23

Advanced Control Schemes for Cement

Fabrication Processes
Susana Arad, Victor Arad and Bogdan Bobora*
University of Petrosani, *Carpatcement Holding,
Romania

1. Introduction
Taking into consideration that the cement market will record an increasing rate of 15 - 20%,
related to the residential buildings development and to the initiation of large infrastructure
projects the cement production is of great interest, both from the point of view of product’s
quality increase and raw material consumption and environmental impact diminishment. The
demands on cement industry in relation with productivity, quality and price, mean an ever-
increasing need to improve the quality products, the productivity increase improvement,
modernization of the technological flow, improvement of environmental quality. Process
automation is where the industrial area offers the biggest and most satisfying challenges in
terms of combining traditional engineering skills with technological innovation.
The primarily goal of our work is to describe the state-of-the-art and future directions in
update of the cement plant from Carpatcement Deva Branch (Casial).
The project aims the development of a series of patterns, more or less independent from each
other, but all of them focused in the improvement of cement quality and diminishing of
environmental impact, by means of models simulation and identification in this process. The
main objective of this paper is to show through examples how process models can be inserted
into advanced controllers to allow the successful control and optimization of the process when
the controlled variables are not on line measured or they are measured infrequently.
The proper implementation of advanced control schemes for complex cement fabrication
processes rely on the availability of the appropriate mathematical models. The control
strategies are illustrated with actual data obtained for one typical process, the clinker
fabrication process. We have done an analysis of rotary kiln from Casial-Deva Branch
through finite elements approximation.

•
Three main stages in the cement production process exist:

•
Raw limestone processing, raw material mixing and milling;
Clinker production (as intermediary by-product) from raw limestone calcinations in a

•
melting rotary kiln;
Clinker mixing and milling with other products, to obtain the cement as final product.
In those three fabrication stages, due to multiple installations and technologies and the
presence of a great number of parameters conditioning, is which are inter conditioning it is
necessary that the process must by controlled a distributed expert system, based on the
intelligent algorithms.

www.intechopen.com
382 Robotics and Automation in Construction

2. Overview and state of art

2.1 Process description
The cement related business in our country is divided between three big manufactures:
Heidelberg Cement (Germany), Lafarge (France), Holcim (Switzerland), each one
controlling about one third of the specific market.
Deva cement plant was established in 1972 and its operation started in 1976. It is located
nearby Deva town, having an output of 1 million tons/year. Carpatcement Holding took
over this plant in 2000.
This process is carried out employing the dry method, which generates the lowest emissions
level and is the most efficient. The raw and auxiliary materials employed in cement and by-
products processing are limestone, clay, pyrite, flying ash, slag, natural puzzolana, gypsum,
sand wastes issued from other industries. The basic raw materials for cement production
(limestone and clay) are mined out in open pits, from where after primary crushing is
transported on the plant’s site for storage and further preparation. Having high moisture, the
clay requires a dehumidification and drying process. External suppliers supply other raw
materials, such as iron ore (pyrite ash), granulated furnace cinder etc.
The cement production stages are involving the following activities, illustrated in Figure 1
(Procese de productie, 2008):

Fig. 1. The technological flow from cement fabrication

•
•
Extraction in open pits of raw materials such as limestone and clay (1 and 2).
Limestone separation and overcrushing (3), where the sizes comprised between 0
– 50 and 100 – 150 mm are transported to the crusher, where their diameter is
reduced to the needed value for cement production, namely 0 – 25 mm, and then

•
are stored in three limestone silos.
Obtaining the raw meal for clinker (4 and 5) is using a laboratory assessed receipt,
the limestone, clay and pyrite ash are closed and transported to the drying tower,

•
and further to the raw mill.
Clinker production (6, 7 and 8) based on raw meal contained in the storage silos is
dosed and transported to the Humboldt four levels heat exchanger, where the raw
meal is preheated from 60°C to about 800°C. The heat is taken over from the kiln’s
hot gasses passing through the heat exchanger, in counter- current with the
supplied raw meal. Preheated and partially decarbonated rawmeal covers all the
areas within the rotary kiln, in a special zone the clinker being obtained, at 1450°C
temperatures. From the rotary kiln, the clinker is discharged in the grid cooler,
where using the air forced into the system by 9 fans, it is cooled from 1350°C to
about 100°C. The heat demands for obtaining the clinker is supplied by burning fuels
as natural gasses, heavy oil, waste oil, SAF, eco-fuel on the hot end of the kiln, and
the rubbers and waste tyres are supplied through the cold end to the ascendant

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 383

column kiln-heat exchanger. Three silos are in place for clinker storage. As mentioned

•
before, the clinker kiln works on the dry procedure and meets the BAT requirements.
Cement production (9, 10 and 11). The slag from the admixture hall is then stored
in the cinder silos from the clinker mills. The cinder is dried with warm gasses
from the grid cooler or, when the clinker kiln does not operate, by burning natural
gasses in the drier’s hotbed. Gypsum is transported from the admixture hall to the
gypsum silos nearby the cement mills. The clinker, cinder and gypsum contained
in silos, after a laboratory receipt, are extracted, dosed and supplied to the cement
mills. The cement mills are tubular mills with balls, having two rooms and
operating in a closed loop. From the mill, the material is brought to a high
efficiency separator where it is separated, the fine fraction (cement) being taken
over in a haulage relay and stored in 9 cement silos, the heavy fraction being

•
recirculated into the mill.
Cement shipping (12), from silos, the cement can be supplied both as bulk material
and in bags, using rotary Mollers equipment. The cement delivery can be done by
trucks or using the railway network.
The basic product is cement, 80% is clinker manufactured using a dry procedure which
allows a production of 1,026,563 t/year clinker. The technological flow may be synthesized
into a scheme presented in Figure 2. The Deva Branch, Casial has two technological lines to
produce cement with 1.2 millions tons a year capacity. The flow sheet of material balance
from Deva Branch, Casial is presented in Figure 3.

Fig. 2. Flow chart of process fabrication

Fig. 3. The flow sheet of material balance

www.intechopen.com
384 Robotics and Automation in Construction

2.2 Kiln operating

Manufacturing cement by dry procedure is a huge energy consuming process. An important
section of cement technological flow, with a great share in finite product (80%) is the making
of the clinker which is presented in Figure 4, where are represented graphically the flow
installations of clinker fabrications.
The calcinations process (transformation of raw material in clinker, the intermediate product
from what the cement is produced) takes place in the rotary kiln. The existing temperature
in the kiln for the process to take place is of 1450°C, while the flame’s temperature is of
about 2000°C.
The existing kiln systems at Carpatcement Holding are on the dry system with suspension
pre-heating in 4 levels, Humbolt types. The material is supplied from the upper part of the
cyclones and they are circulating in decreasing way in counter- current with the gases
resulted from the burning processes which they are circulating through the separation
cyclones from the lower level to the upper level. In this way, the gases are cooling down and
the raw materials are preheating, Figure 1 (6) and it starts the decarbonated process. The
raw material, which is in suspense it is separated from the gases inn each step of the changer
and meets again the gases in the lower step. The repeating of this process (separation-
blending) at each step of the changer until the discharge of the material into kiln it’s ensures
a good thermal transfer.

Fig. 4. Flow sheet of clinker fabrication

The raw meal charging station is equipped with control bins, gravimetric chargers type
Schenk and pneumatic pumps type Fuller.
The control bunkers should maintain a constant raw meal height providing a prescribed
constant material quantity in the supply. Raw meal is taken over from in supply control
bunkers with rampart extractors, being charged according to the centralized commands
given in the central command room, with Schenk gravimetric chargers.

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 385

The material charged in this manner is circulated in the screw pneumatic pump’s bins that
are transported it to the heat exchanger. For each technological line, three Fuller pumps are
in place (2 operating and an auxiliary one) and a four-stage Humboldt suspension heat
exchanger.
The charged material is pneumatically hauled to the exchanger’s upper side, using the
existing joint between stage I and stage II. The material receives heat from the hot gasses
during its traject in the heat exchanger’s cyclones, from upside down in the direction I-II-III-
IV, after that entering the smoke chamber and then the furnace. Inside the exchanger, the
material is heated up to 800°C - 810°C, also being partially decarbonatated. Gasses are then
entering in exchanger at about 1000°C temperature, in his bottom side circulating along his
cyclones in direction IV-III-II-I and finally are exhausted through VRA and VRB exhausters.
The partially decarbonated rawmix is fed into the rotary kiln, having 97 m length and a 5.8m
in diameter. Here takes place the final stage of the clinkerization process, based on specific
thermal and chemical processes.
For burning purposes, liquid fuel (heavy oil), gassy fuels (natural gasses), or both kinds of
fuels are employed, the equipment being well designed to meet this demand. According to
the reactions within the kiln and the resulting compounds, the rotary kiln comprises the

•
following zones:
Decarbonatation area (calcinations), where the alkaline carbonates are decomposed at

•
temperatures comprised between 1000°C and 1100°C.
Transition zone (solid phase reactions area), where the first mineralogical compounds

•
are formed, through solid phase reactions, at temperatures 1000°C - 1350°C.
Clinkerization zone (sinterization area) where, 1350°C - 1500°C temperature values ,
the liquid phase appears, in his presence the tri calcium, silicate (alit) develops, the

•
cement’s most valuable compound.
Cooling zone, where, at temperatures ranging from 1450°C to 1250°C the mineralogical
compounds occurs. Burned gasses are circulated in the kiln backwards related to
material’s advancing direction, than their dust, content is minimized employing a
560,000 m3/h capacity electrostatic precipitator system (EPS).
EPS, works in optimum conditions when the gas temperature doesn’t pass 180° C, because
of this they are cooled down in a water tower. From the kiln, the clinker is discharged at the
warm head of the cooling grate where it is taking place a suddenly cooling to 65° C.
A large volume of gases has to be moved through the kiln system. Particularly in
suspension-preheated systems, a high degree of suction has to be developed at the exit of
the system to drive this. Fans are also used to force air through the cooler bed, and to propel
the fuel into the kiln. Fans account for most of the electric power consumed in the system,
typically amounting to 10–15 kWh per tonne of clinker.
The grate cooler is composed from three grates of different sizes, on which are put high-
melting steel plates with wholes. A bed of clinker up to 0.5 m deep moves along the grate.
These coolers have two main advantages: they cool the clinker rapidly, which is desirable
from a quality point of view, and, because they do not rotate, hot air can be ducted out of
them for use in fuel drying, or for use as preheated combustion air. After cooling the clinker
is crashed until, the granulation is max. 25 mm and then through a transport system made
from two metallic chains with coupes and a relay of three belt conveyors, is transported at
the tree clinker bunker. The dust collection from the cooling grates ensured with a multi -
cyclones batteries with two frames.

www.intechopen.com
386 Robotics and Automation in Construction

3. The rotary kiln modelling

A rotary kiln, the world's largest manufacturing machine - is the major component of the
cement line. Rotary kilns have wide use in industry from the calcinations of limestone to
cement manufacturing to calcining of petroleum coke etc. The kiln is a large rotating furnace
approximately 100 m long, and four to seven m in diameter that weighs over 300 tonnes,
Figure 5. The rotary kiln consists of a tube made from steel plate, and lined with firebrick.
The tube slopes slightly (3%) and slowly rotates on its axis at between 30 and 250
revolutions per hour. Raw meal is feed in at the upper end, and the rotation of the kiln
causes it to gradually move downhill to the other end of the kiln. At the other end fuel, in
the form of gas, oil, or pulverized solid fuel, is blown in through the "burner pipe",
producing a large concentric flame in the lower part of the kiln tube. As material moves
under the flame, it reaches its peak temperature, before dropping out of the kiln tube into
the cooler. Air is drawn first through the cooler and then through the kiln for combustion of
the fuel. In the cooler, the cooling clinker heats the air, so that it may be 400 to 800 °C before

The dimensions and parameters of the oven are: dimensions ∅ 5.8 x 97 m, angle 3 %,
it enters the kiln, thus causing intense and rapid combustion of the fuel.

backing points 4, production capacities Q = 3,125 t/ day, main driving P = 500 KW, n = 750
rot/ min, second driving P = 500 KW, n = 750 rot/ min.

Fig. 5. Rotary kiln

Problems such as low thermal efficiency and low product quality have plagued rotary kiln
operations yet these machines have survived and have been continuously improved (fuel
efficiency, automation) for over a century.
With an ever-increasing focus on reducing greenhouse gas emissions, the continued or
increased use of rotating kilns can only be achieved by reducing the thermal and electrical
energy consumption used in these processes. A fluid bed calciner or dryer achieves rapid
drying by the large heat transfer coefficient obtained through the high air volume being
circulated. The penalty is the increase in electrical energy required to circulate this high air
volume. Rotary kilns on the other hand have poor heat transfer coefficients, hence higher
thermal energy demand, due to the need for larger devices and thus more opportunity for
heat to be lost.
In most rotary kiln operations, the chemical reactions in the bed require high temperature,
for example, cement kilns will require temperatures of approximately 1500°C. The energy to
raise the temperature and drive endothermic reactions is from the combustion of a range of
fuels such as natural gas, coal and more and alternative fuels. Heat transfer from the gas to
the bed is complex and occurs from the gas to the bed surface and kiln wall to bed surface
via conduction, convection and radiation.

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 387

A number of rotary kiln models has been proposed over the years and recent computational
fluid dynamic models can be developed but all have their limitations (Barr, 1989; Bui, et al.,
1995). Most assume isothermal conditions through the bed at any axial position (Majumdar,
& Ranade, 2006). The bed motion regime, cascading, rolling or slumping depends on the

•
rotational speed of the kiln, the percentage fill and the feed physical properties.
They are models which have in the site the thermal processes, models that are
following the thermal transfer between the material bed, gas, kiln walls and
environment, where it appears conduction, convection and radiation phenomenon.
The measures are the material temperatures from supply in those four steps, gas

•
temperature, and walls temperature in the four steps.
They are chemical models who analyze the endo -thermal phenomena that are
taking place at the raw material calcinations. The kiln parameters are the gas
emissions of O2, CO2, NOx, quantities and material compositions (Gorog, et al.,

•
1981).
They are models which have basis the energetically balance of the kilns where they
are appearing energetic aspects in connection with the kiln’s drive, rotation, motor
moment and they are following the automation power adjustment.
A series of equations representing conservation of mass, energy and species averaged over
the cross-section are solved using appropriate numerical methods (He, et al., 1996).
The bed for example is assumed to be well mixed and isothermal in any given transverse
plane (Georgallis, et al. 2001). Although these models have been successfully used in
industry, they are limited for information that can be extracted.
Due to the complex models character, nowadays many software packaging are allowing to
employ numerical analysis of thermal phenomena (FLUX STUDIO, ANSYS,
MULTIPHISICS, FLUENT, COMSOL MULTIPHISICS, QuickField, etc.). A 3D physical
model of the kiln where it can be observed the physical components, walls, material bed and
the burning pipe is given in Figure 6.
As a result a number of researchers have begun the quest for a more encompassing
modelling effort. Boateng and Barr (Boateng, & Barr, 1996) have coupled a conventional
one-dimensional plug flow model with a two-dimensional representation of the bed’s
transverse plane. This improves the ability to simulate conditions within the bed. Alyaser
(Alyaser, 1998) has modelled for axisymmetric conditions. Fully coupled three-dimensional
modeling is applied to the rotary lime kiln (Georgallis, et al., 2001). Three sub-models are
coupled, namely the hot flow model, the bed model and the wall/refractories model. The
model takes into account the major phenomena of interest including the gas flow, all modes
of heat transfer and thermal effects of the refractory.

Fig. 6. 3D- rotary kiln model

www.intechopen.com
388 Robotics and Automation in Construction

A model of rotary kiln heat transfer, which accounts for the interaction of all the transport
paths and processes to the rotary kiln from Casial, Deva Branch is presented in our paper.
Information exchange and directions of transfer are shown in Figure 7 (Barr, et al., 1989).
Two dimensional modelling is applied using finite element method. Heat transfer within the
kiln refractory wall was solved using a finite-element approximation for one-dimensional
transient conduction. Interface temperature boundary conditions for the kiln are used in the
model. Heat flux boundary conditions are used for both the inner and outer surfaces in the
wall model.
The mathematical model of heat-transfer for linear problems is described by the differential
mathematical model of the thermal conduction, Eq.(1) and (2):

Fig. 7. Information exchange of heat transfer

∂T
div ( λ gradT ) + q − ρ c =0
∂t
(1)

∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞ ∂T
⎜ λx ⎟ + ⎜ λy ⎟ = −q − c ρ
∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂t
(2)

Where: T is temperature, t- time, λx(y)-components of heat conductivity tensor; λ - heat

conductivity, q - volume power of heat sources, burner - constant, c(T) - specific heat, ρ -
density of the substance.
In linear case all the parameters are constants within each block of the model. The oneness
of the precedent equation solution of the thermal conduction proposes the knowledge:
a. The heat sources in the domain of calculus, q;
b. The material properties, ρ, c şi λ;
c. Initial conditions T;
d. Limit conditions.
The next limit conditions on surfaces S1, S2, S3, on the frontier domain of calculus is possible.
(Fireteanu, et al., 2004):
Ü Type Dirichlet,

T ( x.z. y, t ) = Ts , t > 0, on Sl (3)

Ü Type Neuman inhomogeneous

∂T
λ = qs , t > 0, on S2
∂n
(4)

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 389

Ü Type mix

= α (T − T0 ) + β k SB (T 4 − T04 ) , t>0, on S3
∂T
λ
∂n
(5)

Where, qs is the superficial specific flow imposed, is the thermal transfer coefficient by
convection, kSB is a Stephan-Boltzmann constant (5.67032·10-8 W/m2/K4), β is an emissive
coefficient, and T0 - ambient radiation temperature.
Convection boundary condition and radiation boundary condition can be specified at
outward boundary of the region.

3.1 Formulate the problem

It will be accomplished the thermal analysis by numerical method with finite elements using
as QuikField software, of the heat transfer problem at the rotary cement kiln.
It will be determined the temperature value in different points of the model, in each block,
thermal gradients, heat flux densities, and temperature on the contour of shell.
The software is based on heat conduction equation with convection and radiation boundary
conditions. The technical characteristics of the rotary kiln are shown in Table 1:

Geometrical Numerical Numerical

Operational variables
parameters value Value
Temperature of clinkerization
Lenght [m] 97 1300
[°C]
Internal kiln radius [m] 1.9 Velocity of kiln [rpm] 1.9
External kiln radius [m] 2.9 Limestone feed temperature[°C] 800
Thermal transfer coefficient
Inclination [%] 3.0 20
[W/Km2 for (S1)
Thermal transfer coefficient
Refractory thickness [m] 0.9 350
[W/Km2] for (S3)
Thermal transfer coefficient
Kiln shell thickness [m] 0.1 0.5
[W/Km2] for (S2)
Emissive coefficient 1
Thermal conductivity [W/K.m]
in bad 0.693
in gas 0.8
in shell 10
in refractory 0.04
Table 1. Operational variables of the rotary kiln for calcining limestone
Defying the thermal transfer problem for this case it was made the geometrical model, the
document describing the problem geometry, the labels of the blocks (bed, gas, refractory,
shell) and it was made the mesh of our model. The model contains specific geometric objects
and establishes the correspondence between the objects and material properties, field
sources and boundary conditions. We gave the properties of each material from the named
blocks (heat conductivity, emissive coefficient for convection and radiation (Table 1), the
thermal field sources were defined and also the boundary conditions and limits.

www.intechopen.com
390 Robotics and Automation in Construction

It was defined three surfaces outlines S1, S2, S3, belonging the calculus frontier domain,
with specified boundary condition like: S1 outer surface as interface between shell and
environment, S2 surface interface between bed-gas and bad-refractory, S3 inner surface
interface between refractory –gas.

3.2 Numerical solutions

The values of temperature T [K], heat flux F [W/m2] and temperature gradient G [K/m] in
some points at interface surfaces between gas and refractory and on the vertical axis of kiln
through each isotherm, in gas, bad and in shell also were calculated and given in Table 2
and Figure 8. Also it was represented the temperature variation on the contour of the inner
surface, Figure 9 and temperature distribution between shell and refractory, in Figure 10.
Inner surface Temperature value T Gradient Value Flux heat
points [K] G[K/m] F[W/m2]
1 1295.6 298.97 239.17
2 1246.0 415.16 332.13
3 1192.2 584.58 467.67
4 1141.8 692.21 553.77
5 1184.8 1021.8 708.08
6 1032.7 1037.4 718.95
7 975.99 1016.9 704.68
8 914.44 991.89 687.38
9 778.01 18578 743.11
10 693.01 18367 734.69
11 636.96 18231 729.23
12 466.04 17808 712.33
13 318.27 70.556 705.56
14 316.21 66.78 667.8
Table 2. Thermal field values
The cross-section model is shown to simulate the measured thermal performance of the kiln
for clinker. The interaction among the heat-transfer processes at cross-sections of the kiln
was examined, and explanations were made for both the observed close coupling of the bed
and inside wall temperatures and the high rates of heat input to the bed occurring near the
kiln entrance and in the presence of an endothermic bed reaction.

Fig. 8. Map of temperature, vectors of heat flux and thermal gradient in cross-section

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 391

The model was validated using thermal measurements from Casial’s kiln. This effort
demonstrates how a model may be used to capture flame phenomena for rotary kilns and to
solve shell fault into the kiln.
A model accepts heat flux values from the hot flow side and temperatures on the wall
interface. Evidence (such as a non-uniform product) has suggested that large temperature
gradients exist near and within the bed.
The work carried out is aimed at understanding and improving the heat transfer in rotary
kiln and to provide a systematic basis for the efficient operating of kilns. It can be noticed
that temperature distribution nearby the kiln’s shell is very close to the trend obtained by
the pyrometer used for temperature monitoring.

Figure 9. Variation of temperature on the contour of inner surface

Fig. 10. Variation of temperature on the interface surface between shell and refractory

www.intechopen.com
392 Robotics and Automation in Construction

The data processed by statistic functions about clinker temperature and automate measured
pyrometer temperature are shown in Figure 11 and Figure 12 and the values of the statistical
parameters we have obtained. The result reflected on prediction performance plot with a
correlation of 0.96 ( Arad &Arad, 2003).

Fig. 11. Clinker temperature

Fig. 12. Pyrometer temperature

4. Cement kiln emissions

The most important gas emissions from cement industry are CO2. The carbon dioxide
emissions which are generated represent about 5% from the world wide CO2 emissions
induced by human activities. These high level emissions are resulting basically from the
specific technology for cement production. The main sources of CO2 in cement industry are:
raw material and fuel burning.
The N2O emissions generated by the cement kilns as a consequence on combustion
processes are relatively low, having no significance if related to CO2 emissions. The last ones

•
are issuing in the following stages:
The calcinations stage:

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 393

• Kiln: direct CO2 emissions having two main sources:

- calcinations (decarbonification) of raw material (60%);

•
- burning of fuel (40%), thermal energy consumption.

•
The milling stage:
Cement milling: indirect CO2 emissions, electric energy consumption.
The direct CO2 emissions in the process are mainly occurring by employed fuel and raw
materials (calcium carbonate), being released during the stage of clinker production in kiln
(the so called calcination). Thermal energy is also used during this stage. So, if natural gas is
used instead of coal, the CO2 emission is decreased with 25%. The thermal consumption and
CO2 emissions also related to the kiln type employed for calcinations and clinkerization.
Apart from this, the emissions are different according the kind of the raw materials
employed. About 60% of total amount of CO2 emissions are depending on the employed
raw materials (during calcination), and the rest of 40% is related to the fuel consumption.
Indirect emission of CO2 from the process are having as main source the use of electric
power for milling purposes, from primary calcinations or from clinker milling (when it is
mixed with additives for the final cement production process).
Three important environmental issues can be outlined as major problems confronting the
European cement industry, two being local and one global, by their nature:
Ü emissions from factories, other than CO2 (SO2, NOx, dust, etc.) and local transportation
of raw materials and products;
Ü raw materials extraction and transportation and their environmental impact (rural
areas, natural resources and biodiversity) and their effect on human life environment
(dust and emissions related to transportation, noise, vibration);
Ü CO2 emissions from plants (emissions from factories and vehicles) and energy
consumption (use of fossil fuels).
A major role in emissions generation is related to the employed fuels. In fuel selection, the
major factor is represented by the cost involved at burner’s level, comprising all the
expenses with acquiring, processing and feeding. The presently employed fuel at
Carpatcement Holding, Deva Branch, is methane, but for economical reasons it was replaced
from 2007 with pet coke and heavy oil.
CO2 emissions reduction measures in the calcinations process (direct emissions) aim the

•
followings:

•
Raw material composition

•
Fuel replacement

•
More efficient technological process
Cement’s final composition (clinker content)

4.1 The diminution of environmental impact

Producing cement has significant positive and negative impacts at a local level. On the
positive side, the cement industry may create employment and business opportunities for
local people, particularly in remote locations in developing countries where there are few
other opportunities for economic development. Negative impacts include disturbance to the
landscape, dust and noise, and disruption to local biodiversity from quarrying limestone
and fabrication process. The cement industry make real efforts to diminish the CO2

•
emissions by implementing actions derived from Kyoto protocol, such as:
improving the production processes through more efficient technologies;

www.intechopen.com
394 Robotics and Automation in Construction

•
•
final cement composition (clinker content);

•
raw material composition;
use of wastes in production processes (European-Union countries have different

•
policies and legal requirements from this point of view);

•
replacing high level CO2 emission fuels with fuels generating lower CO2 emissions;

•
carbon dioxide extraction from emitted gasses;
CO2 emission level reduction at vehicles;
Emissions from cement works are determined both by continuous and discontinuous
measuring methods, which are described in corresponding national guidelines and
standards. Continuous measurement is primarily used for dust, NOx and SO2, while the
remaining parameters relevant pursuant to ambient pollution legislation are usually
determined discontinuously by individual measurements.
Taking into account the significant environmental impact of the cement fabrication, the major
environmental polluting factors by using the mathematical simulation procedures was done.
The environmental impact analyses regarding the activities developed at the Casial cement
factory were realized by simulating the pollutants’ dispersion into the atmosphere from
1995 to 1998 (Arad, 2001).
Referring to air quality in the area we distinguish two categories of pollutants:
Ü Pollutants resulting from cement fabrication processes, represented by dust,
Ü Pollutants resulting from fuel burning processes with SO2, NO2, CO and other
emissions.
The air quality was assessed at the outset of the technology up dating activity.
The analysed area was monitored in 1998 by means of 12 monitoring points by the Agency
of Environmental Protection Deva and also by 5 monitoring points by Casial factory and at a
point (107) situated 450 m away from the pollution source the gaseous pollutants were
assessed. The experimental data obtained regarding the air quality were used to represent
the monthly average concentrations at 10 monitoring locations, depending on the distance
between each assessed location and the cement factory.
The monitored data provided by the beneficiary have pointed out the following:
Ü By analysing the degree of pollution with airborne powder, we find but that the
monthly average values in 1998 did not often exceed the values specified in regulation -
STAS 12574/87, namely the maximum allowable concentration is 0.150 mg/daily. The
maximum annual values exceeded Maximum Admitted Concentration (MAC) with the
frequency of 5.1 %.
Ü Regarding gaseous pollutants we can notice that the average daily values as well as the
maximum values for NO2 do not exceed MAC. The same holds true for SO2, where
neither the daily average values nor the maximum ones were exceeded.
Ü The pollution degree with dust is high. The average monthly values of ten exceed the 17
m2/monthly limit.
In order to assess the pollution sources contribution we resort to mathematic modelling and
also to the results obtained from monitoring, thus having enough information available to
make a study on the pollution impact.

4.2 Predicted air pollutant impact

We have used a software Spreadsheet type, Excel for analyses and pollutant dispersion
prediction in the air. The model assumes that the concentrations spatial distributions of the

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 395

pollutant produced are of Gaussian shape. The model is therefore a Gaussian model
combined with a procedure of averaging concentrations on long time intervals. It is a
Gaussian model because the vertical and horizontal concentrations distributions are the
normal ones.
From the measurements available at Deva Branch, Casial we have simulated the dispersion
model of the pollutants resulting both from the processes of burning fuels and from burning
in clinker kiln and in the lime kiln (Arad & Arad, 2002). The nitrogen dioxide (NO2) and the
sulphur dioxide (SO2) were taken into consideration, the other pollutants being
insignificant. The flow rates are much higher when fuel oil is used. The concentrations are
appraised at ground level, at various distances from the source.
The atmospheric concentrations are appraised by the stability grade from 1 to 6,
representing situations from unstable to stable. With our data, a grade of stability 3 has been
estimated, slightly unstable. In the Figure 11 and Figure 12 are rendered the concentrations
at ground level, at various distances from the source for NO2 and SO2 for the clinker kiln
whose smokestack is 68 m high while the ambient temperature is 10o C in Figures 13,a and
14,a respectively 0oC in Figures 13,b and 14,b.

Predicted Air Pollution Impact Predicted Air Pollution Impact

35 30
30 25
25
1 m/ s 20 1 m/ s
20
3 m/ s 15 3 m/ s
15
5 m/ s 10 5 m/ s
10
5 5
0 0
0,5 0,8 1,5 3 5 10 20 35 60 100 0,5 0,8 1,5 3 5 10 20 35 60 100
D ista nce (Km) D ista nce (Km)

a. b.
Fig. 13. Concentrations at ground level from the source for NO2 at 10 and 0 degree
It can be noticed that the concentration values are higher at temperatures of 10° C. The
situation of the concentrations is far below the allowable values for NO2 sometimes above
them for SO2 at wind velocities of 3.5 m/s and at higher temperatures.

Predicted Air Pollution Impact Predicted Air Pollution Impact

120 100

100 80
80 1 m/ s 1 m/ s
60
60 3 m/ s 3 m/ s
5 m/ s 40 5 m/ s
40

20 20

0 0
0,5 0,8 1,5 3 5 10 20 35 60 100 0,5 0,8 1,5 3 5 10 20 35 60 100
D ista nce (Km) D ista nce (Km)

a. b.
Fig. 14. Concentrations at ground level from the source for SO2 at 10 and 0 degree
The dispersion conditions improve for the pollutants given off by a polluting source 68 m
high (H= 68m) due to horizontal transport conditions, through an increase of wind velocity
over 2 m/s, the atmospheric calm decrease under 25% and because of vertical dispersion.

www.intechopen.com
396 Robotics and Automation in Construction

4.3 The simulation of the dust dispersion using MATLAB

The dust pollutant (settling suspensions) represents the main environmental negative factor
generated during the cement fabrication at the cement factory Casial, Deva Branch.
The simulation’s results enable the comparison with the data obtained by classical analyses
and a high prognosis of the pollutants evolution was done.
The environmental impact analyses regarding the activities developed at the Casial cement
factory were realized by simulating the pollutant’s dispersion into the atmosphere, by using
a computing sequence of the MATLAB space, (Arad & Arad, 2002). We have represented the
concentration fields on a surface area of 100 - 150 km2 for the data collected of 1995 years,
Figure 15 and 1998 year, Figure 16 and 17.
The data measured by Casial factory and Agency of Environmental Protection Deva are
represented by average monthly concentrations in 10 monitoring points depending on the
distance of that point to Casial factory. The results of the simulation for the year 1998 show
that dust pollutant has come down compared with the first period analysed. The
representation of the data obtained in a longer period (1995-1999) is the surface with level
peaks of concentrations on a surface of 10 km 2 from the sources .
The model predictions were verified by comparing them with the experimental data.

Fig. 15. Dust Pollutant dispersion for 1995 year

The integration in the environment legislation of the activities in the analysed site has been
made based on the synthetic assessment of the quality indicators and based on the obtained
prognosis by simulation where have been take actions.

Fig. 16. Dust Pollutant dispersion for 1998 year

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 397

Fig. 17. Dust Pollutant dispersion with data measuring by Casial for 1998 year
As far as the monitoring of the outlets in 1999 there was no system of monitoring them on
the platforms of Casial Plant.
Now, the technologies were up to date and an automatic monitoring system was achieved.
The care for the environment protection was materialized by the modernization of the mill’s
EPS (electrostatic precipitators) and the introduction of the bags filters for the auxiliary de-
dusting. Carpatcement Holding, Deva Branch have already implemented and certified an
integrated management system of quality, occupational health and safety, and environment,
in accordance of the international standard requirements: Quality management system-
requirements-ISO 9001, Environmental Management System, specification and using guide
ISO 14001 and Occupational Health and Safety Management System, after OMSAS 18001
standard. Obtaining the certification of the system after OMSAS 18001 (occupational health
and safety), generates confidence that the society respects: the legal and statutory
requirements, provides a high standard of occupational health safety management as a part
of the firm’s general management system.

•
The total amount of environmental investments is of 8.6 billion Euros, consisting in:

•
dust suppression equipments updating;
purchase an automated equipments of alternative fuel valorisation.

5. Kiln control and advanced control schemes

Effective control of the kiln is complicated by several factors, including nonlinearities in the
dynamics and the large dimensionality of the problem. A linear control scheme has been
already implemented and tested with success. In order to improve the performance of the
control scheme, linear controllers must be replaced by fuzzy controllers.
Fuzzy control is a practical alternative for a variety of challenging control applications since
it provides a convenient method for constructing nonlinear controllers via the use of
heuristic information. Such heuristic information may come from an operator who has acted
as a “human-in-the-loop” controller for a process. In the fuzzy control design methodology,
we ask this operator to write down a set of rules on how to control the process and then we
incorporate these into a fuzzy controller that emulates the decision-making process of the
human. In other cases, the heuristic information may come from a control engineer who has
performed extensive mathematical modelling, analysis, and development of control
algorithms for a particular process. Again, such expertise is loaded into the fuzzy controller
to automate the reasoning processes and actions of the expert. Regardless of where the

www.intechopen.com
398 Robotics and Automation in Construction

heuristic control knowledge comes from, fuzzy control provides a user-friendly formalism
for representing and implementing the ideas we have about how to achieve high-
performance control (Passino & Yurkovich, 1998).
Formation of the desired clinker minerals involves heating the raw meal through the
temperature stages mentioned above. To meet the clinker quality objective, the most
obvious control is that the clinker should reach a peak temperature such that the finishing
reaction takes place to the required degree.
However, for efficient operation, steady conditions need to be maintained throughout the
whole kiln system. The feed at each stage must be at a temperature such that it is "ready" for
processing in the next stage. To ensure this, the temperature of both feed and gas must be
optimized and maintained at every point.
The independent use of fan speed and fuel rate is constrained by the fact that there must
always be sufficient oxygen available to burn the fuel, and in particular, to burn carbon to
carbon dioxide. If carbon monoxide is formed, this represents a waste of fuel, and also
indicates reducing conditions within the kiln which must be avoided at all costs since it
causes destruction of the clinker mineral structure. For this reason, the exhaust gas is
continually analyzed for O2, CO, NO and SO2.
The assessment of the clinker peak temperature has always been problematic. Contact
temperature measurement is impossible because of the chemically aggressive and abrasive
nature of the hot clinker, and optical methods such as infrared pyrometer are difficult
because of the dust and fume-laden atmosphere in the burning zone. The traditional method
of assessment was to view the bed of clinker and deduce the amount of liquid formation by
experience. As more liquid forms, the clinker becomes stickier, and the bed of material
climbs higher up the rising side of the kiln. It is usually also possible to assess the length of
the zone of liquid formation, beyond which powdery "fresh" feed can be seen.
The two-colour pyrometer MPZ 4x is used at Casial factory for the intensity measuring of
infrared radiation at two different wavelength. The series MPZ1x to MPZ5x provide five
efficient, microprocessor-controlled pyrometers for noncontact temperature measurements.
The temperature range covers 700 °C - 3000 °C.
Correct measurement of the temperature on a kiln shell is essential for efficient operation of
the kiln. ECS/CemScanner represents the state-of-the-art in kiln shell infrared scanning. It
combines a robust design with advanced software features, making it an indispensable aid
to the operation and optimisation of kilns.
The CemScaner system controls the kiln (shield) temperatures and it was specially
developed for supervisory control, monitoring and reporting functions from central control
room (CCR) at Casial factory. The adjustment of temperature is done in real time and in the
same time the safe operation of the equipment is assured.
As an exercise in process control, kiln control is extremely challenging, because of multiple
inter-related variables, non-linear responses, and variable process lags.
Since 2000, the technological flow from Cement Plant is controlled automatically by ECS
system (Expert Control and Supervision) from the CCR using the ECS/NTech platform,
complex high level supervisory control system. The ECS environment offers a set of basic
programmed interlocking sequence in process-control. From this point of view, the
installation was equipped with measuring and control devices.

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 399

The operator user interface with full SCADA functionality for technological flow control,
interactive display of results has also been developed and for example the kiln operating is
presented in Figure 18.
Local process automation and monitoring were accomplished. The Casial factory is divided
in areas of automation:
1. Raw Mill
2. Rotary Kiln and Cooler
3. Cement Mills
4. Packing Product.
Automation in Area (2) is based on 2xMain PLC 5/80E Allan Bradley, with 4 and 3 RIO
racks, the communication is DH+. Main PLC-5 communicates also on Modbus with dosage
system Schenck, with the burner furnace based on Allan Bradley SLC 500/4 processor and
with the PLC of Electrostatic precipitator Kiln being the same type as Cooler Main PLC.
Also, a channel of the processor is used to connect the PLC drive system with the PLC of the
kiln. So the PLC Kiln system controls each of the four units.

Fig. 18. Monitoring system and control kiln

Programming for the PLC-5 and SLC 500 used software RSLogix 5. The rest is developed on
the PLC Siemens S7-400 structure with profibus communication. For these PLC’s it is used
Simatic V5.4, Chart 6.1 Programming.
Further information can be obtained from the exhaust gas analyzers.
In our work, an intelligent kiln control system was proposed, with advanced simulation;
Figure 19 shows a schematic of the simulation system. In the system, neural network models
are used in conjunction with advanced high-level controllers based on fuzzy logic

www.intechopen.com
400 Robotics and Automation in Construction

principles. These operate using expert system strategies that maintain a "just sufficient"
burning zone temperature, below which the kiln's operating condition will deteriorate
catastrophically, thus requiring rapid-response, "knife-edge" control.
A local control in each particular operation and an optimum control of entire plant are
enforced. The Expert system is built on a true real-time process control platform that can be
easily linked to the process, through communication with an existing control system.
Simplified structure of a fuzzy control system is presented in Figure 20. Rule-based expert
systems are used to improve production and optimize fuel efficiency.

Fig. 19. Advanced simulation system

The system is not based solely on fuzzy logic but on a hybrid system that utilizes a number
of techniques such us Statistical Process Control (SPC) and Model –based Predictive control
(MPC).
The proposed control scheme of the kiln considers a fuzzy control system in which we have
as rules the chemical model variables of the process as well as the thermal and energetically
ones, look like in Figure 21.

Fig. 20. Fuzzy-Expert system

Control strategies are following:
Ü Fully automatic start-up and shut down
Ü Automatic change of kiln parameters
Ü Control strategies can be completely customised
Ü Self optimising control schemes
Ü Intelligent management of external disturbance,

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 401

The temperature in the clinkerization area is an output variable y(t) needing to be regulated.
The external controls available to achieve this are few:
Ü Feed rate: this defines the kiln output
Ü Rotary kiln speed: this controls the rate at which the feed moves through the kiln tube
Ü Fuel injection rate: this controls the rate at which the "hot end" of the system is heated
Ü Exhaust fan speed or power: this controls gas flow, and the rate at which heat is drawn
from the "hot end" of the system to the "cold end".
The input variables are:
Ü Primary air flow rate
Ü Secondary air flow rate
Ü Natural gas consumption
Ü System total pressure.
Kiln Control System is based on a database, which inherits Fuzzy-technologies for process
optimisation. Based on an ongoing rating of the process, the current state of the facility is
evaluated.
Sophisticated algorithms, which inherit Kiln Control System‘s and the customers experience
it will be implemented then calculate new parameters for subordinate control systems (pre
heater system, burner, kiln cooler, gases analyzer, charger Schenk, clinker silo).

Fig. 21. Kiln Control

For the acquisition and long term storage of process data, the database is used also. A
flexible and very easy to use software system for the management of process data, display of
trends analysis and reporting is available. The software is running on a standard PC, which
can very easy be connected to all automation systems, which are used in plant. Due to the
fact that Kiln Control is designed as a configurable Fuzzy Expert system, new possibilities to
control cement plant arise.
Simple PID-controller can only use a single input to calculate a single output only to be used
for the loop control. Kiln Control‘s Fuzzy-technology processes a multitude of inputs. The
system can be tuned to follow different targets.
This flexibility facilitates new innovative solutions for plant controllers, which render much
better results, than classical PID-controllers.

www.intechopen.com
402 Robotics and Automation in Construction

Of course, modern versions of the „classical” strategies: feed = constant; temperature =

constant, can be chosen.
Our computer control systems make a "calculated" temperature, using contributions from all
information sources, and then set about controlling it.
The new hybrid control system uses predictive-adaptive controllers to keep the main
variables close to the set points and reject disturbances are available (Barreto G., A., 1997).
When doing so, the advantages of the Fuzzy-Expert system are preserved, namely the
consideration of several inputs, which might be contradictory.
The system architecture, as depicted in Figure 22, which is composed of six units
components which are interconnected with Kiln Control System. Greater process knowledge
in the hands of operators, engineers, and managers will lead toward optimized equipment
design and operation, and will have significant impact on cement fabrication economics.
The models must be completed by new software tools that provide the process information
in an interactive user-friendly environment which maximize the benefits of process
modelling. A 3D model of the heat transfer for numerical analyses should be used to obtain
valid results along the entire length of the kiln (Martin, et al., 2003; He, et al., 1996).

Fig. 22. System architecture

6. Conclusions
Process modelling is an effective and economical method of analyzing and diagnosing
process operations. The highly informative and detailed information provided by modelling
cannot be achieved by any other means.
Computer aided finite element modelling was used to predict temperature profiles and heat
fluxes involving linear properties of the exterior insulation materials and internal radiation

www.intechopen.com
Advanced Control Schemes for Cement Fabrication Processes 403

effects. Process modelling provides effective, safe, and economical ways of creating a
detailed database of information about the process. Data includes detailed fields of
temperature, heat transfer and a host of other important information.
The solution indicates that such a model may be used to identify problems of kiln operation.
Our numerical kiln model which uses existing heat transfer correlations for rotating kiln and
incorporates in-kiln combustion provides a tool for modelling temperature, calcining,
drying, in a rotating kiln. We use modelling to improve the clinker kiln operation.
Graphical representations of the results have been developed to facilitate the determination
of the gas temperature and the total heat flux to the wall and to the solids. We have also
achieved the emissions prediction and the simulation of the pollutants dispersion in the air.

7. References
Alyaser, A.H., (1998). Fluid Flow and Combustion in Rotary Kilns, PhD Thesis, Faculty of
Graduate Studies, Dept. of Metals and Materials Eng., University of British
Columbia.
Arad (Vlad), S. (2001). Assessment Environmental Impact from S C Casial SA Deva, M.Sc Thesis,
University of Petroşani, Petrosani, Romania;
Arad, S., Arad, V., (2002), Pollutant Dispersion simulation using Matlab Software for
Cement Factory Casial. Proc. SWEMP2002, Cagliari, Italia. DIGITA University of
Cagliari, ISBN 88-900895-0-4, pg. 963-969
Arad, S., Arad, V., (2003). Probabilistic model simulation in cement process fabrication at
Casial factory, Proceedings of ISARC2003, Technische Universiteit Eindhoven, CD-
ROM, ISBN 90-6814-574-6
Barr, P.V., et al (1989). A heat-transfer model for the rotary kiln: Part II. Development of the
cross -section model. Met. and Material Transaction B, Volume 20, No. 3 : 403-419,
ISSN 1543-1916
Barreto G., A., (1997). Lime kiln hybrid control system. Dynamic Modeling Control
Applications for Industry Workshop, 1997, IEEE Ind. Applications Society, Vol. Issue,
26-27/1997 :44 – 50, Vancouver B.C.
Boateng, A.A., Barr, P.V.,(1996), A thermal model for the rotary kiln including heat transfer
within the bed, Int. J. Heat Transfer Vol. 39 No. 10.
Bui, R.T., Simard, G., Charette, A., Kocaefe, Y., Perron, J., (1995). Mathematical Modelling of
the Rotary Coke Calcining Kiln”, Can. J. Chem. Eng. vol. 73: 534-544
Fireteanu, V., Popa, M., Tudorache, T., (2004). Modele numerice în studiu şi concepţia
dispozitivelor electrotehnice, Matrixrom Ed., Bucuresti
Gorog, J. P., et al. (1981). Radiative heat transfer in rotary kilns. Met. and Materials
Transactions B, Volume 12, Number 1 / March, 1981, 55-70, ISSN 1543-1916
He, P., Salcudean, M., Gartshore, I.S., & Nowak, P., (1996). Multigrid Calculation of the
Fluid Flows in Complex 3D Geometries using Curvilinear Grids, Comp. and Fluids
vol. 25: 395-419
Majumdar, K. S& Ranade, V.V. (2006). Simulation of Rotary Cement Kilns Using a One-
dimensional Model. Chem. Eng. Research and Design Vol 84(A3): 165-177, ISSN 1744-
3563

www.intechopen.com
404 Robotics and Automation in Construction

Martin, J.K., et al. (2003). Modelling heat transfer through a novel design of rotary kiln. Int.
Journal of Comp. Applications in Technology - Vol. 17, No.4 pp.233-243
Georgallis, M., et al. (2001). Mathematical modelling of lime kilns. Internal report, University
of British Columbia, Vancouver, Canada
Passino, K. M., Yurkovich, St., (1998). Fuzzy Control. Addison-Wesley Longman, Inc,
California 94025, ISBN 0-201-18074-X
Procese de productie, (2008). www.heidelbergcement.ro/procese-de-productie.html

www.intechopen.com
 Robotics and Automation in Construction
Edited by Carlos Balaguer and Mohamed Abderrahim

ISBN 978-953-7619-13-8
Hard cover, 404 pages
Publisher InTech
Published online 01, October, 2008
Published in print edition October, 2008

This book addresses several issues related to the introduction of automaton and robotics in the construction
industry in a collection of 23 chapters. The chapters are grouped in 3 main sections according to the theme or
the type of technology they treat. Section I is dedicated to describe and analyse the main research challenges
of Robotics and Automation in Construction (RAC). The second section consists of 12 chapters and is
dedicated to the technologies and new developments employed to automate processes in the construction
industry. Among these we have examples of ICT technologies used for purposes such as construction
visualisation systems, added value management systems, construction materials and elements tracking using
multiple IDs devices. This section also deals with Sensorial Systems and software used in the construction to
improve the performances of machines such as cranes, and in improving Human-Machine Interfaces (MMI).
Authors adopted Mixed and Augmented Reality in the MMI to ease the construction operations. Section III is
dedicated to describe case studies of RAC and comprises 8 chapters. Among the eight chapters the section
presents a robotic excavator and a semi-automated façade cleaning system. The section also presents work
dedicated to enhancing the force of the workers in construction through the use of Robotic-powered
exoskeletons and body joint-adapted assistive units, which allow the handling of greater loads.

How to reference
In order to correctly reference this scholarly work, feel free to copy and paste the following:

Susana Arad, Victor Arad and Bogdan Bobora (2008). Advanced Control Schemes for Cement Fabrication
Processes, Robotics and Automation in Construction, Carlos Balaguer and Mohamed Abderrahim (Ed.), ISBN:
978-953-7619-13-8, InTech, Available from:
http://www.intechopen.com/books/robotics_and_automation_in_construction/advanced_control_schemes_for_
cement_fabrication_processes

InTech Europe InTech China

University Campus STeP Ri Unit 405, Office Block, Hotel Equatorial Shanghai
Slavka Krautzeka 83/A No.65, Yan An Road (West), Shanghai, 200040, China
51000 Rijeka, Croatia
Phone: +385 (51) 770 447 Phone: +86-21-62489820
Fax: +385 (51) 686 166 Fax: +86-21-62489821

www.intechopen.com
www.intechopen.com
© 2008 The Author(s). Licensee IntechOpen. This chapter is distributed
under the terms of the Creative Commons Attribution-NonCommercial-
ShareAlike-3.0 License, which permits use, distribution and reproduction for
non-commercial purposes, provided the original is properly cited and
derivative works building on this content are distributed under the same
license.