electrostatic precipitator performance improvement through ... · electrostatic precipitator...

PowerGen 2002 December 11, 2002

Airflow SciencesCorporation1

Electrostatic Precipitator PerformanceImprovement Through Computational

Fluid Dynamic Modeling

PowerGen InternationalDecember 11, 2002

Authors:Brian J. Dumont

Robert G. Mudry, P.E.

Airflow Sciences Corporation



Introduction

½ Flow modeling is an established practice to optimize gas flow patterns within industrial equipment

½ Little published data exists on the accuracy of flow models, particularly for electrostatic precipitators (ESPs)

½ Available data was analyzed in detail to compare model results to actual plant test data



Fluid Flow Modeling

½ Computational Fluid Dynamics (CFD)½ Physical scale modeling



Testing Methods

½ ESP cold-flow velocity distribution measurement

Flow Flow



Objectives of Analysis

½ Assess all available data for ESP testing and modeling acquired over the past 7 years

• ESP field test data• CFD model results• Physical model results

½ Perform statistical comparisons of the data

½ Obtain quantitative information relating model correlation to test data



Cases Analyzed

½ Ten cases where field data and CFD data exist for the same configuration

½ Five cases where corresponding physical model data also exist

½ All cases from coal-fired electric power stations• U.S. and Canadian plants• Unit size ranges from 326 MW to 952 MW

½ One case study is presented in detail, 326MW Unit in the Western U.S.

½ All cases are summarized



Data Comparisons

½ Contour plots

½ Flow distribution statistics• % RMS• ICAC Standards

½ Point-by-point deviations

½ Overall Correlation Factor(%RMS of point-by-point deviations)



Case Study 2 - Inlet PlaneNormalized Streamwise Velocity Component

½ Both models capture trends½ CFD model captures peak

velocity better, but overpredicts the size of the high velocity region

Phys

ical

Tes

t Dat

a

CFD



Case Study 2 - Inlet PlaneFlow Statistics - Deviations from Goal Velocities

½ Physical model overpredicts flow compliance to goal as shown by all 3 analyses

½ CFD model underpredicts flow compliance to goal as shown by all 3 analyses

%RMS Modified ICAC 115% Modified ICAC 140%0

10

20

30

40

50

60

70

80

90

100

20.1

86.5

99.7

33.4

63.5

93.7

42

61.5

91Physical Model

Test Data

CFD ModelTest Data

CFD Model

Physical Model

42.0

33.4

20.1

63.561.5

86.5

99.793.791.0

% RMS ICACm 115% ICACm 140%



Case Study 2 - Inlet PlaneHistograms – Point-By-Point Deviations from Test Data

½ Correlation Factors:CFD: 37.0Physical: 32.4

½ 65% of CFD points within +/-25% band

½ 61% of physical model points within same band Percent Deviation from Test Data

30

25

20

15

10

5

0

Per

cen

tag

e o

f D

ata

Po

ints

-100 -80 -60 -40 -20 0 20 40 60 80 100

CFD ModelPhysical Model



Case Study 2 - Outlet PlaneNormalized Streamwise Velocity Component

½ Both models capture trends½ CFD model captures peak

velocity betterPh

ysic

al

Tes

t Dat

a

CFD



Case Study 2 - Outlet PlaneFlow Statistics - Deviations from Goal Velocities

½ Both models underpredict %RMS, especially the physical model

½ Both models predict modified ICAC conditions fairly well

%RMS Modified ICAC 115% Modified ICAC 140%0

10

20

30

40

50

60

70

80

90

100

17.8

72.6

92

34.7

78.8

86.8

27.8

76.7

90.6Physical Model

Test Data

CFD Model

Test Data

CFD Model

Physical Model

27.834.7

17.8

78.8 76.772.6

92.0 90.6

% RMS ICACm 115% ICACm 140%

86.8



Case Study 2 - Outlet PlaneHistograms - Point-By-Point Deviations from Test Data

½ Correlation Factors:CFD: 27.2Physical: 40.8

½ 75% of CFD points within +/-25% band

½ 48% of physical model points within same band Percent Deviation from Test Data

30

25

20

15

10

5

0

Per

cen

tag

e o

f D

ata

Po

ints

-100 -80 -60 -40 -20 0 20 40 60 80 100

CFD ModelPhysical Model



Case Study 2 - Summary

½ ESP Inlet� Both models correlate fairly well� Physical model has a lower correlation coefficient� CFD model matches flow statistics better and has a

larger number of points within +/-25% deviation band on histogram

½ ESP Outlet� CFD model agrees better with test data under all

comparisons� Both models capture correct trends



All Case Studies - Correlation Factor½ In some cases, the CFD model has a better correlation factor½ In others, the physical model is better½ There is no clear trend as to why this occurs½ The CFD model correlates better on average

Correlation Factor Summary

0

5

10

1520

2530

35

40

45

1I 1O 2I 2O 3I 3AB 3O 4AB 5O 6I 6O 7I 7O 8I 8O 9I 9O 10AB

Case

Co

rrel

atio

n F

acto

r

CFD Model Physical Model

40

30

20

10

0Co

rrel

atio

n F

acto

r

Correlation Factor Summary



All Case Studies - Overall Summary

½ On average, the Correlation Factor for CFD models is 23.5 (27.8 using only studies where the physical model also existed)

½ On average, the Correlation Factor for physical models is 32.6



Analysis Conclusions

½ On the whole, correlation is not as strong as desired

½ Experience indicates that this level of correlation is enough to make CFD modeling a useful engineering tool for ESPs

½ Additional research and development in CFD technology will allow for increased accuracy



Case Study

½ Two identical 790 MWunits

½ Western U.S. ½ Cold side chevron ESPs½ SCA=125½ Avg. vel 7.2 ft/s½ Both units regularly derated by 240 MW to

operate within opacity limits



Baseline CFD Model Results

½ Velocity pattern at ESP inlet shows non-uniform• High velocity on top and bottom, low velocity in center• RMS deviation = 26 %

½ Outlet plane poor distributionas well• RMS deviation = 20 %



Design Optimization

½ Design modifications developed using CFD model• New inlet and outlet perforated plates• Slight alteration to inlet duct turning vanes



Final Design CFD Model Results

½ ESP inlet velocity profile now uniform• RMS deviation = 7 %

½ Outlet plane also improved• RMS deviation = 7 %

½ Change to system pressure drop = +0.3 ”H2O



Resulting ESP Performance

½ ESP efficiency testing performed• 23% reduction in particulate emissions compared to original

ESP geometry

½ At same opacity, plant output increases by 150 MW per unit

½ Payback for model study and installation costs = 1 year



Questions

electrostatic precipitator performance improvement through ... · electrostatic precipitator...

Documents