Accepted Manuscript

Title: Modified Approach of Total Site Integration for Energy

Conservation: A Case Study of Sponge Iron Cluster

Authors: Venkata Ramanaiah, Shabina Khanam

PII: S0263-8762(18)30120-5
DOI: https://doi.org/10.1016/j.cherd.2018.03.014
Reference: CHERD 3084

To appear in:

Received date: 24-7-2017

Revised date: 10-3-2018
Accepted date: 12-3-2018

Please cite this article as: Ramanaiah, Venkata, Khanam, Shabina, Modified
Approach of Total Site Integration for Energy Conservation: A Case
Study of Sponge Iron Cluster.Chemical Engineering Research and Design
https://doi.org/10.1016/j.cherd.2018.03.014

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

Venkata Ramanaiah* and Shabina Khanam

Indian Institute of Technology Roorkee, Roorkee-247665, Uttarakhand, India

*
Corresponding author: Email: bvramanaiah2000@yahoo.com, Tel.: +919493807763;
Fax: +91-1332-276535

Abstract
In the present work, a modified approach to conserve energy in total site of plants of similar

T
type is developed where conventional methods are not applicable. The approach includes
iterative method, if utility and process streams are same, as well as non-iterative method, if

IP
these are different. Further, if coal is treated as utility and process streams, a revised model to
compute its consumption is also proposed. To illustrate the new approach a cluster of three

R
sponge iron plants are considered where two different strategies are proposed. Results show
that reduction in coal consumption, predicted for Strategy-1, is higher than that of Strategy-2.

SC
Strategy-1 recovers 99.8% of waste heat available in the modified total site. Thus, through
Strategy-1 total amount of energy wasted in the cluster is reduced from 43% to 7.6%. Along
with this, Strategy-1 reduces waste gas emissions significantly while making sponge iron
cluster more environment friendly. Moreover, 97.9% reduction of water in total site is also
U
observed using Strategy-1, which is an added advantage. Results are compared well with that
of the published literature. Further, this approach can be applicable effectively in the site of
N
similar plants irrespective of the operating condition.
A
Graphical Abstract
M
ED
E PT
CC
A

Key words
Energy conservation; Total site integration; Iterative approach; Sponge iron cluster

1
Research Highlights

 Modified approach to conserve energy in a total site of similar plants is developed

 To illustrate the new approach a site of three sponge iron plants is considered
 Two strategies are proposed for utilization of waste heat in cluster and compared
 Most suitable strategy is selected based on %heat recovery and economic analysis

T
R IP
Nomenclature Abbreviations

SC
A Heat transfer area (m2) ABC After Burning Chamber
CP mCp, kW/ᴼC CH Chimney
h Hour DRI Direct Reduced Iron
HX
m
Heat Exchanger
Mass flow rate, kg/h
DSC
EC U Dust Settling Chamber
Evaporating Cooler
N
P Pressure, atm ESP Electrostatic Precipitator
Q Heat load, kW GCC Grand Composite Curve
A
T Temperature HEN Heat Exchanger Network
t Tonne (=1000 kg) HI Heat Integration
M

U Overall heat transfer NHV Net heating value

coefficient (W/m2-ᴼC)/ Utility factor
Subscript PFD Process Flow Diagram
ED

a Air RC Rotary Cooler

c Coal RK Rotary Kiln
de Decomposition SI Sponge Iron
f Feed/Factor SSSP Site Sink Source Profiles
PT

hu Hot utility SUGCC Site Utility Grand Composite Curve

loss Heat losses TAC Total Annual Cost
m Moisture TS Total Site
E

min Minimum TSHI Total Site Heat Integration

CC

p Process TSI Total Site Integration

S Supply Total Site Problem Table
TS-PTA
Algorithm
T Target TSST Total Site Sensitivity Table
A

v-char Volatiles in char WHRB Waste Heat Recovery Boiler

v-coal Volatiles in coal WS Wet scraper

2
1. Introduction
Energy conservation in a process plant is the primary demand of present days. To carry out it,
process integration principles are used successfully since many decades (Klemeš et al., 2013;
Klemeš and Kravanja, 2013). However, instead of a single plant, energy conservation in the
whole site of plants is more profitable, especially in cases where plants are located nearby, as
these sites consider energy conservation opportunities available in all individual plants
collectively. To conserve energy in such clusters of plants simultaneously a concept of total
site (TS) integration is used effectively.

Dhole and Linnhoff (1993) introduced the concept of TS integration where they developed a
graphical method of site profiles, which were analogous to composite curves of individual

T
processes. Klemeš et al. (1997) extended this approach. Raissi (1994) developed analytical
tools that helped in understanding interactions between site fuel, heat recovery and co-

IP
generation. Further, Hui and Ahmad (1994) developed a graphical based procedure for
optimizing the overall site utility system. Goršek et al. (2006) presented modified site sink

R
source profiles (SSSP) and designed cogeneration system in an existing site of specialty
chemicals. Bandyopadhyay et al. (2010) proposed a simplified and novel methodology for

SC
targeting cogeneration potential based on the Salisbury approximation. However, Manesh et
al. (2013) also determined cogeneration potential of TS utility system considering shaft work
and degree of superheat. For integration of renewable energy with variable supply and
demand in industrial sites a graphical targeting methodology (Varbanov and Klemeš, 2011)
U
and numerical algorithmic (Liew et al., 2014b) were proposed. Varbanov et al. (2012)
extended the TS methodology considering individual ΔTmin for each process and utilities
N
whereas, Fodor et al. (2012) considered stream specific ΔTmin inside each process while
setting different ΔT contribution (ΔTcont) and also using different ΔTcont between process
A
streams and utility systems. Liew et al. (2013) extended Total Site Sensitivity Table to
characterize effects of plant shutdown whereas, Liew et al. (2014c) considered water sensible
M

heat to target the TS utility. Chew et al. (2013) listed numerous issues that can significantly
affect integration of TS. Chew et al. (2014; 2015a) developed a Total Site Heat Integration
(TSHI) methodology, which was extended by Chew et al. (2015b) for incorporating
ED

horizontal pressure drop in the steam network. Pouransari et al. (2014) performed site-scale
process integration on a large chemical plant having three different processes. Nemet et al.
(2014) developed a mathematical model for TS integration considering heat losses, piping
PT

cost and capital. Wang et al. (2014) proposed a graphical methodology for determining
energy target of interplant heat integration with different connection patterns. Further, Wang
et al. (2015) developed a strategy that combined direct and indirect interplant heat integration
to reduce total annual cost (TAC). Liew et al. (2017) reviewed developments of heat
E

integration and TSHI until 2016. Tarighaleslami et al. (2017) focused on unifying TSHI for
CC

multiple isothermal and non-isothermal utility targeting.

Hackl et al. (2011) applied the concept of TS to a cluster of five chemical companies and
showed that 50% of savings could be achieved by moderate changes to the existing heat
A

exchanger system whereas, 92% of changes were technically feasible. Matsuda et al. (2009)
applied TS integration concept to Kashima industrial area in Japan that resulted 53% of
energy savings. Matsuda et al. (2012) targeted energy saving potential in a large steel plant
by total site based pinch analysis. An effective TS retrofit framework was proposed and
tested on a petrochemical plant that resulted significant energy savings (Liew et al., 2014a).
Use of a Heat Recovery Loop (HRL) for site-wide heat integration of a typical large dairy
factory with multiple plants on site was investigated by Atkins et al. (2012). Boldyryev and
Varbanov (2015) determined the optimal minimum temperature approach of bromine and
sodium bromide production and specified energy targets.
3
It appears from above discussion that for energy conservation industrial cluster/site
considered includes plants of different process such as heavy chemical complexes (Matsuda
et al., 2009). However, in a number of cases, such clusters are of same process such as
sponge iron, sugar, etc. As all these plants are similar, there is no relative advantage of
conventional TS integration, which is more suitable to integrate heat within the site of
different processes. The traditional approach of generating SSSP is invalid in case of process
plants of similar type where same utility is used in all plants. This is explained in Appendix.
Further, it is observed that in many processes coal is used as utility stream where it is heated
up to reaction temperature and then provides heat through combustion. Thus, for such process
separation of process/utility data is difficult, which makes integration of the utility network
infeasible through conventional TS integration concept. With this backdrop, the present study

T
focuses on energy integration in TS of plants of similar type through a modified approach.

IP
In the present work, the proposed modified approach for TS integration is illustrated through
a case of sponge iron process cluster. Sponge iron, also known as Direct Reduced Iron (DRI),

R
is used widely in steel making process. Today, India is the largest producer of sponge iron in
the world while contributing 25% to world DRI production due to adequate availability of

SC
coal deposits (Midrex, 2015). Sponge iron plants are highly energy intensive. Jena et al.
(1996) observed that about 33% of heat generated in the kiln is lost with waste gas and
authors suggested a waste heat recovery system to utilize this heat. Similar facts were also
drawn by Dey et al. (2015). Loganathan and Sivakumar (2013) developed a waste heat
U
recovery steam generator and saved approximately 50.2 paisa per kg of steam generation as
compared to fired boiler.
N
Prasad et al. (2011a) considered a conventional coal based sponge iron plant of India. They
A
developed a model for coal consumption and proposed design modifications based on process
integration. It reduced coal and water consumptions by 12.5% and 93.4%, respectively.
M

Further, they eliminated use of evaporating chamber (Prasad et al., 2011b). Prasad et al.
(2013) designed a gas carrying duct and rotary drier for utilizing waste heat in sponge iron
plant, which reduced kiln load by 13.15%. Sahu et al. (2013) designed waste heat recovery
ED

system and gas-gas plate heat exchanger to integrate heat of stack gas in the process. Kumar
and Khanam (2012) proposed water bath around rotary cooler to recover heat from kiln outlet
and gas-liquid finned double pipe exchanger for heat utilization. These studies considered
PT

single plant for energy integration. However, in many states of India there are hubs of such
plants, which are placed nearby in the mineral belts of India (Sarangi and Sarangi, 2011).

Many researchers carried out extensive study on sponge iron processes outside India also.
E

Most of these studies are on gas based processes. For coal based processes, a few
CC

experimental and theoretical studies were conducted. Coetsee et al. (2002) found that heat
transfer is not the rate-limiting step in reduction in magnetite-coal pellets. Liu et al. (2004)
investigated reduction behavior of iron ore using Australian coal and Zuo et al. (2013) using
biomass char. Man et al. (2014b) studied mass loss and direct reduction characteristics of iron
A

ore-coal composite pellets. Man et al. (2014a) reported that reduction process was diffusion
controlled below 900ᴼC; however, at higher temperatures up to 1100ᴼC, it was phase-
boundary controlled. All these studies are for iron ore reduction of coal-ore mixtures and its
kinetics. However, it appears that outside India no study is available on energy conservation
of this process.

Further, it is observed that world’s primary energy source comes from fossil fuels such as
coal, oil, natural gas, etc. and its combustion gives major source of global CO2 and other
greenhouse gas emissions. Largest CO2 emissions occur due to coal combustion. Recent
4
studies reveal that overall CO2 emission of the world in 2016 is 32.1 GtCO2 and after China
and USA, India is the top CO2 emitting country in the world (IEA, 2017). Iron and steel
industry is most energy intensive and highest CO2 emitting industry while contributing 27%
of total CO2 emissions produced from global manufacturing sector (Hasanbeigi et al., 2016).
As India is the largest producer of iron, there is a need to pay attention in this plant for
reducing energy consumption and thus, CO2 emissions. Hence, it is thought logical to
illustrate the proposed modified approach of TS integration through sponge iron cluster. For
this purpose, three typical sponge iron plants of different production capacities are
considered, which are located nearby in a state of India.

2. Modified approach for TS integration

The modified approach for TS of similar plants can be better understood while comparing it

T
with that proposed for general problems of process integration. Conventionally TS
Integration is carried out through following steps (Klemeš et al., 1997) (Varbanov et al.,

IP
2012) (Fodor et al., 2012) (Tarighaleslami et al., 2017):

R
Step-1: Data extraction of all individual processes in TS
Step-2: Preparation of stream tables and selection of ΔTmin

SC
Step-3: Energy targeting using PTA for individual processes
Step-4: Plotting GCC for each process and remove pockets
Step-5: Construction of TS profiles
Step-6: Selection of utility data and matching utility generation and consumption targets
Srep-7: Plotting Site Utility Grand Composite Curve
Step-8: Targeting shaft work and cogeneration U
N
Step-9: Designing heat exchanger network (HEN) of TS
Step-10: Modification of TS process flow diagram (PFD)
A

However, steps involved in the proposed approach are:

M

Step-1: Define a strategy for energy conservation

Step-2: Data extraction for all processes together in TS according to the strategy
Step-3: Preparation of stream table considering all processes together and selection of ΔTmin
ED

Step-4: Utility targeting of TS

Step-5: Designing HEN of TS
Step-6: Modification of TS PFD
PT

The detailed computation methodology for the modified approach is presented in Figure 1.
For a process, if utility stream such as coal, coke, char, etc. is one of the feed, stream data for
proposed strategy vary when utility consumption changes. Hence, an iterative approach
E

should be employed, as shown in Figure 1. It requires a model to compute utility

CC

consumption in TS as described below considering coal as utility:

2.1 Model development

Coal consumption depends on heat requirement of air and feed materials (Qa and Qf) for
A

preheating, heat involved in reactions (Qp), heat losses (Qloss), preheating required by coal
itself (Qc) and latent heat required for evaporation of moisture of feed material (Qm). Based
on these factors coal consumption is found using model proposed by Prasad et al. (2011a).
However, they fail to account heat associated with decomposition of sulphur removing
component (primarily dolomite) and vaporization of coal volatiles. These factors, although
minor, not only change amount of coal consumption but also the waste gas. Considering all
these factors a modified model is developed as:
Qhu + Qp + Qf + Qa + Qc + Qm + Qloss + Qde + Qv-coal - Qv-char = mc (NHVUf) (1)

5
 Where, hot utility requirement (Qhu) is computed using pinch analysis (Linnhoff et al., 1982).
Further, heat associated with decomposition of dolomite (Qde) and vaporization of volatile
matter (Qv-Coal and Qv-Char) can be found as:
Heat of dolomite decomposition (Qde)
Dolomite decomposes into oxides and Qde is the heat of reaction.
Heat of vaporization of coal volatile matter (Qv-Coal and Qv-Char)
Amount of volatiles vaporized is computed while subtracting volatiles available with product
i.e. char from total volatiles present in coal. Considering heat of coal devolatization as 122
kCal/kg (Jena et al., 1996) QV-Coal and QV-Char are predicted.
In Eq. 1 coal utility factor (Uf) is also required, which depends on equipment conditions and
quality of coal. It accounts a fact that only a fraction of total energy, released from coal
combustion, is used in the process. Uf is computed using Eq. 1 at existing plant conditions.

T
Further, net heating value (NHV) of coal is considered as 22,930.5 kJ/kg (Jena et al., 1996) in
Eq. 1.

R IP
Start Compute minimum hot and cold utility
requirements using Pinch analysis. Name

SC
them as (Qhu)old & (Qcu)old
Collect data from different plants
including flow diagrams (PFD)

U Compute the utility consumption

using hot utility (Qhu)old and compute
N
(Air)new from air to utility ratio
Physical properties of
raw materials &
A
product streams (Air)old=(Air)new
Is
M

(Air)new = (Air)old?
Carry out material and energy No
balance for each plant and
ED

prepare PFDs with these values Yes

Revise the stream table data considering

new values of utility and air. Compute
PT

Define energy minimum hot and cold utility requirements

conservation strategy using Pinch analysis. Name it as (Qhu)new
E

(Qhu)old=(Qhu)new
According to strategy extract data
CC

Is
of all processes together from PFDs No
(Qhu)new = (Qhu)old?
and prepare stream tables
A

Yes

Are
process and utility
streams different?
Compute minimum hot and Design HEN following
Yes
cold utility requirements the guidelines proposed
using Pinch analysis in section 3 (step 5)

No
Design HEN following Design the modified PFD of total
pinch design rules site following the guidelines
Compute the ratio of air to utility
proposed in section 3 (step 6)
from the existing plant data and 6
name air requirement as (Air)old

Carry out economic analysis

3. Illustration of modified approach with a case study
Actual data of three sponge iron industries such as Plant-1, Plant-2 and Plant-3 being
operated with different capacities in an industrial cluster of India are considered to illustrate

T
the modified approach for energy conservation in TS. The site layout of sponge iron cluster,
with distances between equipment within the same plant and also between different plants, is

IP
shown in Figure 2. All these three plants are SL/RN (Stelco-Lurgi/Republic Steel-National
Lead) process based, which is described in detail by Prasad et al. (2011a). Per day production

R
capacities of Plant-1, Plant-2 and Plant-3 are 100t, 62t and 100t, respectively. Figure 2 shows
that Plant-3, unlike Plant-1 and Plant-2, involves waste heat recovery boiler (WHRB) instead

SC
of Evaporating Cooler (EC).

Step-1: Define a strategy for energy conservation

General guidelines to define a strategy are:
U
(a) Based on PFDs of plants recovering areas should be identified where heat is lost. In the
similar line, utilizing areas are also found in plants where heat available in recovering areas
N
can be transferred.
(b) Using recovering areas different hot streams are identified whereas, cold streams are
A
found based on utilizing areas.
(c) Based on combination of recovering (hot streams) and utilizing (cold streams) areas a
M

strategy can be defined. For maximum heat recovery one should consider all possible
recovering and utilizing areas while defining a strategy.
(d) For defined strategy all associated hot and cold streams of plants are collected with its
ED

supply and target temperatures, mass flow rates and specific heat capacities, etc. and named
these as stream data.
(e) For a strategy, %heat recovery can be preliminarily determined using stream data to find
PT

the efficacy of the strategy.

As mentioned in above guidelines, heat recovering areas identified in plants are: (1) Waste
gas exiting ESP at 220ᴼC is released to atmosphere via chimney. (2) Hot sponge iron is being
E

cooled from 1020ᴼC to 110ᴼC in RC. For this purpose, water is used through an indirect heat
CC

exchanger and vapor generated from it goes directly to atmosphere. (3) Waste gas coming
from ABC at very high temperature i.e. 1050ᴼC, 850ᴼC and 950ᴼC for Plant-1, Plant-2 and
Plant-3, respectively, is cooled to around 250ᴼC in EC using water. Similarly, heat utilizing
areas are identified as: Feed materials like (1) kiln feed (2) slinger coal (3) kiln air enters the
A

kiln at ambient temperature, which are to be preheated up to reaction temperature before

reduction process. This preheating could be done using heat available in recovering areas.

Considering these working areas two different energy conservation strategies, Strategy-1 and
Strategy-2, are proposed. The proposed strategies should be able to recover maximum
amount of energy from these areas. Considering this aspect, Strategy-1 is proposed in the
present work. However, Strategy-2 is identified to carry out comparative analysis only. These
strategies vary in terms of modification and investment.

7
For Strategy-1 waste gas exits from ABC of Plant-1 is used for preheating the kiln feed and
slinger coal of all three plants, individually, from 30ᴼC to 300ᴼC. Kiln outlet stream is cooled
from 1020ᴼC to 110ᴼC using kiln air of its respective plant. Further, kiln air is preheated
using waste gas to maximum possible temperature (421ᴼC in this case) such that ΔTmin equal
to 50ᴼC is maintained. In addition to it, waste gas streams from ABC exits of Plant-2 and
Plant-3 are combined and used for power generation. The partial schematic of Strategy-1 is
shown in Figure S1 in supplementary data.

T
IP
Waste gas 13m 12m 40m 5m 20m

R
44.17 t/h
Water: Kiln Feed: 30C Air: 30C Slinger coal:
12.54 t/h Iron ore: 6 t/h 24.80 t/h 2.2 t/h, 30C

SC
Waste gas:
30C Feed Coal: 3.16 t/h Water: 30C
33.18 t/h
Dolomite: 0.2 t/h

4.49 t/h
213.63 t/h

2.33 t/h
Waste gas 1050C
CH

45.72 t/h Flue gas: 17.98 t/h

30.97 t/h At 110C
250C EC
ABC
900C
U
Rotary Kiln
Kiln out
RC
SI: 4.17 t/h
Char: 0.55 t/h
N
5.40 t/h Ash: 0.61 t/h
DSC 1020C Lime: 0.07 t/h
ESP Air:
A
2.92 t/h Water: 34.74C
Waste gas: 30C 213.63 t/h
220C
M

Dust
1.55 t/h WS

Dust: 0.7 t/h Plant-1

ED

75 m 75 m

75 m
Lime: 0.07 t/h
Lime: 0.07 t/h

Char: 1.17 t/h

PT Char: 0.55 t/h

Ash: 0.30 t/h

Ash: 0.86 t/h

SI: 2.60 t/h

SI: 4.17 t/h

At 110C
At 110C

Water: 34.74C
Water: 34.74C
Water: 30C

Water: 30C
251.38 t/h

193.48 t/h
E

193.48 t/h
251.38 t/h
20m

20m

2.6 t/h, 30C

Slinger coal:

Slinger coal:
RC

RC
CC 2 t/h, 30C

Kiln out

Kiln out
1020C
5.65 t/h

1020C
4.14 t/h

4.08 t/h 4.08 t/h

5m

5m

6.7 t/h 5.83 t/h

Air: 30C

Air: 30C
37.60 t/h

33.23 t/h
A
26.82 t/h

23.32 t/h
Rotary Kiln

Rotary Kiln
40m

42m

Feed Coal: 3.21 t/h

Dolomite: 0.12 t/h

Dolomite: 0.12 t/h

Feed Coal: 4.5 t/h
Kiln Feed: 30C

Kiln Feed: 30C

Iron ore: 6.5 t/h

Iron ore: 5.4 t/h

Dust: 0.7 t/h

Dust: 0.5 t/h

Flue gas:

Flue gas:
45.07 t/h

40.43 t/h
900C

750C
DSC

DSC
12m

WS

WS
12m

8
ABC
ABC
aste gas:

aste gas:
7.75 t/h

.59 t/h
.38 t/h

.66 t/h
50C

0C
0C

0C
ir:

ir:
 T
IP
In Strategy-2 waste gas streams exiting ESPs of Plant-1, Plant-2 and Plant-3 are mixed and
used as hot stream to preheat kiln feed and slinger coal of all three processes at Plant-1 from
30ᴼC to 150ᴼC. Then preheated kiln feed and slinger coal is carried from Plant-1 to respective

R
plants. Further, kiln outlet stream is cooled from 1020ᴼC to 110ᴼC using kiln air of its
respective plant. It is noted from Figure 2 that amongst three plants in TS only Plant-3 has a

SC
unit of WHRB. Thus, Strategy-2 proposes WHRBs, for power generation, in Plant-1 and
Plant-2 using waste gas exiting from ABC of respective plants. Similar to Figure S1, in
supplementary data, the partial schematic can also be prepared for Strategy-2. Thus, all

U
recovering and utilizing areas are considered in Strategy-1 and Strategy-2.
N
Step-2 and Step-3: Data extraction, preparation of stream tables and selection of ΔTmin
Stream data for both strategies are shown in Table 1, which includes data of all three plants
A
together. Properties of ore and coal, shown in Table 1, are constant and taken from literature
(Green and Perry, 2008; Gronvold and Samuelsen, 1975) whereas that of air and waste gas
M

are computed at respective average temperatures

(“http://www.peacesoftware.de/einigewerte/luft_e.html,” n.d.). The number shown as a
subscript in Table 1 denotes the specific plant in TS. For example, kiln outlet of Plant-1 is
ED

denoted as Kiln outlet1. Similarly, combined ABC exit waste gas streams of Plant-2 and
Plant-3 is denoted as ABC exit gas23 and so other streams. In Strategy-1 total 13,850.4kW
heat is required to preheat kiln inlet streams i.e. kiln feed, slinger coal and kiln air of all
plants, to target temperatures as can be seen through Table 1. Here, 2,770.6kW heat is
PT

provided by kiln outlet streams and remaining 11,079.8kW is supplied through ABC exit gas1
stream. Consequently, target temperature of ABC exit gas1 stream is found as 43.2ᴼC as
shown in Table 1. In the similar line target temperatures of streams are found and reported in
E

Table 1.
CC

Utility requirements for both strategies are targeted using pinch analysis (Linnhoff et al.,
1982) where waste gas is used either as hot stream or cold stream. Due to lower heat transfer
coefficient of waste gas it requires substantially large heat transfer area, which can be
A

somewhat compensated by increasing temperature difference between waste gas and other
streams. Thus, large ΔT i.e. 50ᴼC is preferred for the present study (Linnhoff et al., 1982).

Step-4: Utility targeting for TS

For Strategy-1 hot and cold utilities are found as zero and 23,617.2kW, respectively, using
pinch analysis (Linnhoff et al., 1982) whereas, these are zero for Strategy-2. It shows that in
case of Strategy-2 the complete problem is in energy balance.

9
For all existing plants, total coal consumption is 17.67t/h whereas, kiln air is 95.6t/h as shown
in Table 2. Requirements of coal and air are to be estimated for proposed strategies. As
discussed in Section 2 that if utility and process streams are same, which is the present case,
iterative approach is required to compute amounts of coal and air. Thus, calculations of all
iterations are carried out using proposed model for coal consumption and the computation
methodology shown in Figure 1. Final amounts of coal and air, in Strategy-1, are 10.9t/h and
59t/h, respectively, as shown in Table 2. Through Strategy-1 38.3% of coal requirement is
reduced in comparison to the existing process. Similar iterations are carried out for Strategy-2
where final coal and kiln air requirements are 13.8t/h and 74.9t/h, respectively. Strategy-2
reduces 21.7% of coal in comparison to the existing process.

T
IP
Table 1 Stream data of proposed strategies

R
Stream no.
Stream Mass Flow CP

SC
Stream used in Ts (ᴼC) Tt (ᴼC)
Type (t/h) (kW/ᴼC)
HEN
Strategy-1
Kiln outlet1
Kiln outlet2
Hot
Hot
H1
H2
1020
1020
U 110
110
5.4
4.1
1.05
0.87
N
Kiln outlet3 Hot H3 1020 110 5.7 1.12
ABC exit gas23 Hot H4 901 120 92.3 30.24
A
ABC exit gas1 Hot H5 1050 43.2 33.2 11.01
Kiln air1 Cold C1 30 421 24.8 7.10
M

Kiln air2 Cold C2 30 421 33.2 9.51

Kiln air3 Cold C3 30 421 37.6 10.76
Slinger coal1 Cold C4 30 300 2.2 0.84
ED

Slinger coal2 Cold C5 30 300 2.6 0.10

Slinger coal3 Cold C6 30 300 2.0 0.77
Kin feed1 Cold C7 30 300 9.4 2.87
PT

Kin feed2 Cold C8 30 300 8.7 2.71

Kin feed3 Cold C9 30 300 11.1 3.49
Strategy-2
E

Kiln outlet1 Hot H1 1020 110 5.4 1.05

Kiln outlet2 Hot H2 1020 110 4.1 0.87
CC

Kiln outlet3 Hot H3 1020 110 5.7 1.12

ESP exit gas123 Hot H4 220 189.2 151.7 45.54
ABC exit gas1 Hot H5 1050 250 33.2 11.27
ABC exit gas2 Hot H6 850 240 44.6 14.79
A

Kiln feed123 Cold C1 30 150 29.2 9.07

Slinger coal123 Cold C2 30 150 6.8 2.61
Kiln air1 Cold C3 30 166.6 24.8 6.98
Kiln air2 Cold C4 30 115.2 33.2 9.33
Kiln air3 Cold C5 30 126.8 37.6 10.56

Table 2 Iteration results for Strategy-1 and Strategy-2

10
 Strategy-1 Strategy-2
Iteration Coal Kiln air Coal Kiln air
no. consumption requirement consumption requirement
(kg/h) (kg/h) (kg/h) (kg/h)
0 17,670 95,636.5 17,670 95,636.5
1 10,902.7 59,009.5 13,831.9 74,863.3
2 10,902.7 59,009.5 13,861.7 75,024.3
3 13,835.1 74,880.5
4 13,832.2 74,865.1
5 13,831.9 74,863.5
6 13,831.9 74,863.3

T
7 13,831.9 74,863.3

IP
Reduction in coal consumption is more for Strategy-1 in comparison to Strategy-2. This is
because in Strategy-1 kiln feed is preheated, outside the kiln, from 30ᴼC to 300ᴼC utilizing

R
2,448kW of waste heat whereas in Strategy-2 it is only from 30ᴼC to 150ᴼC considering
1,088kW of waste heat. Consequently, for Strategy-1 and Strategy-2 feed is heated from

SC
300ᴼC and 150ᴼC, respectively, to reaction temperature inside the kiln using coal. For this
purpose, Strategy-1 utilizes 1,820kg/h of coal; whereas, 2,199.8kg/h is consumed in Strategy-
2, which is 20.9% more. On the other hand, this value of coal consumption is 2,503.2kg/h in

U
the existing cluster. Similarly, other inlet streams i.e. slinger coal and kiln air require less coal
inside the kiln in Strategy-1 as compared to Strategy 2. In fact, in Strategy 1 total
N
13,850.4kW of waste heat is utilized to preheat the kiln inlet streams; whereas, in Strategy 2
it is only 4,171.3kW. Thus, coal consumption to heat these streams inside the kiln reduces
A
more in Strategy-1 than that in Strategy-2.
M

Step-5: Designing HEN of TS

As utility and process streams are same in the present case, usual pinch design rules are not
applicable. Thus, following guidelines are proposed for such cases, which are based on actual
ED

practice:
1. All streams should be considered as part of a single plant
2. Tmin should not be violated for any match
3. Solid streams should exchange heat with gas and liquid streams but not with solid.
PT

4. If options are available, exchanger should be placed between near most streams.

Based on the proposed guidelines HEN is designed for TS through Strategy-1 as shown in
E

Figure 3 where temperatures and heat loads are based on values of final iteration reported in
Table 2. Figure 3 includes twelve new heat exchangers in the modified design of TS, which
CC

do not violate ΔTmin. Amongst twelve heat exchangers nine are gas-solid heat exchangers and
three are gas-gas heat exchangers. In Strategy-1 power generation is carried out using waste
gases from ABC exits of Plant-2 and Plant-3. For this purpose, these two streams of waste
gas are mixed (Stream no. H4 in HEN) having CP 18.44kW/ᴼC. For producing steam first
A

water is preheated from 30ᴼC to 345ᴼC where saturated steam is generated at 155.5 bar and
then it is superheated up to 700ᴼC and 300 bar. In this process 14.93t/h superheated steam is
generated, which is expanded through steam turbine having 85% isentropic efficiency (Smith,
2016). Consequently, 5.19MW of power is produced, which is assumed to be exported. Once
superheated steam is generated, temperature of waste gas reduces to 120ᴼC. For this
modification 14.93t/h water is used, which is 705.94t/h in the existing site as shown in Figure
2. Hence, 97.9% less water is required in Strategy-1.

11
Similarly, HEN for Strategy-2 is designed and shown in Figure S2 in supplementary data. It
includes five heat exchangers, which maintain ΔTmin. Further, Strategy-2 proposes WHRBs,
for power generation, in Plant-1 and Plant-2 using waste gas exiting from ABC of respective
plants. Total power produced through heat of waste gas is 2.59 MW and 2.66 MW for Plant-1
and Plant-2, respectively. It requires 25.84t/h water, which is 96.4% less in comparison to the
existing site. Thus, these modifications eliminate water requirement in RC of all three plants
and EC in Plant-1 and Plant-2.

CP, kW/ᴼC
1020ᴼ

T
110ᴼ
1 1.05
H1

IP
1020ᴼ 110ᴼ
H2 2 0.87

1020ᴼ

R
110ᴼ
H3 3 1.12

SC
900.5ᴼ 120ᴼ
H4 WHR 18.42
596.7ᴼ 512.8ᴼ D2 D3B
4 5 6 1.67

1050ᴼ
H5
D1
D5
7
438.7ᴼ
8
371ᴼ
U D6
9
D7
125ᴼ
2.45
N
D1
D9 459.8ᴼ 411.8ᴼ D1 1
A
1 1 1 2.67
0
421ᴼ 0 1 2 30ᴼ
D4 247.7ᴼ
M

4 1 C1 4.38
758.9 kW 953.1 30ᴼ
421ᴼ
165.6ᴼ kWD8 2 5.87
7 C2
30ᴼ
ED

1498.6 kW 795.2
421ᴼ kW
184ᴼ D1
1 3 C3 6.64
2
0 kW
1573.2 1022.4 30ᴼ
300ᴼ kW
5 C4 0.52
PT

140.5 30ᴼ
300ᴼ
kW C5 0.62
8
30ᴼ
E

300ᴼ 166 kW
1 C6 0.47
30ᴼ
CC

300ᴼ 1
127.7
kW 6 2.41
C7
649.2 30ᴼ
300ᴼ
kW
9 C8 2.23
A

603.1 30ᴼ
300ᴼ
kW 1 2.83
C9
2
764.6
Figure 3. Heat Exchanger Network for kW
Strategy-1

12
Step-6: Modification of TS PFD
As distance between plants is larger than that within a plant, following guidelines are
proposed for movement of streams within TS:
1. Liquid stream can be transported through pipes of different diameter. Pressure drop in each
pipe should be computed to propose the required number of pumps
2. Gas stream can be transported through ducts where pressure drop in each duct is to be
found to propose number of forced draft (FD) fans
3. Solid stream should be transported through conveyors either open or closed.
4. For exchange of heat between gas/liquid and solid, preference should be given to transport
gas and liquid streams over solid.

T
Based on these guidelines the modified PFD of TS for Strategy-1 is proposed as shown in
Figure 4. In this strategy, as in most of exchangers heat transfer occurs between gas and solid

IP
streams, ducts and conveyors are used to transport. However, following guideline no. 4,
movement of gas is preferred over solid and so, ducts are employed. Thus, Figure 4 includes

R
total sixteen ducts for Strategy-1, three for carrying kiln air and thirteen for waste gas to
different places before it is directed to the chimney. As distance between ABC of Plant-1 and

SC
kiln outlet of the same plant is 52m, as shown in Figure 2, an insulated duct, D1, of same
length is employed to transfer first branch of waste gas, having CP of 1.67kW/ᴼC, from ABC
to kiln outlet of Plant-1. While passing through this duct, temperature of waste gas drops
from 1050ᴼC to 1038.72ᴼC as shown in Figure 4. At this temperature waste gas enters to a
U
gas-gas heat exchanger, HX-4 shown in Figure 3 and Figure 4, where kiln air of Plant-1 is
preheated while exchanging heat of 758.9kW. The methodology for computing temperature
N
profile and pressure drops inside a duct is adopted from the work of Prasad et al. (2013) and
Prasad (2010), respectively. Details of all sixteen ducts are shown in Table S1 in
A
supplementary data. Mild steel is used for casing of the duct and for insulation ceramic fibre
blanket and glass wool, at higher and lower temperatures, respectively, are placed at outer
M

surface of the duct. FD fans are used to compensate pressure drops in ducts where one fan
can sustain 1.26 atm pressure drop (“www.pewi-fans.com/industrial-blowers.html,” n.d.).
Accordingly, sixteen FD fans are required, one for each duct.
ED

Similarly, Strategy-2 requires eleven numbers of ducts, eight for carrying waste gas streams
from exits of ESP to other sections of plants and three for carrying kiln air to kiln outlet in the
PT

individual plant. Details of these are also shown in Table S1 in supplementary data. The
modified PFD of TS for Strategy-2 can be drawn in the similar line as prepared for Strategy-
1.
E

4. Comparative analysis of two Strategies

CC

Results of Strategy-1 and Strategy-2 for TS integration of sponge iron plants are compared
with the existing system through Table 3. It indicates that both strategies offer better results
over the existing system. Table 3 shows that Strategy-1 and Strategy-2 consume 38.3% and
21.7%, respectively, less coal. Reasons of difference in savings in coal consumptions are
A

already explained in Section 3, Step-4. After final design it is found that Strategy-1 recovers
99.8% of waste heat available in the modified site whereas, Strategy-2 recovers 86%. It is
computed as: total heat that can be possibly recovered in the existing site is 25,602.7kW and
Strategy-1 is able to recover it completely whereas Strategy-2 recovers only 22,211.5kW.
After energy conservation through Strategy-1, total heat that can be recovered from modified
site is reduced to 15,481.9kW, out of which, 15,448.2kW (99.8%) of heat is recovered.
Similarly, through Strategy-2 total recoverable heat is reduced to 20,768.1kW where
17,890.2kW (86%) is recovered. Also, through Strategy-1 overall amount of energy wasted in

13
the cluster, in terms of coal consumption, is reduced from 43% to 7.6%. As more % of heat is
recovered the requirement of coal is reduced more in Strategy-1 than in Strategy-2.
Results further indicate that both strategies generate less waste gas in comparison to the
existing system; however, Strategy-1 produces lesser than Strategy-2 due to less coal
consumption. Thus, Strategy-1 gives more environment friendly sponge iron cluster.

Table 3 shows that TAC is 14% less for Strategy-1 as compared to Strategy-2. Both
Strategies offer huge profits. Though payback period is slightly more for Strategy-1, it offers
more profit than that of Strategy-2. Thus, based on energy consumption, water requirement,
waste gas generation, TAC and profit, Strategy-1 is selected as best heat recovery option.
Detailed economic analysis for Strategy-1 is shown in Appendix S3 in supplementary data.

T
Capital cost of gas-gas shell and tube heat exchangers and gas-solid exchangers are computed
using correlation proposed by Shenoy (1995) and the method shown in the work of Prasad

IP
(2010), respectively. Further, cost of ducts and FD fans are computed as given by Prasad
(2010). As these costs are older, cost index factors are used to predict recent costs of these
components. Cost index factors (Vavatuk, 2002) (“http://www.chemengonline.com,”) in

R
1995, 2010 and 2017 are 381.1, 550.8 and 566.6, respectively. Considering these factors,

SC
total capital cost is found as 7.6 $million and profit as 8.14 $million/year instead of 6.95
$million and 7.88 $million/year as reported in Table 3. The payback period thus found is 11.4
months instead of 10.7 months for Strategy-1.

U
In the present work, though only two strategies are proposed, Strategy-1 shows maximum
heat recovery. However, Strategy-2 is chosen to perform the comparative analysis. Though
N
both strategies are first defined and then whole analysis is carried out, it could be possible
that better strategies are missed if the selection of strategies is not done systematically. This
A
can be avoided while identifying strategies using guidelines proposed in Section 3, Step-1
and computing % recovery of heat, for each strategy. After final design it was found that
M

Strategy-1 recovers 99.8% of waste heat available whereas, Strategy 2 recovers only 86%.
Though other strategies can also be proposed as shown in Table S2, it is difficult to recover
more than 99.8% of heat through any other strategies. Thus, there is no chance to miss the
ED

best possible strategy if % recovery of heat is computed.

5. Comparative analysis of results of Strategy-1 with that of published literature

PT

As conventional method, developed for TS integration, is used to handle clusters of different

type plants only, these methods cannot be used directly for energy integration in site of
similar type plants. Further, it appears from literature that site integration of sponge iron
cluster is not carried out. Thus, Strategy-1 cannot be compared with the available literature.
E

However, a few investigators, developed energy conservation scenarios for sponge iron
CC

plants of different capacities. Results of these studies can be compared with that of Strategy-1
on the basis of %reduction in coal, water and waste gas and profit per ton of sponge iron to
observe benefits of TS integration. Kumar and Khanam (2012) proposed preheating of kiln
air to 80ᴼC by kiln outlet using water bath. Prasad et al. (2011a; 2011b) proposed preheating
A

of feed materials to kiln and air using waste gas exiting ABC whereas, Prasad et al. (2011a)
also proposed recovery of heat from kiln outlet by cooling it from 1020ᴼC to 110ᴼC using
waste gas exiting ESP.

Advantages of Strategy-1 over the published literature are shown graphically in Figure 5.
Figure 5(a) shows that % of coal and water saved by TS integration through Strategy-1 is
more than that achieved by various design modifications reported in the literature. Maximum
reduction of waste gas generation is found as 27.4% in the published literature (Prasad et al.,

14
2011b) whereas, Strategy-1 decreased it to 51.7%. Reasons for such variation can be better
understood through following points:
1. To recover maximum possible waste heat the present work considers all heat recovering
and utilizing areas mentioned in Section 3, Step-1. Though Prasad et al. (2011a) considered
all areas, they were able to recover only 33.4% of waste heat that was available in the plant as
shown in Figure 5(a). However, Kumar and Khanam (2012) and Prasad et al. (2011b) are
able to recover maximum 67% and 73%, respectively, of waste heat only.

2. For Strategy 1, waste heat available from ABC exit of one plant is completely used for
preheating kiln inlet streams of all plants; whereas, that of other two plants is used for power
generation simultaneously. This ensures maximum heat recovery in the site, which is not
possible in the case of single site integration considered in published literature.

T
3. The model developed in published literature, for coal utility factor and coal consumption,

IP
did not consider all factors such as preheating of dolomite, heat released in its decomposition
and also heat required to vaporize coal volatiles as explained in Section 2.1. The modified

R
model, used in the present work, further increased coal savings by 4.3%. Thus, the modified
model ensures better coal utility factor, which in turn results in better coal saving.

SC
As shown in Figure 5(b), Strategy-1 offers profit of 30,030$/year/tonne of sponge iron
whereas design modifications, proposed in published literature, offer minimum
2,046$/year/tonne of sponge iron (Kumar and Khanam, 2012) to maximum
U
12,677$/year/tonne of sponge iron (Prasad et al., 2011a) only. These huge profits through
Strategy-1 are found at the cost of high capital investment, which also includes the capital
N
required for power generation. In spite of higher capital requirement the payback period for
strategy-1 is only 10.7 months, which is comparable with that of Kumar and Khanam (2012)
A
and Prasad et al. (2011b) whereas, it is higher in comparison to Prasad et al. (2011a) as
shown in Figure 5(b). Thus, based on above discussion it is observed that TS integration
M

through Strategy-1 can be considered as the best design modification over that offered in the
published literature, which makes sponge iron cluster more environment friendly.
ED

6. Generalization of the results

In the present work the modified approach of TS integration is illustrated through a case
study of sponge iron cluster and thus, results found is applicable to these plants. However, a
PT

few data may change with capacity such as coal and water consumptions, heat duties of
exchangers, waste gas generation, operating and capital costs, profit and payback period, etc.
The limitation of the proposed approach is that it can be employed to the industrial cluster
where all available plants are of same process. However, these plants may have same process
E

and utility streams or different. Sponge iron, steel, cement, thermal power plants, etc. used
CC

coal as utility and so, in these plants process and utility streams are same. Thus, the approach
illustrated through sponge iron cluster, can be applicable effectively in these plants also
irrespective of variation in operating conditions. On the other hand, in plants such as oil
refineries, petrochemical, etc. process and utility streams are different. For such cases,
A

another approach, shown in Figure 1, is applicable. As it works on pinch analysis principles,

it is also employed effectively in different clusters. However, a few factors like design
parameters of new equipment, utility consumption, economic data, etc. may vary according to
operating conditions.

15
 ABC exit gas1-b: 7.30 t/h; 1050C

I
D5

R
Plant-1 ABC exit gas1-a: D1

SC
4.98 t/h Kiln Feed: Slinger coal:
8.15t/h 1.36t/h
1050C
D3 D2
Waste Gas HX6 HX5 HX4
1038.7 C

U
18.67 t/h
300C
ABC 300C

421 C
N
exit 900C
gas1 -c: D4
ABC Rotary Kiln

Chimney
7.93t/h

A
1050C
DSC Kiln Air:

M
106C Air:
15.30 t/h 247.7
3.38 t/h HX1
30C
30C
ED ESP Sponge iron:
Wet Scraper 5.4 t/h
110C
Dust: Dust:
1.55 t/h 0.7 t/h
D11
PT
D9
Plant-3 Kiln Feed: Slinger coal:
Kiln Feed: Slinger coal: Plant-2
1.23t/h 7.5t/h 1.60t/h
9.4t/h
Waste Gas: D7 D6
E

D10
Waste Gas 29.09t/h HX11 HX10 Waste Gas HX9 HX8 HX7
HX12
27.79 t/h 950C 26.77 t/h
CC

300C Waste Gas: 300C

300C

421 C
300C 27.17t/h
421 C
D8
900C 850C 750C
ABC Rotary Kiln D12 ABC Rotary Kiln

Chimney
Chimney

118.61C
118.62C
DSC DSC Kiln Air:
Kiln Air: Air:
Air: 184C 20.50 t/h 165.6
23.20 t/h 3.38 t/h HX2
3.38 t/h HX3 30C
30C 30C
30C ESP Sponge iron:
ESP Sponge iron:
5.65 t/h Wet Scraper 4.14 t/h
Wet Scraper
Water Steam 110C
110C Dust: Dust:
Dust: D15 D13
Dust: 0.7 t/h 0.4 t/h 0.5 t/h
1.3 t/h D14
WHRB
16
D16
Figure 4. Modified process flow diagram of Total Site through Strategy-1
 Table 3 Comparative analysis of two strategies

Resource
Consumption Operating Waste
Capital TAC Power (%) Profit Payback
( t/h) cost gas
Case cost ($million/ generated Heat ($million/ period
($million/ generated
($million) year) (MW) recovery year) (months)
Coal Water year) ( t/h)

Existing 17.7 705.9 - - - - 151.7 - - -

Strategy-1 10.9 14.9 4.15 6.95 4.85 5.19 73.2 99.8 7.88 10.73
Strategy-2 13.8 25.8 4.94 5.95 5.53 5.25 93.4 86.1 7.59 9.55

T
120
97.9 99.8

IP
100
80

R
51.7 Kumar and Khanam (2012)
% 60
Prasad et al. (2011b)

SC
38.3
40 Prasad et al. (2011a)
20 Strategy-1

0
Coal saved Water saved Waste gas Heat recovery U
N
generation
(a) reduced
A
35
30.03
M

30 26.48
25
ED

20 Kumar and Khanam (2012)

15 Prasad et al. (2011b)
10.73
Prasad et al. (2011a)
PT

10
Strategy-1
5
E

0
Capital investment Profit Payback period
CC

(0.001 $million / (0.001 $million/ (months)

tonne of SI) tonne of SI)
(b)
Figure 5. Comparative analysis of Strategy-1 with published literature
A

17
7. Conclusions
In this paper, a modified approach to conserve energy in TS of plants of similar type is
developed. Approaches discussed in the literature are applicable to the TS of different
plants. However, when these techniques are applied to the site of similar plants, these fail
to conserve energy in such sites. It gives motivation to develop the present approach.
Further, it addresses two issues: (1) utility and process streams in the process are same and
(2) utility and process streams are different. Thus, to develop an approach, which
addresses these issues in TS of similar plants, is the primary and scientific contribution of
the present work.

Based on results obtained, it can be concluded that through economic analysis best energy
conservation strategy can be selected. However, that can also be chosen at initial stage

T
while comparing %heat recovery of each strategy. Further, while conserving energy
through the present approach, more environment friendly network can be obtained as it

IP
reduces waste gas generation significantly. The modified approach, illustrated through
sponge iron cluster, can be applicable effectively in the site of similar plants irrespective

R
of the operating condition and thus, it is a generalized approach also.

SC
The present approach can further be improved while considering variations in physical
properties of different streams. Based on this approach, other commodities such as water,
separating agents, etc. can be conserved as it requires areas from where these commodities
can be recovered (as being wasted) and where these can be utilized. A mathematical model
U
to optimize TAC in sites of similar plants can be proposed and if options are available,
how to select best one, can be understood through the present approach.
N
A
M
ED
E PT
CC
A

18
References
Atkins, M.J., Walmsley, M.R.W., Neale, J.R., 2012. Process integration between individual
plants at a large dairy factory by the application of heat recovery loops and transient
stream analysis. J. Clean. Prod. 34, 21–28. doi:10.1016/j.jclepro.2012.01.026
Bandyopadhyay, S., Varghese, J., Bansal, V., 2010. Targeting for cogeneration potential
through total site integration. Appl. Therm. Eng. 30, 6–14.
doi:10.1016/j.applthermaleng.2009.03.007
Boldyryev, S., Varbanov, P.S., 2015. Low potential heat utilization of bromine plant via
integration on process and Total Site levels. Energy 90, 47–55.
doi:10.1016/j.energy.2015.05.071
Chew, K.H., Klemeš, J.J., Alwi, S.R.W., Manan, Z.A., 2015a. Process modifications to
maximise energy savings in total site heat integration. Appl. Therm. Eng. 78, 731–739.

T
doi:10.1016/j.applthermaleng.2014.04.044
Chew, K.H., Klemeš, J.J., Alwi, S.R.W., Manan, Z.A., 2014. Process modification of Total

IP
Site Heat Integration profile for capital cost reduction. Appl. Therm. Eng. 89.
doi:10.1016/j.applthermaleng.2015.02.064
Chew, K.H., Klemeš, J.J., Alwi, S.R.W., Manan, Z.A., 2013. Industrial implementation

R
issues of total site heat integration. Appl. Therm. Eng. 61, 17–25.

SC
doi:10.1016/j.applthermaleng.2013.03.014
Chew, K.H., Klemeš, J.J., Alwi, S.R.W., Manan, Z.A., Reverberi, A. Pietro, 2015b. Total site
heat integration considering pressure drops. Energies 8, 1114–1137.
doi:10.3390/en8021114
U
Coetsee, T., Pistorius, P.C., De Villiers, E.E., 2002. Rate-determining steps for reduction in
magnetite-coal pellets. Miner. Eng. 15, 919–929. doi:10.1016/S0892-6875(02)00120-6
N
Dey, N.R., Prasad, A.K., Singh, S.K., 2015. Energy survey of the coal based sponge iron
industry. Case Stud. Therm. Eng. 6, 1–15. doi:10.1016/j.csite.2015.04.001
A
Dhole, V.R., Linnhoff, B., 1993. Total site targets for fuel co-generation, emissions, and
cooling. Comput. Chem. Eng. 17, 101–109. doi:10.1016/0098-1354(93)80214-8
M

Fodor, Z., Klemeš, J.J., Varbanov, P.S., Walmsley, M.R.W., Atkins, M.J., Walmsley, T.G.,
2012. Total site targeting with stream specific minimum temperature difference. Chem.
Eng. Trans. 29, 409–414. doi:10.3303/CET1229069
ED

Goršek, A., Glavič, P., Bogataj, M., 2006. Design of the optimal total site heat recovery
system using SSSP approach. Chem. Eng. Process. Process Intensif. 45, 372–382.
doi:10.1016/j.cep.2005.10.003
Green, D.W., Perry, R.H., 2008. Perry’s Chemical Engineers’ Handbook, 8th ed. McGraw-
PT

Hill, New York.

Gronvold, F., Samuelsen, E.J., 1975. Heat capacity and thermodynamic properties of a-
Fe2O3 in the region 300-1050 K. antiferromagnetic transition. J. Phys. Chem. Solids 36,
E

249–256. doi:10.1016/0022-3697(75)90017-7
CC

Hackl, R., Andersson, E., Harvey, S., 2011. Targeting for energy efficiency and improved
energy collaboration between different companies using total site analysis (TSA).
Energy 36, 4609–4615. doi:10.1016/j.energy.2011.03.023
Hasanbeigi, A., Arens, M., Cardenas, J.C.R., Price, L., Triolo, R., 2016. Comparison of
A

carbon dioxide emissions intensity of steel production in China, Germany, Mexico, and
the United States. Resour. Conserv. Recycl. 113, 127–139.
doi:10.1016/j.resconrec.2016.06.008
http://www.chemengonline.com [WWW Document], n.d.
http://www.peacesoftware.de/einigewerte/luft_e.html [WWW Document], n.d.
Hui, C.-W., Ahmad, S., 1994. Total Site heat integration using the utility system. Comput.
Chem. Eng. 18, 729–742. doi:10.1016/0098-1354(93)E0019-6
IEA, 2017. CO2 emissions from fuel combustion, OECD/IEA.
Jena, S.C., Patnaik, N.K., Sarangi, A., 1996. Heat and Mass Balance in Rotary Kiln Sponge
19
 Iron Making, in: International Conference on Alternative Routes on Iron and Steel under
Indian Conditions. Jamshedpur, India, pp. 59–64.
Klemeš, J., Dhole, V.R., Raissi, K., Perry, S.J., Puigjaner, L., 1997. Targeting and design
methodology for reduction of fuel, power and CO2 on total sites. Appl. Therm. Eng. 17,
993–1003. doi:10.1016/S1359-4311(96)00087-7
Klemeš, J.J., Kravanja, Z., 2013. Forty years of Heat Integration: Pinch Analysis (PA) and
Mathematical Programming (MP). Curr. Opin. Chem. Eng. 2, 461–474.
doi:10.1016/j.coche.2013.10.003
Klemeš, J.J., Varbanov, P.S., Kravanja, Z., 2013. Recent developments in Process
Integration. Chem. Eng. Res. Des. 91, 2037–2053. doi:10.1016/j.cherd.2013.08.019
Kumar, V., Khanam, S., 2012. Recovery and utilization of waste heat in a coal based sponge
iron process. Chem. Eng. Process. Process Intensif. 56, 19–28.

T
doi:10.1016/j.cep.2012.03.002
Liew, P.Y., Lim, J.S., Wan Alwi, S.R., Abdul Manan, Z., Varbanov, P.S., Klemeš, J.J.,

IP
2014a. A retrofit framework for Total Site heat recovery systems. Appl. Energy 135,
778–790. doi:10.1016/j.apenergy.2014.03.090
Liew, P.Y., Theo, W.L., Wan Alwi, S.R., Lim, J.S., Abdul Manan, Z., Klemeš, J.J.,

R
Varbanov, P.S., 2017. Total Site Heat Integration planning and design for industrial,

SC
urban and renewable systems. Renew. Sustain. Energy Rev. 68, 964–985.
doi:10.1016/j.rser.2016.05.086
Liew, P.Y., Wan Alwi, S.R., Klemeš, J.J., Varbanov, P.S., Abdul Manan, Z., 2014b.
Algorithmic targeting for Total Site Heat Integration with variable energy
U
supply/demand. Appl. Therm. Eng. 70. doi:10.1016/j.applthermaleng.2014.03.014
Liew, P.Y., Wan Alwi, S.R., Lim, J.S., Varbanov, P.S., Klemeš, J.J., Abdul Manan, Z.,
N
2014c. Total Site Heat Integration incorporating the water sensible heat. J. Clean. Prod.
77, 94–104. doi:10.1016/j.jclepro.2013.12.047
A
Liew, P.Y., Wan Alwi, S.R., Varbanov, P.S., Manan, Z.A., Klemeš, J.J., 2013. Centralised
utility system planning for a total site heat integration network. Comput. Chem. Eng. 57,
M

104–111. doi:10.1016/j.compchemeng.2013.02.007
Linnhoff, B., Townsend, D.W., Boland, D., Hewitt, G.F., Thomas, B.E.A., Guy, A.R.,
Marsland, R.H., 1982. User guide on process integration for the efficient use of energy,
ED

1st ed. Institution of Chemical Engineers, Rugby, UK.

Liu, G.S., Strezov, V., Lucas, J.A., Wibberley, L.J., 2004. Thermal investigations of direct
iron ore reduction with coal. Thermochim. Acta 410, 133–140. doi:10.1016/S0040-
PT

6031(03)00398-8
Loganathan, R., Sivakumar, P., 2013. Waste Heat Recovery Steam Generator in Sponge Iron
Plant. SIJ Trans. Ind. Financ. Bus. Manag. 1, 23–28.
Man, Y., Feng, J., Ge, Q., Li, F., Chen, Y. mei, Zhou, J. zhi, 2014a. Influence of temperature
E

and time on reduction behavior in iron ore–coal composite pellets. Powder Technol. 6,
CC

361–366. doi:10.1016/j.powtec.2014.02.039
Man, Y., Feng, J. xiao, Chen, Y. mei, Zhou, J. zhi, 2014b. Mass Loss and Direct Reduction
Characteristics of Iron Ore-coal Composite Pellets. J. Iron Steel Res. Int. 21, 1090–1094.
doi:10.1016/S1006-706X(14)60188-6
A

Manesh, M.H.K., Amidpour, M., Khamis Abadi, S., Hamedi, M.H., 2013. A new
cogeneration targeting procedure for total site utility system. Appl. Therm. Eng. 54,
272–280. doi:10.1016/j.applthermaleng.2013.01.043
Matsuda, K., Hirochi, Y., Tatsumi, H., Shire, T., 2009. Applying heat integration total site
based pinch technology to a large industrial area in Japan to further improve
performance of highly efficient process plants. Energy 34, 1687–1692.
doi:10.1016/j.energy.2009.05.017
Matsuda, K., Tanaka, S., Endou, M., Iiyoshi, T., 2012. Energy saving study on a large steel
plant by total site based pinch technology. Appl. Therm. Eng. 43, 14–19.
20
 doi:10.1016/j.applthermaleng.2011.11.043
Midrex, 2015. World direct reduction statistics, Midrex Technologies, Inc.
Nemet, A., Klemeš, J.J., Kravanja, Z., 2014. Mathematical Programming Approach to Total
Site Heat Integration. Comput. Aided Chem. Eng. 33, 1795–1800. doi:10.1016/B978-0-
444-63455-9.50134-3
Pouransari, N., Bocquenet, G., Maréchal, F., 2014. Site-scale process integration and utility
optimization with multi-level energy requirement definition. Energy Convers. Manag.
85, 774–783. doi:10.1016/j.enconman.2014.02.005
Prasad, A.K., 2010. Application of Pinch Technology for Process Integration in Coal based
Sponge Iron Plant, PHD Thesis. ed. NIT Jamshedpur, India.
Prasad, A.K., Dey, N.R., Singh, S.K., 2013. Evaluation of new equipments for utilization of
waste heat in sponge iron industry. Energy Effic. 7, 591–608. doi:10.1007/s12053-013-

T
9242-6
Prasad, A.K., Prasad, R.K., Khanam, S., 2011a. Development of energy conservations

IP
scenarios for sponge iron industry using process integration. Energy Effic. 4, 321–333.
doi:10.1007/s12053-011-9108-8

R
Prasad, A.K., Prasad, R.K., Khanam, S., 2011b. Design Modifications for Energy
Conservation of Sponge Iron Plants. J. Therm. Sci. Eng. Appl. 3, 15001–11.

SC
doi:10.1115/1.4003506
Raissi, K., 1994. Total site integration, PHD Thesis. ed. UMIST, UK.
Sahu, S., Tirumalaraju, D., Khanam, S., Sahoo, A., 2013. Energy Conservation in Sponge
Iron Production Process through Proper Utilization of Waste Heat. Int. J. Eng. Sci.
Innov. Technol. 2, 359–366.
U
Sarangi, A., Sarangi, B., 2011. Sponge iron production in Rotary kiln. PHI, New Delhi.
N
Shenoy, U.V., 1995. Heat Exchange Network Synthesis. Gulf Publishing Company.
Smith, R., 2016. Chemical Process Design and Integration. John Wiley & Sons.
A
Tarighaleslami, A.H., Walmsley, T.G., Atkins, M.J., Walmsley, M.R.W., Liew, P.Y., Neale,
J.R., 2017. A Unified Total Site Heat Integration targeting method for isothermal and
M

non-isothermal utilities. Energy 119, 10–25. doi:10.1016/j.energy.2016.12.071

Varbanov, P.S., Fodor, Z., Klemes, J.J., 2012. Total Site targeting with process specific
minimum temperature difference (DT min). Energy 44, 20–28.
ED

doi:10.1016/j.energy.2011.12.025
Varbanov, P.S., Klemeš, J.J., 2011. Integration and management of renewables into Total
Sites with variable supply and demand. Comput. Chem. Eng. 35, 1815–1826.
PT

doi:10.1016/j.compchemeng.2011.02.009
Vavatuk, W.M., 2002. Updated the CE Plant Cost Index. Chem. Eng. 62–70.
Wang, Y., Chang, C., Feng, X., 2015. A systematic framework for multi-plants Heat
Integration combining Direct and Indirect Heat Integration methods. Energy 90, 56–67.
E

doi:10.1016/j.energy.2015.04.015
CC

Wang, Y., Feng, X., Chu, K.H., 2014. Trade-off between energy and distance related costs
for different connection patterns in heat integration across plants. Appl. Therm. Eng. 70,
857–866. doi:10.1016/j.applthermaleng.2014.06.012
www.pewi-fans.com/industrial-blowers.html [WWW Document], n.d.
A

Zuo, H., Hu, Z., Zhang, J., Li, J., Liu, Z., 2013. Direct reduction of iron ore by biomass char.
Int. J. Miner. Metall. Mater. 20, 514–521. doi:10.1007/s12613-013-0759-7

21
Appendix

In an industrial site if all plants are similar, there is no relative advantage of conventional TS
integration, which is more suitable to integrate heat within the site of different processes. To
illustrate this point, stream data from three similar plants are collected and shown in Table A.
For all three plants hot utilities are found to be zero whereas cold utilities are 0.38kW,
0.40kW and 0.27kW as shown in Section S1. SSSP for this data is shown in Figure A, which
indicates that hot and cold utilities of TS are 0kW and 1.05kW, respectively. Details are
discussed in Section S2. These values are simply addition of hot and cold utilities of
individual plants. This shows that the traditional approach of generating SSSP is invalid in
the case of process plants of similar type where same utility is used in all plants.

T
Table A Stream data of individual plants

IP
Stream Type Ts (oC) Tt (oC) CP (kW/C)

R
Plant-1
Hot 550 100 10.21

SC
Hot 620 553.9 26.32
Cold 40 220 20.21
Cold 40 220 15.12

Hot 550
Plant-2
100 U 12.41
N
Hot 480 348.8 10.67
Cold 40 220 18.31
A
Cold 40 220 20.49
Plant-3
M

Hot 550 100 8.93

Hot 520 371.5 20.42
Cold 40 220 18.76
ED

Cold 40 220 20.41

PT

Figure A. Site source sink profiles

E

640
CC

Site Sink Profile 620

Temperature (degree C)

600

580
A

Site Source Profile

560

540
Cold utility
520

500
-1.2 -1 -0.8 -0.6 -0.4 -0.2 0
Enthalpy, kW
22