INTERNATIONAL JOURNAL OF ENERGY RESEARCH
Int. J. Energy Res. 2008; 32:1092–1106
Published online 17 July 2008 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/er.1447
A model on CO
2
emission reduction in integrated steelmaking by
optimization methods
C. Wang
1
, M. Larsson
1
, C. Ryman
1,2,
,y
, C.-E. Grip
2,3
, J.-O. Wikstro
¨
m
1,2
,
A. Johnsson
1
and J. Engdahl
4
1
Centre for Process Integration in Steelmaking, MEFOS—Metallurgical Research Institute AB, P.O. Box 812, SE-97125 Lulea
˚
, Sweden
2
Lulea
˚
University of Technology, SE-97187 Lulea
˚
, Sweden
3
SSAB Tunnpla
˚
t AB, SE-971 88 Lulea
˚
, Sweden
4
SSAB Tunnpla
˚
t AB, SE- 78184 Borla
¨
nge, Sweden
SUMMARY
The iron and steel industry is a large energy user in the manufacturing sector. Carbon dioxide from the steel industry
accounts for about 5–7% of the total anthropogenic CO
2
emission. Concerns about energy consumption and climate
change have been growing on the sustainability agenda of the steel industry. The CO
2
emission will be heavily
influenced with increasing steel production in the world. It is of great interest to evaluate and decrease the specific CO
2
emission and to find out feasible solutions for its reduction. In this work, a process integration method focusing on the
integrated steel plant system has been applied. In this paper, an optimization model, which can be used to evaluate CO
2
emission for the integrated steel plant system, is presented. Two application cases of analysing CO
2
emission reduction
possibilities are included in the paper. Furthermore, the possibility to apply the model for a specific integrated steel
plant has been discussed. The research work on the optimization of energy and CO
2
emission has shown that it is
possible to create a combined optimization tool that is powerful to assess the system performance from several aspects
for the steel plant. Copyright r 2008 John Wiley & Sons, Ltd.
KEY WORDS: process integration; modelling; CO
2
emission; optimization; steel industry
1. INTRODUCTION
The iron and steel industry is the largest energy-
consuming manufacturing sector in the world.
Therefore, concerns about energy con sumption
and climate change have been growing on the
sustainability agenda of the steel industry. The
world’s annual steel production has been steadily
*Correspondence to: C. Ryman, Centre for Process Integration in Steelmaking, MEFOS—Metallurgical Research Institute AB,
P.O. Box 812, SE-97125 Lulea
˚
, Sweden.
y
E-mail: Christer.Ryman@mefos.se
Contract/grant sponsor: Swedish Governmental Agency for Innovation Systems
Contract/grant sponsor: Knowledge Foundation
Contract/grant sponsor: Swedish Foundation for Strategic Research
Contract/grant sponsor: Luossavaara-Kirunavaara AB
Contract/grant sponsor: SSAB Tunnpla
˚
tAB
Contract/grant sponsor: Rautaruukki Oyj
Received 15 October 2007
Revised 18 February 2008
Accepted 18 March 2008Copyright r 2008 John Wiley & Sons, Ltd.
increasing during the past decades, and a particu-
larly rapid increase is noticed since 2000. Acco rd-
ing to IISI [1,2], the world’s crude steel
production was 1058 Mt in 2004, exceeding
1000 Mt for the first time in the steel production
history. The BF/BOF (blast furnace/basic
oxygen furnace) route and the electric arc
furnace (EAF) route are the two dominating
process routes. The share of the B F/BOF and
EAF-based production of crude steel in 2005 was
65.4 and 31.7%, respectively. The CO
2
emission
from the steel industry links to the production
process. As for the BF/BOF route, the reduction
and melting of iron ore to hot metal (HM) in the
BF is almost entirely based on c oal. Conse-
quently, the steel production industry em its large
amounts of carbon dioxide, accounting for about
5–7% of total anthropogenic CO
2
emission [3].
The CO
2
emission will be heavily influenced by
increasing steel production in the world. It can
be anticipated that CO
2
emission from the
steel industry will increase with the increase in
crude steel production in the near futu re un less
significant changes in the current process
route shares or significant energy/production
efficiency can be made, or some effective CO
2
emission reduction technologies, e.g. carbon
capture and storage, can be employed widely in
the iron and steel industry. It is of great
significance to develop a method to analyse
potential CO
2
reduction possibilities in the steel
industry. Some studies [4 –6] have a nalysed CO
2
emission re duction options within th e iron and
steel industry. Most of these studies applied
a simple top –down econometric approach,
neglecting complex interactions of different
process units for the steelmaking. In this study,
a process integration (PI) method focusing on
the integrated steel plant system (the conven-
tional system of the BF/BOF) has been applied.
An optimization model, which can be used to
evaluate CO
2
emission by optimizing ferrous
burden material use in the BF –BOF system, is
presented. The study also cov ers carbon-trading
schemes in order to find out the lower abatement
cost option(s). Finally, a possibility of applying
the model for a specific integrated steel plant is
discussed.
2. MODEL DESCRIPTION
The model developed is based on a PI technique,
mathematical programming, to analyse CO
2
emis-
sion by optimizing material and energy systems in
the steel industry. A survey on mathematical
programming applications indicates that a broader
application of optimization has been focusing on
chemical and pe troleum engineer ing. For the
metallurgical industry it has been mainly restricted
to the application of linear programming for
inventory control, blending, scheduling and simi-
lar purposes [7]. Deo et al. [8] described the
possibilities to use either mathematical program-
ming or genetic algorithms to find the optimum
operating co nditions in integrated steelmaking.
However, till now unexpectedly few reports on
how to solve the complex steelmaking by PI tools
are available. In this paper, the method described
is based on the mixed integer linear programming
(MILP). The method uses a graphical interface
equation editor ReMIND, which was developed
in cooperation between two Swedish Universities
of Linko
¨
ping University and Lulea
˚
University of
Technology, to generate the mathematical equa-
tions to be optimized. Figure 1 shows the flow
chart of the model structure. There are several
numerical solvers avail able, which can be used for
optimization. In the presented work, the ILOG
CPLEX linear programming solver is used.
Microsoft Excel is used to analyse the modelling
results with some MACRO commands.
The principle of ReMind model is presented in
Figure 2. The model is represented by nodes and
branches where the branches represent energy or
material flows and a node may represent a process
GUI/Equation editor
Equation Solver
Spreadsheet
Model design tool
Optimizing tool
Analyzing tool
Results
Figure 1. Flow chart of the optimization model.
A MODEL ON CO
2
EMISSION REDUCTION IN INTEGRATED STEELMAKING 1093
Copyright r 2008 John Wiley & Sons, Ltd. Int. J. Energy Res. 2008; 32:1092–1106
DOI: 10.1002/er
unit as well as a production line or a whole factory.
Each process node has its own energy demand in
the form of electricity and/or heat demand. These
demands depend on the amount of material
processed in the unit and may be described by
linear or piecewise linear relations. The variations
are described in the system with boundary
conditions, for instance, production capacity,
limited availability for various resources such as
fuels, electricity or raw materials. Each system is
adjusted to the situation in each individual case.
The adjustment is made to answer the questions in
the individual case and to make the model as
efficient as possible.
In this work, ReMIND has been used for the
integrated steelmaking system, which co vers
processes of coke oven plant (COP), lime
furnace, BF, BOF, ladle metallurgy, continuous
casting (CC) and combined heat and power
(CHP). The model includes four kinds of nodes:
material flow nodes, energy flow nodes, process
nodes and end product nodes. Material and energy
flow nodes are the input nodes for the model. The
core nodes for the model are the process nodes
that contain the basic metallurgy process es.
Processes are described by mass and energy
balance to link ingoing material and energy
flows, thereby connecting the different processes.
An example process node, the BF node, is shown
in Figure 3. The end nodes include the main
product from the processes, for instance, slabs for
the whole system, HM or liquid steel if we are only
looking at the BF or the BF1BOF, etc. The other
end nodes could be heat and power generation, gas
to flare, etc.
Production
demand node
Process nodes
Energy supply node
Energy flow
Material flow
Material supply node
1
2
3 4
5 6
7 8
9
Figure 2. Schematic description of the principle of the
ReMind model.
Figure 3. An example of the function editor in the BF node.
C. WANG ET AL.1094
Copyright r 2008 John Wiley & Sons, Ltd. Int. J. Energy Res. 2008; 32:1092–1106
DOI: 10.1002/er
2.1. Objective function(s)
There is a possibility of defining several objectives
in the model depending on the obj ective problem
studied. These can either be analysed one a time,
i.e. single objectives, or combined, i.e. as multi-
objective function. Generally, the objective can be
expressed in mathematical terms as follows:
min zðx; yÞ¼
X
c
j
x
j
þ b
j
y
j
; j ¼ 1; ...; n ð1Þ
s:t:
A
1
x b
1
A
2
x þ By b
2
x 2 R
n
; y 2f1; 0g or integer
where z is the objective function for minimizing
CO
2
emission, x represents the studied variables
(x
i
means the i th variable), y represents the binary
variables, c
j
is the coefficient for the jth variable in
the objective function and b
j
is the coefficient for
the jth binary variable in the objective function.
Engineering design often deals with mult iple,
possibly conflicting, objective functions or design
criteria. For instance, one may want to maximize
the performance of a system while minimizing its
cost. Such design problems are the subject of
multi-objective optimization. Thus, the multi-
objective function is needed when optimizing
more objectives at a time is required. It is useful
to find out an optimum solution with a lower
production cost and at the same time with a lower
CO
2
emission. There are several different
approaches for multi-objective optimization,
e.g. weighted sum, e-constraint and goal
programming. A more detailed description on
each approach can be found in [9]. In this work,
e-constraint method is used for mult i-objective
optimization. For the e-constraint method, only
one objective is optimized, whereas the other
objectives are bounded by some constraints.
In this study, for the multi-objective
optimization problem based on Cost and CO
2
emission minimization, it can be expressed by the
following equation:
min
X
n
a
n
b
n
X
t
X
m
ðC
m;t;n
x
m;t
Þð2Þ
where n denotes objective, e.g. the objective will be
the cost when n 5 1, and CO
2
emission when n 5 2,
etc., x
m,t
is the flow m for the time step t, c
m,t,n
is
the coefficient for the flow m of objective type n in
time step t, a
n
is a coefficient making it possible to
normalize each objective function n, whereas b
n
is
a coefficie nt making it possible to weight the
objectives (note that one constant could also have
been used, but two constants were used to
facilitate the study). The constants also provide
the possibility to exclude any objectives from the
optimization by setting them to zero. a
n
and b
n
correspond to K1 and K2 in Figure 4.
The studied objective is bounded according to
the following equation:
X
t
X
m
C
m;t;n
x
m;t
C
n;
t 8n ð3Þ
Figure 4. Scope and time step of the model.
A MODEL ON CO
2
EMISSION REDUCTION IN INTEGRATED STEELMAKING 1095
Copyright r 2008 John Wiley & Sons, Ltd. Int. J. Energy Res. 2008; 32:1092–1106
DOI: 10.1002/er
where C
n,t
is a constraint for the objective, either
cost or CO
2
emission, during the time step t.
As shown in Figure 4, the scope of the CO
2
emission can be defined locally for direct emission
from a specific plant or globally including both
upstream and downstream emissions. The latter
can be used when doing a life cycle assessment
(LCA) for the studied system. The model can
simulate CO
2
emission for a fixed time or during a
time span; therefore, a time-step function is
needed, see Figure 4. For example, the time-step
function is needed when analyzing the CO
2
emission for different periods for the steel plants
in the emission-trading program.
In connection with the multi-objective
optimization, it is possible to find Pareto-optimal
solutions [10]. A Pareto-optimal solution is a
solution where no objective can be improved
without another deteriorating. The plot of the
objective functions is called the Pareto front, an
example of a Pareto front is shown in Figure 5. As
for the bi-objective optimization problem, the
Pareto front curve represents all the solutions
from minimizing one objective with upper-level
constraints bounded by the other objective, and
vice versa. This allows the decision maker to
choose an acceptable trade-off between the two
goals by con sidering the different solutions along
the Pareto front.
2.2. System definition
Figure 6 shows the system boundary. At the first
step, the model boundary covered the main
process units of the BF and the BOF, i.e. System
I. The model was further extended to cover COP,
CC and CHP in System II. Finally in System III a
sub-model of a rolling mill (RM) is included; thus,
the model boundary has covered a fully integrated
steel plant, i.e. COP-BF-BOF-CC-RM.
The model can be used to analyze the CO
2
emission either for the whole system jointly or
for one or a few sub-models separately depending
on the research interests. Two application cases
covered by this paper correspond to different
system bounda ries in Figure 6, optimizing ferrous
burden materials in BF–BOF [11] and emission-
trading schemes’ (ETS) influence on CO
2
emission
reduction [12]. A customized model for a Swedish
steelmaker, SSAB Tunnpla
˚
t AB, with two inte-
grated production sites of steelmaking and RM, as
an example of a fully integrated steel plant, will be
discussed in the paper as well (System III).
2.3. Validation
The model used in this work is based on an
existing model that was initially developed for
analysis of the energy use for an integrated steel
plant, and the model has been validated by using
actual production data [13]. This model has been
successfully used in several studies mainly focusing
on material, energy use and production cost
Figure 5. Example of a Pareto front for a bi-objective
minimization problem.
Coke oven
CC
BF
BOF
System I
CHP
Rolling
Mill
System II System III
Figure 6. Scheme of the model development layout.
Note: Material and energy flows between and within
processes/sub-models are not included in the figure.
C. WANG ET AL.1096
Copyright r 2008 John Wiley & Sons, Ltd. Int. J. Energy Res. 2008; 32:1092–1106
DOI: 10.1002/er
![Table I. Coefficients used for different objective function [11].](/figures/table-i-coefficients-used-for-different-objective-function-35x24gxc.webp)
