CC BY
6«CQyyOMUM-JMTMaiL» #30053), 20222 / TECHNICAL SCIENCE
15
TECHNICAL SCIENCE
UDC 66-9
Bakhteev E.M.,
Post-graduate student Rutkovskiy A.L.,
Professor, Doctor of technical sciences
Butov H.A.,
Post-graduate student
Department of Non-Ferrous Metallurgy and Automation of Metallurgical Processes
North Caucasian Institute of Mining and Metallurgy
Vladikavkaz, Russia DOI: 10.24412/2520-6990-2022-30153-15-19
OPTIMIZATION OF THE CHARGE PREPARATION PROCESS IN THE PRODUCTION OF
TITANIUM PELLETS
Abstract.
The problem of mathematical modeling and control of the charge preparation process in order to optimize the technological regime is considered. Using the methods of regression analysis, the relationship between the gas permeability of the charge layer, the shrinkage of the layer, the bulk mass and the speed of movement to the drying zone on its moisture content is established. An expression is found for the optimal moisture content of the charge.
On the basis of theoretical and experimental data, a computer simulation program has been compiled which makes it possible to determine the estimate of the optimal parameters in the process ofpreparing the charge for the production of titanium pellets.
Keywords: mathematical modeling, charge, pellets, gas permeability, optimization, roasting machine.
Introduction. Titanium-containing materials are used in many areas of industrial production. In particular, they are widely used in transport and chemical engineering, aerospace engineering and other industries due to their high specific strength, fatigue resistance, fracture toughness, and corrosion resistance [1]. At the same time, the production of such materials is characterized by high energy intensity and a significant amount of hard-to-recycle waste from titanium metallurgical production [2].
One of the main directions in the development of metallurgy in modern conditions is not only the re-equipment of the industry through the introduction of new technologies, but also the reconstruction, modernization, automation and computerization of existing and construction of new metallurgical units. To ensure the growth of production and the competitiveness of products in the metal market, it is necessary to constantly improve the organization of preparation in high-quality raw materials [3].
As a rule, mathematical modeling has long been of decisive importance in understanding and predicting technological processes in heavy industry and metallurgy.
The purpose of modeling any technological process is to establish a quantitative dependence of the output parameter on one or a group of input parameters that can change randomly [4].
An effective solution of the problem of optimal control of any complex object, as a rule, is associated with a fairly complete consideration of the physical nature of the processes occurring in it and an analysis of their features.
In this case, the primary task is to build a mathematical model of the control object, which allows the choice of the structure and parameters of the control system, the formation of optimality criteria and constraints, the solution of forecasting problems, etc. [5, 6].
The optimal gas permeability of the charge production. The initial charge, consisting of return, concentrate and finely ground limestone, is fed from the bunkers to the conveyor. The amount of charge is determined by the dispensers located at the bottom of the bins. The mixture is mixed in the drum, then bentonite is added to the mixture for bonding and water. Pelletiz-ing takes place on the plate granulator. Finished raw pellets are fed to a belt sintering machine. On the conveyor of the machine, the pellets pass through three zones: I - heating, drying; II - firing; III - cooling. In the drying zone, the pellets are heated up to 400 °C by the heat of gases from the roasting and cooling zones. In the firing zone, the temperature reaches 1200-1350 °C due to gas combustion.
The charge is heated from above, in the cooling zone, cold air is blown through the pellet bed. The cooled pellets are unloaded on the screen. Fractions <10 mm are sent for recycling [7].
An important step in the production of pellets is the process of moistening the charge (feedstock) to achieve its optimal gas permeability in order to increase the productivity of sintering machines and improve the quality of the finished product. The gas permeability of the charge depends to a large extent on its absolute humidity. For a material of a given mineralogical and granulometric composition, there is only one optimal moisture value, which must be maintained manually or
16
TECHNICAL SCIENCE / «<g©yy©®UOUM=J0Umay» #3@Î1H), 2022
automatically with the possible degree of accuracy [8, 9].
The moisture content has a very strong influence on the gas permeability of the layer of pelletized charge. The results of the study by G.I. Volkovitsky, N.M. Yakubtsiner, Yu.P. Smirnov, N.Z. Plotkin, V.I. Korotich [10] showed that when the humidity changes, the properties of the pelletized charge change significantly: bulk mass, lump strength, layer structure and its gas permeability. The typical nature of the dependence of these parameters on the moisture content of the mixture showed that with a gradual increase in the moisture content of the mixture, its bulk density decreases (the porosity of the layer increases), and gas permeability increases. The charge reaches a minimum value upon reaching a certain moisture content of the bulk mass, with a further increase in moisture content, the porosity of the layer begins to fall, and the gas permeability continues to grow. This dependence is characterized by a change in the pelletizing regime and, accordingly, a change in the charge structure [11, 12]. Subsequently, after analyzing the results, the authors came to the conclusion that the optimum moisture content of the charge at which the maximum vertical sintering rate is obtained is approximately 1% less than the moisture content corresponding to the highest gas permeability of the initial charge.
Experimental details. The results of this study make it possible to obtain mathematical models with the help of which it is possible to find, close to the initial value, the optimal values of moisture content and gas-dynamic parameters of the charge.
The construction of approximating and interpolating functions is carried out using the capabilities of the Mathcad program. To process the results based on the known experimental data, the methods of general regression and polynomial regression were used [13, 14].
The models with the highest accuracy were finally accepted.
As a rule, regression is very effective when the data distribution law (x1, y1) is known in advance (or at least well guessed) [15].
The following are the results performed in the Mathcad system, preparation of the initial data, general regression, polynomial regression, calculation of interpolation errors and a graphical illustration of the results of data processing calculations.
On the basis of experimental and calculated data, the coordinates of the initial points are formed in the form of a matrix. Mathematical processing of the experimental results made it possible to obtain the following regression equations relating the moisture content of the charge to the rate of air suction through the charge layer (1), bulk mass (2), layer shrinkage (3) and the speed of movement of the drying zone (4).
Results and discussion. The experimental data given in [10] give graphical expressions for the relationship between the following process parameters: air suction rate d, bulk mass d, layer shrinkage Sh, and drying zone movement speed t91 depending on the batch moisture content w.
As a result of processing experimental data, the following regression equation was obtained:
i9(w) = 2,493 • 10-3 •1 + 3.877 • 10" 4.066 • 10-4 • e0>
• œ
"4-ew (1)
where R = 0,99 - correlation coefficient, Fcalc = 47,422, Ftable = 4,876 - calculated and tabulated values of the Fisher criterion for the confidence level y = 0,95.
On Fig. 1, on the basis of equation (1), a graph of the dependence of the charge moisture œ on the air suction rate t9 is shown. From Fig. 1 that the resulting model agrees with the experimental data.
Fig.1. Graph of influence of moisture content of sinter charge w (%) on air suction rate d (m/sec).
The influence of the charge moisture w on the bulk mass d is approximated by the following regression equation (2):
d(w) = 8.678 • 10-4 • w3 + 4.357 • 10-3 • m2 - 0.125 •w + 2.262 (2)
where R = 0,998 - correlation coefficient, fca£c = 322,434, Ftafcie = 4,876 - calculated and tabulated values of the Fisher criterion for the confidence level y = 0,95.
The calculation results are shown in Fig. 2.
«CQyyOMUM-JMTMaiL» #3(01151)), 2022 / TECHNICAL SCIENCE
17
Fig.2. Graph of influence of moisture content of sinter charge m (%) on bulk mass of charge d (kg/dm3).
The effect of charge moisture w on layer shrinkage Sh is represented by the following regression equation (3):
Ah(w) = -0.05 • œ3 + 0.348 • œ2 + 0.26 • œ + 0.646 (3) where R = 0,992 - correlation coefficient, fca£c = 61,758, FtaMe = 3,388 - calculated and tabulated values of the Fisher criterion for the confidence level y = 0,95.
On Fig. 3, based on the regression equation (3), a graph of the dependence of the charge moisture w on the shrinkage of the layer 5ft is shown.
(4.956:4.396) *
Charge moisture, %
-Model
♦ ♦ Experiment
• •• Extremum point
Fig.3. Graph of the influence of moisture content of the sinter charge m (%) on the shrinkage of the charge layer
5h (mm).
The effect of charge moisture m on the speed of movement of the drying zone is represented by the following regression equation (4):
^(w) = 0.064 • w3 - 2.242 • w2 + 20.386 •
M
42.536
(4)
where R = 0,995 - correlation coefficient, Fcaic = 96,03, Ftafcie = 4,876 - calculated and tabulated values of the Fisher criterion for the confidence level y = 0,95.
The calculation results are shown in Fig.4.
» Extremum point
Fig.4. Graph of the influence of moisture content of the sinter charge m (%) on the speed of movement of the
drying zone (m/sec).
18
TECHNICAL SCIENCE / «<g©yy©®UOUM=J0Umay» #30053), 2022
Installation. However, when the process is affected by uncontrolled perturbations (granulometric composition of the charge, its temperature, the height of the charge layer), the position of the extremum will shift. To compensate these effects, it is proposed to equip the roasting machine with a measuring vacuum chamber (MVC), which is installed at the machine inlet in front of the first working chamber. MVC is equipped with an individual pump for air suction (Fig. 5).
The vacuum in the MVC is stabilized by an automatic control system. Under these conditions, the air flow through the MVC fully corresponds to the gas permeability of the charge layer. The amount of air flow through the MVC is fed to the input of the extreme regulator that controls the water supply to the pelletizer. Such a system continuously maintains the maximum gas permeability of the charge bed and the best process performance.
The device for controlling the preparation of the charge in the production of titanium pellets includes a
granulator 1, a screen 2 and a roasting machine containing heating and drying zones, a roasting zone and a cooling zone with the corresponding vacuum chambers 3, 4 and 5, as well as individual vacuum pumps 6, 7 and 8. The device is additionally equipped with a measuring vacuum chamber 9 connected to an individual vacuum pump 10 and installed on the roasting machine in front of the vacuum chamber 3 of the drying zone. Measuring vacuum chamber 9 is equipped with a vacuum stabilization system and an air flow control system. At the same time, the vacuum stabilization system includes a vacuum control sensor 11, an actuator 12 and a damper 13 installed on the vacuum chamber 9, and the air flow control system includes an air flow control sensor 14 and an extreme regulator 15 connected through the actuator 16 with the control damper 17 water consumption.
Fig.5. A fragment of a roasting machine with a measuring vacuum chamber (MVC) installed
Working principle. The device operates as follows: the mixture is fed to the granulator 1, where it is pelletized. The raw pellets then enter the drying zone of the roasting machine. Using a vacuum pump 10, air is sucked through a layer of raw pellets into a measuring vacuum chamber 9, in which the vacuum is stabilized by a vacuum stabilization system. At the same time, the vacuum value is recorded by the vacuum control sensor 11, which, in turn, sends a signal to the actuator 12 that controls the damper 13. By changing the position of the damper 13, the vacuum stabilization system maintains its value in the vacuum chamber 9 constant. At a constant vacuum in the vacuum chamber 9, the air flow through it depends on the gas permeability of the pellet bed. To control the gas permeability, the signal from the sensor 14 for controlling the flow of air passing through the vacuum chamber 9 is fed to the input of the extreme regulator 15, which through the actuator 16 with the help of the damper 17 regulates the water flow to the inlet of the granulator 1. The air flow control system provides the supply of such an amount water on the
granulator 1, which achieves the optimal gas permeability of the layer of pellets, determined by the maximum air flow through the vacuum chamber 9 at a constant vacuum in it. Pellets with optimal gas permeability after the heating and drying zone move through the roasting machine through the roasting and cooling zones, in which pellets are sintered and hardened. The high-quality finished product after the roasting machine is unloaded onto screen 2. The device for preparing the charge in the production of titanium pellets will allow you to quickly adjust the gas permeability of the layer even when the charge composition changes, which will significantly improve the quality of the pellets at the exit of the roasting machine and reduce the amount of product sent to circulation.
This device will simplify the management of the firing process, improve the quality of the finished product due to the operational control of the elements of the device and optimization of the gas permeability of the charge layer.
Conclusion. Thus, the use of the experimental data in combination with modeling made it possible to
«cqyyomum-jmtmail» #mmm, mm / technical science
19
obtain a relationship between the moisture content of the charge and the rate of air suction; dependence of charge moisture content with bulk mass; layer shrinkage and the speed of movement of the drying zone, and a correlation between them is found.
During the processing of the initial data by means of Mathcad functions, two versions of the regression equation were obtained: general and polynomial.
The adequacy of the regression equations (1)-(4) was confirmed using the Fisher criterion.
The high value of the correlation coefficient (fl = 0,99) confirms this result.
The mathematical model of the pelletizing process can be used to develop mathematical models of many technological processes carried out in sinter machines. To compensate for uncontrolled influences, it is proposed to equip the firing machine with a measuring vacuum chamber (MVC), which is installed at the inlet of the machine in front of the first working chamber. MVK is equipped with an individual pump for sucking air. The vacuum in the MVC is stabilized by an automatic control system. Under these conditions, the air flow through the MVC fully corresponds to the gas permeability of the charge layer. The amount of air flow through the MVC is fed to the input of the extreme regulator that controls the water supply to the pelletizer. Such a system continuously maintains the maximum gas permeability of the charge bed and the best process performance.
References
1. Ilyin A.A., Kolachev B.A. and Polkin I.S. Titanium alloys. Composition, structure, properties: reference book. Moscow, VILS-MATI, 2009, 520 p.
2. Andreev A.L., Anoshkin N.F. and Bochvar G.A. Titanium Alloys. Melting and Casting of Titanium Alloys. Moscow, Metallurgiya, 1978, 383 p.
3. Torgovets A.K., Pikalova I.A. Agglomerate production: technology and equipment. Temirtau, KSIU, 2013, 134 p.
4. Nuradinov A.S., Uzdieva N.S., Balatkhanova E.M. and Tepsaev A.N. Mathematical modeling of the
processes of the formation of cast units in the continuous casting of steel. Herald of GSTOU, Technical sciences Volume XV, № 2 (16), 2019, pp. 56-66.
5. Arunyants G.G., Rutkovskiy A.L., Alborov E.K., Sidorov D.V. and Didenko A.A. Estimation and correction of parameters of mathematical models in adaptive control of non-stationary technological processes. Izv. Univer. North Caucasus region, Tech. sciences, Special issue, 2004, pp. 47-50.
6. Ageev N.G. Modeling of Processes and Objects in Metallurgy. Ekaterinburg, UrFU, 2016, 108 p.
7. Linchevsky B.V., Sobolevsky A.L. and Kal-menev A.A. Metallurgy of ferrous metals: textbook for technical schools - 2nd ed., revised and additional. Moscow, Metallurgy, 1986, 360 p.
8. Moraes S., Lima T. and Ribeiro JR. Iron Ore Pelletizing Process: An Overview, Iron Ores and Iron Oxide Materials, Chapter 3, London, IntechOpen, 2018, 280 p.
9. Deepak N., Nigamananda R., Nilima D., RathSwagat S., PatiSoobhankar S. and Partha De. An Optimal Route for the Preparation of Metallized Composite Pellets from Ilmenite Concentrate. Journal of Sustainable Metallurgy, № 7, 2021, pp. 1102-1115.
10. Korotich V.I. and Puzanov V.P. Gas dynamics of the sintering process. Moscow, Metallurgy Publishing House, 1969, 208 p.
11. Rutkovskiy A.L., Salikhov Z.G., Kovaleva M.A. and Bakhteev E.M. An optimized process of drying titanium pellets in a tunnel kiln. Tsvetnye Metally, № 5, 2021, pp. 89-94.
12. Komiyama K., Guo M. and Zoughbi B-Y.H. Numerical analysis of titanium compounds in blast furnace hearth during titania addition. Steel Research International, № 6, 2014, pp. 592-603.
13. Vegman E.F. Theory and Technology of agglomeration. Moscow, Metallurgy, 1974, 208 p.
14. Lvovskiy E.N. Statistical methods for constructing empirical formulas. Moscow, Vysh. shk. Publ., 1988, 239 p.
15. Kiryanov D.V. Mathcad 14. Sankt Petersburg, BHV-Petersburg, 2007, 704 p.
