding Site Sink enthalpies at utilities level, MWhH corresponding Site Source enthalpies at utilities level,

MWQ cooling utilities heat owrate, MW

Table 7TSP analysis for application of PluseMinus principles (Case Study) (Refer to Fig. 6).

Segment Temperature Location Utility Usage/G

SSiPA1 T > 260 C Above overlap region VHPS

A2 260 > T > 205 C Above overlap region HPS

B 205 > T > 150 C At overlap region MPS

C 150 > T > 105 C At overlap region LPS

SSoPD1 205 > T > 105 C At overlap region LPS

D2 T < 105 C Below overlap region CW

Cs

1321M

he db values represent an increase in utility consumption compared to base case.c Excess LPS has be accounted for in the increase in CW consumption.

K.H. Chew et al. / Applied Thermal E738Table 8Summary of results (case study).

Utility level Utility consumption, MW

Base study Changesegment A2

Changesegment D1

VHPS 14.8 13.3 14.7HPS 30.0 28.4 29.2MPS 32.5 32.0 31.9LPS 6.5a 6.5 5.2CWc 95.9 95.9 90.2Targeted reduction as per Table 7 HPS LPS

a values represent excess LPS generation. The CW consumption has included tFig. 6. TSP with PluseMinus principles.eneration G Principle Priority Potential process modications

Y duty () QH Y 3 Y CP and/orY target temperature

Y duty () QH Y 1 Y CP and/orY target temperature

Y duty () QH Y 2 Y CP and/orY target temperature

[ duty () QC Y e None, since there isalready excess LPS

Y duty () QC Y 4 Y CP and/or[ target temperature

Y duty () QC Y 4 Y CP and/or[ target temperature

Overall saving

hangeegment B

Changesegment D2

Combined changeof A2, B, D1 & D2

MW %

4.8 14.8 13.3 1.6 10.80.0 30.4 30.4 0.4b 1.48.0 32.6 28.6 4.0 12.17.7 5.8 4.5 ec e14.1 91.8 88.8 7.0 7.4PS CW

issipated of this excess heat.

ngineering 78 (2015) 731e739cQh heating utilities heat owrate, MW

DH process heat owrate, MWDTmin minimum approach temperature, CDTmin (processeprocess) minimum approach temperature between

process and process, CDTmin (processeutility) minimum approach temperature between

process and utility, CT* shifted temperature for process PTAT** double shifted temperature for TSP plot and TS-PTA, CTPH the highest Process Pinch temperature on site, CTPL the lowest Process Pinch temperature on site, C

AbbreviationsCC Composite CurveCCS Carbon capture and Carbon storageCUCC Cold Utility Composite CurveGCC Grand Composite CurveEU European UnionHC hydrocarbonsHI Heat IntegrationHPS high pressure steam, bargHUCC Hot Utility Composite CurveKHSH Keep Hot Stream HotKCSC Keep Cold Stream ColdLPS low pressure steam, barg

Control), (accessed 14.05.13).

[2] V.R. Dhole, B. Linnhoff, Total site targets for fuel, co-generation, emissions, andcooling, Comput. Chem. Eng. 17 (1993) S101eS109.

[3] J.J. Klemes, V.R. Dhole, K. Raissi, S.J. Perry, L. Puigjaner, Targeting and designmethodology for reduction of fuel, power and CO2 on total sites, Appl. Therm.Eng. 17 (8e10) (1997) 993e1003.

[4] J.J. Klemes, F. Friedler, I. Bulatov, P. Varbanov, Sustainability in the ProcessIndustry e Integration and Optimization, McGraw-Hill, New York, USA, 2010.

[5] S. Perry, J. Klemes, I. Bulatov, Integrating waste and renewable energy toreduce the carbon footprint of locally integrated energy sectors, Energy 33(10) (2008) 1489e1497.

[6] P.Y. Liew, S.R. Wan Alwi, P.S. Varbanov, Z.A. Manan, J.J. Klemes, Centralisedutility system planning for total site heat integration network, Comput. Chem.Eng. 57 (2013) 104e111.

[7] J.J. Klemes, Z. Kravanja, Forty years of heat integration: pinch analysis (PA) andMathematical Programming (MP), 2013, Curr. Opin. Chem. Eng. 2 (4) (2013)461e474.

[8] T. Harkin, A. Hoadley, B. Hooper, Reducing the energy penalty of CO2 captureand compression using pinch analysis, J. Clean. Prod. 18 (9) (2010) 857e866.

[9] B. Linnhoff, D.R. Vredeveld, Pinch technology has come of age, Chem. Eng.Prog. 80 (7) (1984) 33e40.

[10] C. Lee, S.R. Wan Alwi, Z.A. Manan, D.M. Fraser, Targeting maximum heat re-covery for systems with heat Losses and Heat Gains, 8th Conference on Sus-tainable Development of Energy, Water and Environment Systems (SDEWES2013), 22e27 Sept 2013, Dubrovnik, Croatia, lecture SDEWES2013-0480.

[11] R. Smith, Chemical Process e Design and Integration, McGraw-Hill, New York,USA, 2005.

[12] Kemp, I.C. 2007 (authors of the rst edition B. Linnhoff, D.W. Twonsend, G.F.Hewitt, B.E.A. Thomas, A.R. Guy, R. Marsland), Pinch Analysis and ProcessIntegration: A User Guide on Process Integration for the Efcient use of En-ergy. Elsevier, Amsterdam, The Netherlands.

[13] J.J. Klemes, P.S. Varbanov, Z. Kravanja, Recent developments in process inte-

K.H. Chew et al. / Applied Thermal Engineering 78 (2015) 731e739 739MPS medium pressure steam, bargPTA Problem Table AnalysisSCC Site Utility Composite CurvesSSiP Site Sink ProleSSoP Site Source ProleTS Total SiteTS-PTA Total Site Problem Table AnalysisTSHI Total Site Heat IntegrationTSP Total Site ProleUGCC Utility Grand Composite CurveVHPS very high pressure steam, barg

References

[1] EU, Directive2010/75/EU of the European Parliament and of the Council of 24November 2010 on Industrial Emissions (Integrated Pollution Prevention andgration, Chem. Eng. Res. Des. 91 (10) (2013) 2037e2053.[14] R. Hackl, E. Andersson, S. Harvey, Targeting for energy efciency and

improved energy collaboration between different companies using total siteanalysis (TSA), Energy 36 (8) (2011) 4609e4615.

[15] A. Nemet, J. Klemes, P.S. Varbanov, M. Atkins, M. Walmsley, Total site meth-odology as a tool for planning and strategic decisions, Chem. Eng. Trans. 29(2012) 115e120.

[16] B. Linnhoff, J.R. Flower, Synthesis of heat exchanger networks: I. Systematicgeneration of optimal network, AIChE J. 24 (4) (1978) 633e642.

[17] P.Y. Liew, S.R. Wan Alwi, P.S. Varbanov, Z.A. Manan, J.J. Klemes, A numericaltechnique for total site sensitivity analysis, Appl. Therm. Eng. 40 (0) (2012)397e408.

[18] T.G. Walmsley, M.R.W. Walmsley, M.J. Atkins, J.R. Neale, Improving energyrecovery in milk powder production through soft data optimisation, Appl.Therm. Eng. 61 (1) (2013) 80e87.

[19] P.S. Varbanov, Z. Fodor, J.J. Klemes, Total site targeting with process specicminimum temperature difference (DTmin), Energy 44 (1) (2012) 20e28.

Process modifications to maximise energy savings in total site heat integration1. Introduction2. Application of pinch strategies for process modifications of a single process to TS3. The PlusMinus principles for TS4. Methodology5. Illustrative example6. ConclusionAcknowledgementsNomenclatureReferences