Control and Optimization for the Cowpers of a Steelmaking Plant
D. Popescu, I. I. Luican, S. Stamatescu
University “Politehnica” of Bucharest
Department of Automatic Control and Computer Science
313 Splaiul Independentei, Sector 6, 060041-Bucharest, Romania
E –mail :dpopescu@indinf.pub.ro
Abstract-The paper presents the results of the
research performed by the authors on systems
control and optimization for the operating process of
the cowpers from the ISPAT-SIDEX steel plant. This
system was developed on two relevant levels
interconnected in a hierarchical control structure.
The acquisition and control level was designed using
specialized microcontrollers. The supervisor level for
the optimization of the combustion process was
implemented on an operator console. The solution of
the optimization problem represents the optimal
decision, translated in real-time procedure to the
acquisition and level control.
I. INTRODUCTION
The complexity of the metallurgical installations and
the difficulties of planning and technological functioning
are well-known. Significant improvements in MMM area
have been obtained when numerical equipments and
modern theory of automatic control were introduced
[Taloi, 1993], [ Avoy, et al., 2002].
In economic and commercial environments, quality
and performance are very important criteria. The
priorities in MMM processes are productivity, raw
materials and the quality of products.
In this context, at ISPAT-SIDEX, an important steel
plant in Eastern Europe, a program of modernization was
launched in order to feed the plant’s blast furnaces with
hot air from the cowpers ensemble.
The cowper’s operating process has three work
phases: heating, aeration and cooling. Using efficient
synchronization and adequate technological switching,
the cowpers are continuously feeding the blast furnace
with hot air.
Some particularities of the process can be noticed.
The large dimensions of the installation imply a plant
model with large delays and distributed parameters,
engaging important flow materials.
The used fuel has many components: methane gas,
coke gas and furnace gas, with different caloric powers.
A convenient recipe must be calculated in order to feed
the burners.
The quality of the combustion gas and the process
nonlinearities introduce important disturbances in
exploitation. To evaluate the combustion process, the
composition of the flue gases is analyzed; more precisely,
the concentration of O2 and CO are measured and
computed.
Our major interest was to improve the cowper’s
efficiency using an adequate automation solution.
The work has been focused on two main directions:
- Design of a data acquisition and control system in order
to maintain the installation in a nominal operating point;
- Optimization of the burning process, important
consumer of fuel gas.
The old conventional control solution, based on
analogical systems [Weight and Scrimgeour, 1962], was
replaced with numerical control. The numerical solution
was conceived using the model based - control design
procedure, by poles-allocation methods for PID
algorithms [Popescu et al., 1989].
During the identification step, LSR methods were
introduced, using the standard algorithms:
0
0
ˆ ˆ ( 1) ( ) ( 1) ( ) ( 1),
( ) ( ) ( ) ( ) ( 1) ( ) ,
1 () ( )( )
ˆ ( 1) ( 1) ( ) (),
T
T
T
k k Fk k k k
F k k k F k F k F k k
k F k k
k yk k t k
? + = ? + + f e + ? ?
f f
+ = - ? ?
+ f f
e + = + -? f ? ?
¥
¥
¥
(1)
with the following initial conditions:
1 (0) ( ) ,0 1 F I GII = = <d<
d
(2)
The estimated ) ( ˆ k ? represents the parameters of the
polynomial plant model.
In the control design phase, the RST control
algorithms were evaluated by poles allocation methods,
covering both reference tracking and disturbances
rejection:
1 1 ( ) ( ) ( ) ( ) ( ) 1 1 ( ) ( )
T z R z u k r k y k
S z S z
- -
= - - -
(3)
Before the implementation stage, the performances of
the designed systems have been verified by simulation.
The authors had to make some improvements of the
.....