close

Вход

Забыли?

вход по аккаунту

?

Close-to-streamline numerical study on gas velocity distribution in industrial scale electrostatic precipitator gas inlet hood.

код для вставкиСкачать
ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
Published online 29 May 2010 in Wiley InterScience
(www.interscience.wiley.com) DOI:10.1002/apj.460
Special Theme Research Article
Close-to-streamline numerical study on gas velocity
distribution in industrial scale electrostatic precipitator gas
inlet hood
Jie Zhang*
Curtin Centre for Advanced Energy Science and Engineering, Department of Chemical Engineering, Curtin University of Technology, GPO Box U1987,
Perth, WA 6102, Australia
Received 15 October 2009; Revised 2 April 2010; Accepted 20 April 2010
ABSTRACT: The uniformity of gas velocity distribution in gas inlet hoods is important to guarantee high dedust
efficiency in electrostatic precipitators (ESPs), especially for high solid concentration applications. The close-tostreamline numerical method was developed to geometrically 1 : 1 simulate an industrial scale ESP gas inlet hood,
which had complicated internal structures, such as dedust angle irons and gas distribution perforated plates with up to
10 000 holes. The realizable k-ε model was used for the gas flow simulation. The numerical results of the gas velocity
distribution show reasonable agreement with the field measurements. The outlet gas velocity distribution was nonuniform, which was lower in the central region and higher in the near-wall regions. The gas flow characteristics were
analysed to reveal the main influence factors on gas velocity distribution and put forward corresponding modification
methods. After the retrofit, the field measurements of the outlet gas velocities showed that the gas velocity distribution
improved. The non-uniform index of outlet gas velocity distribution decreased greatly from 0.355 to 0.244. The visible
dust emissions from the chimney disappeared which indicated increased dedust efficiency. Therefore, the close-tostreamline method can simulate gas velocity distribution in complicated structures of ESP gas inlet hoods.  2010
Curtin University of Technology and John Wiley & Sons, Ltd.
KEYWORDS: numerical simulation; electrostatic precipitator; gas inlet hood; gas velocity distribution
INTRODUCTION
Gas inlet hoods with large included angles are
commonly used in the inlet section of electrostatic precipitators (ESPs). The main purpose is to reduce the
gas velocity entering ESPs and prolong the gas residence time for better dedust efficiency in ESPs.[1] Due
to space limitations, the area ratio between the outlet
and inlet sections for gas inlet hoods is usually quite
large. The large area ratio makes it very difficult to
realize uniform gas velocity distribution at the outlet
section of gas inlet hoods and also the inlet section
entering ESP electric fields. For this reason, perforated
plates with a certain opening ratio, i.e. perforated plate
porosity, are commonly used to improve gas velocity
distribution in gas inlet hoods.[2] However, uniform gas
flow distribution in gas inlet hoods is difficult to be realized due to large included angles caused by the large
*Correspondence to: Jie Zhang, Curtin Centre for Advanced Energy
Science and Engineering, Department of Chemical Engineering,
Curtin University of Technology, Perth, WA 6102, Australia.
E-mail: jie.zhang@curtin.edu.au
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
cross-sectional area ratio. Any localized high velocity
region would result in dust particles at certain solid concentration escaping through the collector plates without
effective collection, which would contribute to higher
dust emissions from the downstream chimney. ESPs
are commonly used in coal-fired power plants, which
have an inlet solid concentration of 20–50 g/m3 from
upstream flue gas ducts.[3] In addition, ESPs have been
increasingly used in industrial applications with higher
solid concentration entering ESPs. The solid concentration can increase 20–50 times, up to 1000 g/m3 in
some industrial applications, such as cement production and dry flue gas desulfurization processes.[4,5] With
increased solid concentration entering ESP gas inlet
hoods, the uniformity extent of the gas velocity distribution became more important to guarantee high dedust
efficiency and meet stricter dust emission regulations.
For industrial scale ESP gas inlet hoods, the external
structure is quite large with the outlet cross-sectional
area of up to 200 m2 and the included angle of up to
120◦ . The internal structures are also very complicated,
such as the perforated plates with up to 10 000 holes
and various local structures of baffles and slots. Current
638
J. ZHANG
researches mostly made some geometrical assumptions
to simplify the complicated structures of industrial
scale ESP gas inlet hoods, which made it difficult to
accurately reveal the detailed gas flow characteristics in
gas inlet hoods.
Experimental researches, such as reduced-scale physical models and on-site field measurements, have been
widely used to investigate gas velocity distribution
in ESP gas inlet hoods. Reduced-scale physical models were built based on the similarity principle and
were studied by visualization techniques and velocity measurements.[6] On-site measurements were always
conducted to verify the gas velocity distribution performance after the industrial ESP installation. However,
detailed and reliable gas velocity measurements inside
gas inlet hoods were very time-consuming and expensive due to the large and complex structural characteristics.
As an alternative method, numerical simulations have
been developed to study the gas flow characteristics in
ESP gas inlet hoods.[7,8] One important part of numerical simulations was the treatment of the gas inlet
hood geometry, especially the perforated plate structures. Most numerical studies adopted the equivalent
porous media assumption by using certain local resistance coefficients to replace the perforated plate structure, which could greatly reduce the computational grid
number.[9,10] The simulation accuracy of the predicted
gas flow characteristics largely depended on the local
resistance coefficient that was calculated by the known
pressure drop through the perforated plate.[11] However,
the real pressure drop through the perforated plate was
difficult to be determined in advance, especially for nonuniform gas velocity distribution across the perforated
plate. Therefore, this simulation method was difficult to
be widely used for the design and retrofit of ESP gas
inlet hoods.
The close-to-streamline numerical method has been
successfully developed in predicting the gas–solid flow
in industrial scale tangentially fired boilers with complicated structures of pulverized coal burners and plate
superheaters.[12,13] This numerical method was mainly
the structural grid construction accompanied with the
combined coordinate grid system. The structural grid
systems were constructed with different grid directions
and grid sizes based on the geometric structure and the
flow characteristics of various flow regions in the computational domain. The close-to-streamline method had
the advantages of effectively reducing the numerical
false diffusion and accurately simulating detailed local
structures. Therefore it is possible to simulate the gas
flow characteristics in detail in the complicated structures of industrial scale ESP gas inlet hoods.
In this article, the close-to-streamline numerical
method was adopted to geometrically 1 : 1 simulate
the gas flow characteristics in an industrial scale ESP
gas inlet hood and its complicated internal structures,
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pacific Journal of Chemical Engineering
including the perforated plates. The gas velocity
distributions were compared with field measurements
and analysed in detail to provide modification guidance
to the structure retrofit. The retrofit was successful in
improving the gas velocity distribution uniformity in the
gas inlet hood and results in the disappearance of the
visible dust emissions from the downstream chimney.
NUMERICAL SIMULATION PROCEDURE
Gas inlet hood geometry
The gas inlet hood was quite large that contained large
included angles. The schematic structure of the gas inlet
hood is shown in Fig. 1. The inlet section was 5 m
× 3.8 m and the outlet section was 10.8 m × 15.1 m.
The cross-sectional area ratio between the outlet and
inlet cross sections was 8.55. The included angle was
horizontally 2α = 84◦ and vertically 2β = 120◦ , which
indicated that in the gas inlet hood it was very difficult
to realize uniform gas velocity distribution even with
the addition of various internal structures.
The adopted internal structures mainly included the
dedust angle iron section, the first stage perforated plate
section and the second stage perforated plate section.
The dedust angle iron section consisted of two layers of
staggered angle irons with the internal surfaces towards
the gas flow direction to collect the dust particles in the
flue gas and then reduce the solid concentration entering
the ESP. The collected dust was knocked off the angle
irons and removed from the bottom dust slot in the gas
inlet hood. The first and second stage perforated plate
sections comprised of small pieces of perforated plates
with certain porosity to facilitate the perforated plate
fabrication and installation processes. Each perforated
plate section had up to 10 000 holes to realize uniform
gas flow distribution in the ESP gas inlet hood, which is
beneficial to guarantee the dedust efficiency by realizing
the full capacity of the downstream ESP electric fields.
Mathematical models
The gas flow in the ESP gas inlet hood was assumed
to be viscous, incompressible, steady, isothermal and
turbulent flow.[14] The realizable k-ε model was used
to simulate the turbulent gas flow, which consisted
of a modified model dissipation rate equation and an
improved realizable eddy viscosity formulation.[15]
Numerical approach and computational grids
The close-to-streamline numerical method was mainly
the structural grid construction method with the combined coordinate grid system. The structural grid systems were constructed with different grid directions and
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
CLOSE-TO-STREAMLINE NUMERICAL STUDY
Figure 1. Structure schematics of the ESP gas inlet hood. (a) Dedust
angle iron; (b) First stage perforated plate; (c) Second stage perforated
plate. This figure is available in colour online at www.apjChemEng.com.
grid sizes based on the geometric structure and the flow
characteristics of various flow regions in the computational domain. The computational data were transferred
by interpolation calculations on the interfaces between
the structural grid systems.
Using different grid directions for various flow
regions, the numerical false diffusion could be reduced
due to the decreased slope angle between the flow direction and the grid direction, which resulted in improved
numerical simulation accuracy. Using different grid
sizes for various flow regions, it is possible to reduce
the grid size of certain detailed structures for high precision simulation and then use larger grid sizes for other
flow regions to reduce the total grid number.
In addition, the computational grid set-up time was
quite long for industrial scale equipment with complicated internal structures, especially in equipment structure design and retrofit applications. As for the common grid set-up method, all the computation grids
were regenerated when certain localized structure or
dimension was modified. As for the close-to-streamline
method, only the computational grid for the flow regions
with structural modifications would be regenerated and
the computational grid for other flow regions did not
require regeneration, which greatly shortened the computational grid set-up time.
Therefore, due to these advantages of the close-tostreamline numerical method, it is possible to geometrically 1 : 1 simulate industrial scale ESP gas inlet hoods
and its complicated internal structures such as the dedust
angle irons and the perforated plates with up to 10 000
holes.
In this article, the close-to-streamline method was
used to simulate the ESP gas inlet hood. The computational domain was divided into three flow regions,
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
i.e. the angle iron section, the first and second stage
perforated plate sections. Each flow region was constructed with certain non-uniform staggered structural
grid system with suitable grid direction and grid size
according to the geometric structure and the flow characteristics. The effect of grid sizes and grid distributions
was investigated to obtain the optimum combination
of grid independence and computational economy. The
total grid number was up to 2.5 million with 30 grids
along the gas flow direction. The grid diagram of the
three flow regions is shown in Fig. 2.
The grid number on each of the first and second stage
perforated plates was roughly 80 000 with at least four
grids on each perforated plate hole. Figure 3 shows the
grid diagram of the first stage perforated plate and its
detailed bottom structure.
The discretized equations for the pressure–velocity
coupling were solved by using the SIMPLEC algorithm.
Pressure was solved using a second-order discretization scheme. A power law discretization method was
used for momentum, turbulent kinetic energy and turbulence dissipation rate. The flow field was numerically
calculated, respectively, in the three flow regions with
different grid systems. The calculation data were then
transferred by interpolation calculations on the interfaces between the structural grid systems.
Boundary conditions
The outlet boundary condition was the pressure outlet
with a fully developed flow. The exit pressure was
set to 0 and all computed pressures were relative to
the exit pressure. Non-slip boundary conditions were
assumed at the gas inlet hood walls. The standard wall
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
639
640
J. ZHANG
Asia-Pacific Journal of Chemical Engineering
Figure 2. Grid diagram of various structural sections of ESP gas inlet hood.
Figure 3. Structure schematics of the first stage perforated plate. This
figure is available in colour online at www.apjChemEng.com.
functions were adopted in the near-wall regions. The
inlet boundary condition was a gas velocity profile that
consisted of experimentally measured data at the gas
flow rate of 400 501 Nm3 /h. The gas temperature was
75 ◦ C and the gas density was 0.9825 kg/m3 .
RESULTS AND DISCUSSION
Several cross sections along the gas flow direction were
selected at a distance of about 200 mm before and after
the internal structures, such as the angle iron and the
perforated plates. The 200 mm distance was chosen
to reduce the gas velocity fluctuation caused by the
internal structures and reveal relatively stable gas flow
distribution.
Variation of average gas velocity along the
gas flow direction
Figure 4 shows the variation of average gas velocity on
the selected typical cross sections. It can be seen that
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
the average gas velocity decreases gradually along the
gas flow direction due to the enlarged cross-sectional
area. The average gas velocity ratio between the inlet
and outlet sections was 8.44, which was close to the
aforementioned cross-sectional area ratio of 8.55. The
difference between the two parameters indicated nonuniform gas velocity distributions in the inlet and outlet
sections.
Gas velocity distribution at the outlet
At the outlet of the gas inlet hood, the gas velocity distribution was measured by hot-wire anemometers. The measurement points were ten rows evenly
distributed along the height direction (Z axis) and eight
rows evenly distributed along the width direction (X
axis).
Figure 5 shows the outlet gas velocity comparison between the field measurements and the numerical results. The numerical results showed reasonable
agreement with the field measurements. The outlet gas
velocity distribution was non-uniform with lower gas
velocity in the central region and higher gas velocity
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
CLOSE-TO-STREAMLINE NUMERICAL STUDY
Figure 4. Average gas velocity along the gas flow
direction on typical cross sections.
in the near-wall regions. The non-uniform index, i.e.
the rms velocity deviation from the average outlet gas
velocity, was used to evaluate the non-uniform extent of
the outlet gas velocity distribution. The field measurements showed that the non-uniform index of the outlet
gas velocity distribution was up to 0.355, which was
far beyond the acceptable limit of 0.25.[8] This resulted
in visible dust emissions from the downstream chimney
during the first stage power plant operation.
Gas velocity distribution in the gas inlet hood
The non-uniform gas velocity distribution in the gas
inlet hood was analysed to reveal the main influence
factors and put forward corresponding modification
methods for the gas inlet hood retrofit.
Figure 5.
Outlet gas velocity comparison between
numerical and experimental results.
Gas velocity distribution near the inlet section
The gas velocity distribution of the cross section Y =
300 mm is shown in Fig. 6, which is similar to the inlet
gas velocity distribution. The gas velocity distribution
was higher in the left region (minus the X direction). In
addition, it took on counterclockwise rotation tendency
with three-dimensional velocity components. In ESP
gas inlet hood design processes, the inlet gas velocity
distribution was commonly assumed to be uniform
with the gas velocity normal to the gas inlet hood.
Figure 6. Gas velocity distribution of the cross section Y = 300 mm. This figure is available in colour online at
www.apjChemEng.com.
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
641
642
J. ZHANG
Asia-Pacific Journal of Chemical Engineering
Gas velocity distributions near the angle iron section. This figure is available in colour online at
www.apjChemEng.com.
Figure 7.
However, due to the upstream non-uniform gas flow
characteristics, the real inlet gas velocity distribution
was usually non-uniform and not normal to the inlet,
which had great influence on the gas inlet hood design.
In order to solve this problem, it is helpful to add
guide vanes, such as assembled honeycomb structures,
to reduce the gas rotation tendency and improve the gas
velocity distribution entering the gas inlet hood. However, the horizontal plates of the honeycomb structures
should be kept short to avoid dust deposition, especially
in high solid concentration applications.
Gas velocity distribution near the angle iron
section
In order to analyse the influence of the angle iron
section that was located at Y = 1000 mm, the gas
velocity distributions of the cross sections at Y = 800
and Y = 1200 mm are shown in Fig. 7.
The gas velocity distributions of the cross sections
at Y = 800 and Y = 1200 mm were similar, which
indicated that the angle iron section had little effect
on the gas flow distribution. The gas velocity in the
left region was higher than the right region and the gas
velocities in the top and bottom regions were lower.
Compared with the cross section at Y = 300 mm, the
non-uniform extent of the gas velocity distributions near
the angle iron section gradually decreased due to the
diffusion effect of enlarged cross-sectional area along
the gas flow direction. In addition, the gas velocity
distributions took on stripe-like profiles along the height
direction due to the blockage effect of the angle iron
structures.
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
Gas velocity distribution near the perforated
plate sections
The gas velocity distribution near the first stage perforated plate was presented in detail due to the similar
influence of the two-stage perforated plates on the gas
velocity distribution. The gas velocity distribution of
the cross section at Y = 1800 mm is shown in Fig. 8,
which is located before the first stage perforated plate.
Due to the blockage effect of the perforated plate, the
gas flow had an obvious tendency to move from the
central region to the near-wall regions. This resulted in
higher gas velocity in the near-wall regions. In addition,
both the gas velocity magnitude and the rotation tendency decreased gradually along the gas flow direction
due to the diffusion effect of enlarged cross-sectional
area.
The gas velocity distribution of the cross section at
Y = 2200 mm is shown in Fig. 9. The gas velocity
adjacent to the left, right and top walls was very low
due to the baffle plate structures in these areas. The gas
velocity near the bottom wall was quite high due to
the bottom dust slot structure. Compared with the cross
section at Y = 1800 mm, the gas velocity distribution at
Y = 2200 mm was even more non-uniform with higher
gas velocity in the near-wall regions and lower gas
velocity in the central region.
The influence of the first stage perforated plate on
the gas velocity distribution was qualitatively analysed
as follows. As the gas flow approached the perforated plate, the flow streamlines tended to diverge and
then finally became nearly normal to the perforated
plate. Due to the blockage effect of the first stage
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
CLOSE-TO-STREAMLINE NUMERICAL STUDY
Figure 8. Gas velocity distribution of the cross section Y = 1800 mm. This figure is available in colour online at
www.apjChemEng.com.
Figure 9. Gas velocity distribution of the cross section Y = 2200 mm. This figure is available in colour
online at www.apjChemEng.com.
perforated plate, the gas velocity in the central region
decreased rapidly which resulted in the kinetic energy
loss transformed into higher static pressure. Due to
the large included angles of the gas inlet hood, the
static pressure in the near-wall regions was originally
low. The pressure difference between the central region
and the near-wall regions also induced the gas flow to
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
change the flow direction towards the near-wall regions.
However, when the gas flow reached the near-wall
regions, it was difficult for the gas flow to overcome the
movement inertia and change the flow direction again
to move towards the central region due to the relatively
large turnover angle in the near-wall regions. Therefore,
the gas velocity distribution after the perforated plate
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
643
644
J. ZHANG
Asia-Pacific Journal of Chemical Engineering
Figure 10. Gas velocity distributions near the second stage perforated plate. This figure is available in
colour online at www.apjChemEng.com.
was higher in the near-wall regions and lower in the
central region.
Figure 10 shows the gas velocity distributions before
and after the second stage perforated plate that was
located at Y = 3000 mm. The gas velocity distributions
near the second stage perforated plate were even more
non-uniform than those near the first stage perforated
plate. This further verified the aforementioned analyses
about the influence of the perforated plates on the gas
velocity distribution.
Two feasible retrofit methods could achieve better
gas velocity distribution for the simulated perforated
plates. One method was to increase the porosity of
the perforated plates, especially the porosity in the
central region. It was practical to selectively blank the
perforated plates in the near-wall regions or remove the
perforated plates in the central region. This would result
in increasing gas flow in the central region and reducing
gas flow towards the near-wall regions. Another method
was to install guide vanes on the perforated plates in
the near-wall regions, which would force the gas flow
to overcome the movement inertia and change the gas
flow direction towards the central region.
Effect of structure retrofit on gas velocity
distribution
Due to the space limitations of the existing engineering application, the external structures of the gas inlet
hood were impossible to be retrofitted. According to the
aforementioned feasible retrofit methods, the internal
structures were modified in the retrofit. This mainly
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
included the addition of the honeycomb structures and
the guide vanes, and the perforated plate porosity adjustment, which were well designed and simulated for optimum retrofit performance. The trial and error method
was mainly used during the structure optimization process. The optimization was finally realized by several
rounds of numerical simulations and on-site field measurements for various structure designs, such as the
positions and the dimensions of the internal structures.
The low cost and highly efficient close-to-streamline
numerical simulation method facilitated carrying out the
structure retrofit and on-site measurements.
After the retrofit, the field measurements of the outlet
gas velocity distribution were carried out to verify the
retrofit performance, which are listed in Table 1. The
measurement points were eight rows evenly distributed
along the height direction (Z axis) and eight rows
evenly distributed along the width direction (X axis).
Table 1. Field measurements of outlet gas velocities
of the retrofitted gas inlet hood (m/s).
Z1
Z2
Z3
Z4
Z5
Z6
Z7
Z8
X1
X2
X3
X4
X5
X6
X7
X8
0.90
0.80
0.80
0.95
0.70
0.90
1.20
1.30
1.00
0.85
0.70
0.75
0.65
0.70
1.00
1.20
0.90
0.80
0.60
0.60
0.75
0.70
0.65
1.20
0.80
0.55
0.55
0.85
0.80
0.85
0.90
1.40
0.85
0.60
0.65
1.00
0.65
0.75
0.85
0.80
0.70
0.70
0.90
0.75
0.45
0.60
0.90
1.20
1.00
0.95
1.00
0.95
0.60
0.60
0.85
1.30
1.10
1.00
0.85
0.70
0.55
0.70
0.95
1.10
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
The outlet gas velocity distribution of the retrofitted
gas inlet hood was much better than that of the original structure. The non-uniform index of the outlet gas
velocity distribution decreased greatly from 0.355 to
0.244, which was helpful in realizing higher ESP performance. The later operating results demonstrated that
the dedust efficiency increased with the disappearance
of visible dust emissions from the chimney.
CONCLUSIONS
The close-to-streamline numerical method was used to
geometrically 1 : 1 simulate an industrial scale ESP gas
inlet hood, which had complicated internal structures,
such as dedust angle irons and perforated plates. The gas
velocity distributions in the gas inlet hood were analysed in detail and the corresponding retrofit methods
were put forward to be adopted in the structure retrofit.
The outlet gas velocity distribution showed reasonable
agreement with on-site gas velocity measurements and
showed the following conclusions:
(1) The close-to-streamline method simulated the gas
velocity distribution in the industrial scale ESP gas
inlet hood in a detailed manner, which provided
important guidance for the structure retrofit of the
gas inlet hood.
(2) The original gas inlet hood was poorly designed
with large included angles due to space limitations.
The gas velocity decreased gradually through the
gas inlet hood with non-uniform outlet gas velocity
distribution that was lower in the central region
and higher in the near-wall regions. This resulted
in visible dust emissions from the downstream
chimney.
(3) The inlet gas velocity distribution was higher in
the left region and took on counterclockwise rotation tendency with three-dimensional velocity components. The angle iron section had little effect
on the gas velocity distribution. The perforated
plate sections had great influence on the gas velocity distribution due to the blockage effect of the
 2010 Curtin University of Technology and John Wiley & Sons, Ltd.
CLOSE-TO-STREAMLINE NUMERICAL STUDY
perforated plates to move the gas flow from the
central region to the near-wall regions.
(4) The internal structures were modified in the retrofit,
which mainly included the addition of honeycomb
structures and guide vanes, and the perforated
plate porosity adjustment. After the retrofit, the gas
velocity field measurements showed that the nonuniform index of the outlet gas velocity distribution
decreased greatly from 0.355 to 0.244. In addition,
this was verified by increased dedust efficiency with
the disappearance of visible dust emissions from the
chimney.
REFERENCES
[1] A. Jaworek, A. Krupa, T. Czech. J. Electrostat., 2007; 65(3),
133–155.
[2] B. Sahin, A.J. Ward-Smith. Heat Fluid Flow, 1987; 8(2),
124–131.
[3] A. Suriyawong, C.J. Hogan, J.K. Jiang. Fuel, 2008; 87(6),
673–682.
[4] S.H. Huang, C.C. Chen. Aerosol Sci. Technol., 2003; 37(2),
109–121.
[5] F.J. Gutierrez, P. Ollero, A. Cabanillas, J. Otero. Adv. Environ. Res., 2002; 7(1), 73–85.
[6] J. Mizeraczyk, M. Kocik, J. Dekowski, et al . J. Electrostat.,
2001; 51–52, 272–277.
[7] S.M.E. Haque, M.G. Rasul, A.V. Deev, M.M.K. Khan,
N. Subaschandar. Appl. Therm. Eng., 2009; 29(10),
2037–2042.
[8] J. Zhang, Z. Zhang, Y.F. Hu. J. Chem. Ind. Eng., 2004; 55(9),
1448–1454.
[9] M.J. Schwab, R.W. Johnson. Proc. Am. Power Conf., 1994;
56(1), 882–888.
[10] I. Gallimberti. J. Electrostat., 1998; 43(2), 219–247.
[11] S.M.E. Haque, M.G. Rasul, M.M.K. Khan, A.V. Deev,
N. Subaschandar. Exp. Therm. Fluid Sci., 2009; 33(2),
322–328.
[12] Z. Zhang, X.C. Xu. J. Combust. Sci. Technol., 2002; 8(2),
97–102.
[13] Z. Zhang, X.C. Xu, S.H. Wu, Y.K. Qin. Proc. CSEE, 2001;
21(8), 108–113.
[14] J. Zhang, C.F. You, C.H. Chen, H.Y. Qi, X.C. Xu. Environ.
Sci. Technol., 2007; 41(12), 4415–4421.
[15] T.H. Shih, W.W. Liou, A. Shabbir, Z.G. Yang, J. Zhu.
Comput. Fluids, 1995; 24(3), 227–238.
Asia-Pac. J. Chem. Eng. 2010; 5: 637–645
DOI: 10.1002/apj
645
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 845 Кб
Теги
streamline, distributions, hood, velocity, industries, numerical, gas, scala, electrostatic, closer, stud, precipitated, inlet
1/--страниц
Пожаловаться на содержимое документа