Mei C., Zhou J., Peng X. Simulation and Optimization of Furnaces and Kilns for Nonferrous Metallurgical Engineering
Подождите немного. Документ загружается.
5 Hologram Simulation of Aluminum Reduction Cells
current/voltage load in aluminum reduction cells (in Chinese).
Light Metals: 131
Liu Yexiang, Mei Chi et al (1996) Computer simulation of electromagnetic force field on
the melt in alumina reduction cell (in Chinese).
Transactions of Nonferrous Metals
Society of China
, 6(1):27
Lu Jiayu et al (1986) Computation of current field in large aluminum reduction cells with
use of boundary element method (in Chinese). In:
Thesis Collection of Aluminum
Reduction Cells “Three fields”
, 9: 73
Mei Chi, Tang Hongqing (1986) Computation analysis and simulation test of electric and
thermal fields in aluminum reduction cells (in Chinese).
Journal of Central South
Institute of Mining and Metallurgy
, (6): 29
Mei Chi, Wang Qianpu et al (1992) Study of thermal field in aluminum reduction cells (in
Chinese).
Light Metals, (1): 29
Mei Chi, Wang Qianpu et al (1996) Optimization of inner lining structure in 80kA
self-baking aluminum reduction cells (in Chinese).
Light Metals, (4): 21
Mei Chi, You Wang et al (1997) Study and development of the simulation software for the
freeze profile in aluminum reduction cells (in Chinese).
Journal Of Central South
University of Technology
, 28(2): 138
Mei Chi, Zhou Naijun et al (1998) Dynamic prediction of bath temperature and online
simulation of freeze profile in aluminum reduction cells (in Chinese).
Light Metals,
(supplement): 16
Morris S J Sˈ Davidson P A (2003) Hydromagnetic edge waves and instability in
reduction cells.
Fluid Mech., 493(2):121~130
Peacey J G, Medlin G W (1979) Cell sidewall studies at Noranda aluminum.
Light Metals,
(1): 475
Perutne (1982) Freeze profile in side-break, cells-calculations and measurements.
Light
Metals
: 359
Qiu Jie (1992) Computation of nonlinear magnetic field in aluminum reduction cells (in
Chinese).
Nonferrous Metal, (3): 55
Qiu Zhuxian (1982) Aluminum Electrolysis (in Chinese). Metallurgical Industry Press,
Beijing, 292
Qiu Zhuxian (1988) Aluminium Metallurgy In Pre-Baked Cells (in Chinese). Metallurgical
Industry Press, Beijing ,244
Qiu Zhuxian et al (1981) Mathematic model of current efficiency in industrial aluminum
Reduction Cells (in Chinese).
Light Metals, (9): 23
Richard M C (1976) A new approach for comparing the impact on magnetic fields from pot
design alterations.
Light Metals, (1): 109
Romanov B P et al (1980)
Non-Ferrous Metals (in Russian),(1):66
Robl R F (1978) Influence by shell steel on magnetic fields within Hall-Heroult cells.
Light
Metals
, (1):1
Robl R F, Haupin W E, Sharma D et al (1977) Estimation of current efficiency by a
mathematical model including hydrodynamics parameters
. Light Metals: 185
Naijun Zhou
Sele T (1974) Computer model for magnetic fields in electrolytic cells including the effect
of steel parts.
Metal. Trans: 2145
Severo D S et al (2005) Modeling magnetohydrodynamics of aluminum electrolysis cells
with ANSYS and CFX (in Chinese).
Light Metals: 475~480
Shen Hongyuan, Mei Chi, Cai Qifeng (1992) Mathematic model of current efficiency
determined by two dimensional turbulent mass transfer in aluminum reduction cells (in
Chinese).
Journal of Central South Institute of Mining and Metallurgy, 23(6): 693
Shen Xianchun et al (1994) Physical simulation and experimental determination of
magnetic field in aluminum reduction cells (in Chinese).
Light Metals, (2): 26
Sneyd A D (1985) Stability of fluid layers carrying a normal electric current.
Fluid Mech:
223
Solhiem A et al (1983) Heat transfer coefficients between bath and side ledge in aluminum
cells.
Light Metals: 425
Sun H, Zikanov O, Donald P et al (2004) Non-linear two-dimensional model of melt flows
and interface instability in aluminum reduction cells.
Fluid Dynamics Research,
35:255~274
Sun H, Zikanov O, Finlayson B A (2005) Effect of background melt flow and interface
distortion on the stability of Hall-Heroult Cells.
Magnetohydrodynamics, 41(3):273~287
Sun Ru, Mei Chi (1989) Numerical simulation of the circulation of molten metal In
aluminum reduction cell.
Journal of Central South Institute of Mining and Metallurgy
(in Chinese), 20 (6): 618
Tarapore E D (1982) The effect of operating variables on flow in aluminum reduction cells.
Metals, 34 (2): 541
Vanvoren C et al (2001) AP50: the pechiney 500kA cell.
Light metals: 221~226
Vao Shihuan (1990) Selection of bus bar optimum section in high amperage reduction cells.
Light Metals: 453
Wang Hua (1993) Mathematic model for optimization of current efficiency in 60kA side
stud cells (in Chinese).
Light Metals, (1): 24
Wang Huazhang et al (1982) Mathematic model of current efficiency in 24kA self-baking
anode aluminum reduction cells (in Chinese).
Light Metals, (8): 24
Wang Ruliang, Yang Xiaodong et al (1998) Optimization of bus bar configuration and cell
structure in 280kA pre-baked cells (in Chinese).
Light Metals, (Supplement): 58
Wu Jiangkang, Huang Ming, Huang Jun et al (2002) Finite element analysis of
incompressible viscous flow with moving free surface by selective volume of fluid
method.
Light Metals: 511~514.
Xue Yuqing et al (2006) Normal temperature analogue experiment of anode bubble’s
behavior in aluminum electrolysis cells (in Chinese).
The Chinese Journal of Nonferrous
Metals
, 10:1823~1828
Yang Jinxin (2007) Where is the right way for China’s non-ferrous metals industry to
save energy and reduce consumption (in Chinese).
Energy Saving of Non-Ferrous
Metallurgy
:3
5 Hologram Simulation of Aluminum Reduction Cells
You Wang (1997) Online dynamic simulation of freeze profile in large pre-baked
aluminum Reduction cells (Docteral Dissertation, in Chinese). Central South University
of Technology , Changsha
You Wang, Wang Qianpu et al (1998) On line dynamic simulation of freeze profile in
aluminum electrolysis (in Chinese).
The Chinese Journal of Nonferrous Metals, 8(4):
695
Zeng Shuiping (1996) Computation of electro-magnetic field in aluminum reduction cells
and continuous monitor of current efficiency (Docteral Dissertation, in Chinese).
Changsha: Central South University of Technology
Zeng Shuiping (2004) Mathematical model of current efficiency with anode current
distribution in aluminum reduction cells(in Chinese).
The Chinese Journal of Nonferrous
Metals
, 4:681~685
Zeng Shuiping, Cai Qifeng et al (1996) Study of bus bar configuration in vertical-stud
aluminum reduction cells (in Chinese).
Light Metals, (2): 32
Zhou Keding (1983) Basic theory of boundary element analysis of electro-magnetic field
(in Chinese).
Micro Motor, (3)
Zhou Naijun (2006) Effect of electromagnetic force and anode gas on electrolyte flow in
aluminum electrolysis cell.
Journal of Central South University of Technology,
5:496~500
Zhou Naijun, Mei Chi et al (1998) Dynamic numerical simulation of melt temperature in
large pre-baked aluminum reduction cells (in Chinese).
Proceedings of China
Engineering Thermophysics Annual Conference,
Hefei
Zhou Ping, Mei Chi (1992) Continuous monitor of freeze profile in aluminum reduction
cells (in Chinese).
Light Metals, (4):19
Zuo Jun (1996) Dynamic simulation of thermal field in aluminum reduction cells (Master
Dissertation, in Chinese). Central South University of Technology, Changsha
Simulation and Optimization of
Electric Smelting Furnace
Jiemin Zhou and Ping Zhou
In this Chapter, the simulation models, including sintering process model of
self-baking electrode, flow field model and temperature field model in molten pool
of electric smelting furnace are built, and the cases in practice are employed to
optimize them. Eventually, the computational software is developed to solve the
physical fields and production condition of the electric furnace is improved.
6.1
Introduction
Electric smelting furnace, also called submerged arc furnace or electric arc
resistance furnace, is mainly applied in the smelting of nonferrous metals,
ferroalloys, calcium carbide and yellow phosphorus. Ferroalloy furnaces are usually
designed to be circular, and three electrodes insert into the charge forming a regular
triangle. Most electric furnaces for smelting of nonferrous metal are rectangular,
others are circular or elliptic. In these rectangular or elliptic electric furnaces, three
or six electrodes are distributed in line. Current is conducted to the furnace slag or
charge through electrodes, and the voltaic arc is produced at the interface of
electrode and slag or charge, where most of the heat is released. When the currents
flow through the slag or charge, the triangular and star-shaped circuits are formed,
and Joule heat is produced. The resulting high temperature then melts the furnace
charge, accompanied by various physical and chemical reactions, which enriches the
valuable element. In nonferrous metallurgy, the electric smelting furnace is mostly
used in the smelting of copper, nickel and tin, and the impoverishment of slag, or
used as electrically-heated settler for separation of metal and slag.
Jiemin Zhou and Ping Zhou
Electric smelting furnace has the merits of high smelting temperature, relatively
little fume, high recovery rate of metal, easy control and so on. However, it
consumes electric energy that is converted from other kind energy, therefore, the
high power consumption and high cost are its primary disadvantages. Meanwhile,
large electric smelting furnaces commonly use self-baking electrode, which may
result in accidents such as soft-broken and hard-broken, and the operation is often
affected. In recent years, as the price of energy increases because of the short
supply, investigation on saving energy and improving the quantity and quality of the
production has become a hot topic for all technicians and theoretical researchers
working in the field of electric smelting furnace.
The performances, as well as technical and economic indices, of electric
smelting furnace are affected by many factors, such as chemical compositions and
physical characters of the charge, the way to prepare the charge, the electric
condition, operation condition, the furnace configuration and so on; that is, the
operation parameters and the configuration parameters of the furnace. These
variables influence the furnace thermo-technical process parameters including
current distribution, electricity-heat conversion, fluid flow, heat transfer,
temperature distribution, etc. Such thermo-technical process parameters directly
determine the furnace production indices, for instance, the production capacity,
power consumption per unit production, and recovery of valuable metals, etc.
The development of computer and calculation method makes it possible to study the
heat transfer process, momentum transfer process and mass transfer process in the
furnace quantitatively, which means we can simulate the correlation between operation
parameter, configuration parameter and thermo-technical process parameter as well as
various production indices, and further detect rules of the production process in electric
smelting furnaces to improve the equipments, optimize the operations and improve the
furnace performances and production indices.
This chapter mainly investigates the sintering process of self-baking electrode in
large scale electric smelting furnace, the melt flow process in slag impoverishment
˄depriving˅electric furnace and the heat transfer process in molten pool of ore
smelting electric furnace. Then, relevant software is introduced to simulate and
optimize the real electric smelting furnace, and specific approach is presented to
improve the configuration parameters and operation condition.
6.2
Sintering Process Model of Self-baking Electrode in Electric
Smelting Furnace
The self-baking electrode(Fig.6.1), or Soderberg electrode, which has a lower cost
than the prebaked electrode, is constituted of electrode paste and steel plate. It is
the most important part of the electric smelting furnace and is baked to be a solid
6 Simulation and Optimization of Electric Smelting Furnace
electrode during the smelting process inside the furnace. Having to conduct very
huge current, and to endure conditions such as oxidization of high temperature in
the furnace, sharp cooling of clamp water-cooled system, temperature variation
during the start-up and shut-down of the furnace, and the mechanical impact of the
charge, the electrodes must have excellent electrical conductivity, enough
high-temperature mechanical strength and relatively high oxidizing temperature.
To a great extend, the normal operation of the furnace depends on the electrodes’
working situation.
In order to reduce the malfunction of electrodes and to improve the energy
efficiency, a rational electrode baking rule, that is, a rational temperature
distribution must be maintained.
There are two ways to study the temperature distribution in the electrode: by
measurement and by analytical modeling.
Generally, measurements of the electrode temperature are carried out by
inserting thermocouples into the electrode at different depth. During the descent of
the electrode, the temperature from the thermocouple can be recorded continuously
by the secondary meter, and credible data can be obtained. However, it is difficult
to maintain a stable electrode temperature during several days’ measurements
when the electrode is descending. Thus the testing error is unavoidable due to the
variation of the processing conditions. As a result, when to exam the influence of
different operating conditions on the temperature field of the electrode, a large
amount of experiments have to be carried out, and it’s time-consuming. Especially,
when the equipment configurations are to be changed, such as the dimension and
the amount of the fin, or the distance between the clamp and the roof of the
furnace, it is tedious and needs more human and financial resources. In condition
that we adopt mathematical method to computer and display the temperature
distributions which will appear in different design approaches, and then predict the
electrode baking process as well as its operation performance, both the time and
the cost of the experiments will be greatly reduced. This is what we called
analytical modeling method.
The electrode temperature field mainly depends on the power distribution in the
electrode, i.e. the electrical field, therefore, the electric analysis and thermal
analysis are bonded together.
R. Innvaer et al. of Elkem Metal Company in Norway are the vanguards in the
research of electric analytical and thermal analytical models of self-baking
electrode in the electric smelting furnace. Since 1972, they have developed five
self-baking electrode mathematical models, including two static models, one
dynamic model, one stress model, and one three-dimensional analytical model. Of
the static models, one is based on direct current and called Elkem-S, and the other
is based on alternating current, called Elkem-X. These two models are generally
used to compute the electrode temperature field and baking zone during the stable
Jiemin Zhou and Ping Zhou
operation in order to solve the electrode soft-broken problem. The dynamic model
(Elkem-D) is used to compute the temperature field when operation parameters
varying with time when to investigate the electrode soft-broken and hard-broken
problem. The stress model (Elkem-T) can calculate the thermal stress in both
steady and unsteady state to study the hard-broken problem for the baked
electrodes. These models have become powerful tools in production and design
(Olsen et al., 1972, 1976; Innvaer et al., 1976, 1980). However, they are all
two-dimensional models, and did not take the electric and thermal uniformity
along the perimeter of the electrode caused by the fins into account. As an
improvement, R.Innvaer et al. developed a three-dimensional analytical model
(Elkem-3X) of the self-baking electrode, in which the influences of electrode
descent velocity, clamp current, furnace temperature, material and thickness of the
electrode shell and the fin’s length on the temperature distribution of the
electrode’s cross section and the electrode baking process were considered. They
also used the Elkem-3X model to simulate a new electrode holder, and obtained
some valuable data(Innvaer et al., 1986). This type of electrode holder was later
sold to China (Vatland et al., 1987).
Because in the solution procedure of the Elkem-3X model, the influence of
alternating magnetic field on alternating electric field is considered, its result is more
precise. However, as it needs to solve the electric field, magnetic field and temperature
field simultaneously, the computing time and cost all
increase. Therefore, Elkem-3X model is usually used only
to calculate part of the self-baking electrode.
There are some other literatures that reported
research of the self-baking electrode, but they are not
as systematic and consummate as the Innvaer’s
Elkem-3X model. For such literatures, please refer to
the references (Heiss, 1981; Bochmann et al., 1968
Cavigli, 1978; Ambrosio, 1981).
There is no investigation on the analytical model of
self-baking electrode before in China.
6.2.1
Electric and thermal analytical
model of the electrode
With the three-dimensional model, multiple adjustable
parameters, large amount of information and more
precise result can be obtained. However, if we
excessively pursue the precision of the solution, the
calculation time and cost will be remarkably increased,
which will limit the model’s application. Therefore, a
Fig.6.1
Schematic of the
self-baking electrode
6 Simulation and Optimization of Electric Smelting Furnace
relatively coarse grid is adopted to simplify the calculation of the electric field
when the model is developed. Compared to literatures, further investigation and
development have been carried out to improve the output form of the result
(Zhou,1991).
6.2.1.1
Physical model
The self-baking electrode of 16500 kVgA nickel electric smelting furnace in
Jinchuan Nonferrous Metal Company was taken as an example in the analysis.
Its schematic is shown in Fig. 6.1. In furnace No. 2, the electrode diameter is 1m,
the electrode current is about 11̚ 18 kA, and the electrode consumption is 0.3
m/d.
The compositions and properties of several kinds of domestic electrode pastes
are shown in Table 6.1.
Table 6.1 Component and properties of the electrode paste
Producing
area
Fixed
carbon
/%
Volatile
/%
Ash
/%
M
oisture
/
%
Density
/ggcm
−3
Compression
strength/Pa
Porosity
/%
Jilin 69.26 19.23 11.07 0.45 1.50 260×10
5
20.0
Guiyang 70.25 18.38 10.52 0.85 1.61 135.5×10
5
25.0
Shanghai 79.23 11.99 8.78 1.57 270×10
5
23.9
Kunming 72.64 19.34 8.02 1.71 37.75×10
5
15.9
The tested electrical resistance of the electrode paste at different temperature are
given in Table 6.2.
Table 6.2 Electrical resistance of electrode paste (¡gmm
2
/m)
Category 100ć 150ć 200ć 250ć 300ć
Standard paste 1.07×10
8
3.4×10
7
1.4×10
7
9.8×10
6
7.93×10
6
Inclosed paste 7.45×10
5
4.54×10
5
3.4×10
5
4.69×10
5
1.67×10
6
Category 350ć 400ć 450ć 500ć 650ć
Standard paste 5.22×10
6
1.72×10
6
9.30×10
4
1.37×10
4
2.4×10
3
Inclosed paste 5.44×10
5
9.45×10
4
1.86×10
4
4.72×10
3
2.49×10
3
The real electrode system is very complicated. To obtain a physical model
that can be described in mathematical way, some simplifications are made as
follows:
Jiemin Zhou and Ping Zhou
a) The electrode system is in a steady state, and
its electric field and thermal field do not change
with time.
b) The electrode descends at a certain speed.
c) The influence of skin effect on the
electrode electric field is ignored. According to
reference (Bochmann et al., 1968), the
influence of skin effect on the current
distribution is inapparent when the electrode
diameter is less than 1.2 m.
d) Assumption is made that when the
electrode shell (casing) in the slag is burned out,
it will be replaced by the electrode paste, which
means the electrode diameter is unchanged, The
burning loss occurs only at the tip of the
electrode, and the speed corresponds to the
electrode descent speed. This assumption
simplifies the calculation without inducing big
error (Olsen et al., 1972).
Cylindrical coordinates are adopted in this
model, and its computational domain is shown in
Fig. 6.2. Due to the symmetry of the geometry,
the electric field and the thermal field, only 1/2n
of the cycle area of the electrode cross section
needs to be calculated, where n is the number of
the brass clamps. In z direction, the analytical
domain is divided into five zones along the paste
surface, the upper and downside edges of the
clamp, the inner surface of the furnace roof, the
slag surface and the tip surface of the electrode.
While in the radial direction, the analytical
domain is divided into two zones along the fin.
Thus, suitable discretization can be applied to
different zone according to the importance of
each zone and the requirement for calculation
precision, if necessary.
6.2.1.2
Mathematical model
In the cylindrical coordinates system, when the electrode descends at a constant
speed, the differential equation of the thermal conduction that dominates the heat
transfer process in the electrode is given as follows:
Fig. 6.2 Anal
y
tical domain and
grids arrangement of
self-baking electrode