Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
IMPROVING THE ACCURACY OF CEMS BY MEANS OF MULTIPATH ULTRASONIC FLOWMETER
Toralf DietzR&D Divison Flow Solutions20. April 2015
2Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
MEASUREMENT PRINCIPALULTRASONIC FLOW METER
cospAB vc
Lt
cospBA vc
Lt
BAABp tt
Lv 1-1cos2
vp - average velocity on measuring pathtAB.tBA- transit timesL - length of measuring pathD - diameter - installation angle to flow axisc - speed of soundk - calibration factor
pV vDkQ 2
4
3Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
MEASUREMENT PRINCIPALULTRASONIC FLOW METER
picture: nuclearpowertraining.tpub.com
·
2 ·
0.9
0.92
0.94
0.96
0.98
2 4 6 8 10
k-fa
ctor
lg Re
4Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
FLOW PROFILEUNCERTANTY CONSIDERATIONS
5Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
Swirl adds a velocity component!
FLOW PROFILEUNCERTAINTY CONSIDERATIONS
flow
Vnonaxial
Vaxial
Vsound
meter body bore
1.A
1.B
))tan(()cos( nonaxial
vvvLt
axialsoundAB
))tan(()cos( nonaxial
vvvLt
axialsoundBA
cospAB vc
Lt
cospBA vc
Lt
6Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
FLOW PROFILEUNCERTANTY CONSIDERATIONS
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
2 4 6 8 10
k-fa
ctor
lg Re
INVESTIGATION SINGLE PATH SYSTEMUNCERTAINTY CONSIDERATIONS
-8
-6
-4
-2
0
2
4
6
8
0 200 400 600 800 1000
mea
surin
g de
viat
ion
f →
flow rate Q →
10D straight input length+ single bend pipe 0°10D straight input length+ single bend pipe 45°10D straight input length+ single bend pipe 90°10D straight input length+ single bend pipe 135°10D straight input length+ single bend pipe 180°10D straight input length+ single bend pipe 225°10D straight input length+ single bend pipe 270°10D straight input length+ single bend pipe 315°
± 3%
20./21. April 2015 Toralf Dietz | NIST Workshop Smoke Stack Measurement 7
single bend10D upstream
INVESTIGATION SINGLE PATH SYSTEMUNCERTAINTY CONSIDERATIONS
-8
-6
-4
-2
0
2
4
6
8
0 200 400 600 800 1000
mea
surin
g de
viat
ion
f →
flow rate Q →
10D straight input length + single bend pipe 0°10D straight input length + single bend pipe 45°10D straight input length + single bend pipe 90°10D straight input length + single bend pipe 135°10D straight input length + single bend pipe 180°10D straight input length + single bend pipe 225°10D straight input length + single bend pipe 270°10D straight input length + single bend pipe 315°20D straight input length + single bend pipe 0°20D straight input length + single bend pipe 45°20D straight input length + single bend pipe 90°20D straight input length + single bend pipe 135°20D straight input length + single bend pipe 90° (2)20D straight input length + single bend pipe 180°20D straight input length + single bend pipe 225°20D straight input length + single bend pipe 270°20D straight input length + single bend pipe 315°
single bend20D upstream
± 1%
20./21. April 2015 Toralf Dietz | NIST Workshop Smoke Stack Measurement 8
9Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
Size: 2“ to 48“ Pressure: 100bar Dry, clean gas Machined meter body Accurately measured geometry Meters are calibrated at a flow
lab
FISCAL METERING GAS TRANSMISSIONULTRASONIC FLOW METER
10Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
FISCAL METERINGULTRASONIC FLOW METER
ACCURACY CLASSES
DOCUMENTS ISO 17089-1&2.A.G.A. Report 9OIML R137
Installation EffectsClass 1.0: less than 0.33%Class 0.5: less tan 0.16%
EXAMPLEPROJECT DIRECT CO2 MONITORING
11
12Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
SYSTEM OVERVIEWPROJECT DIRECT CO2 MONITORING
QAadjustment
QAadjustment
3D Laser –Adjustment
mastertools
A
CO2 Concentration Flow
MAC
CO2 mass
Stack
13Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
INSTALLATIONPROJECT DIRECT CO2 MONITORING
14Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
Installation FLOWSIC100:▸ Target uncertainty (as found): Uk=2 ≤ ±1.0%▸ 2-path system
― 60° path angle― Chordal layout, mid radius position
▸ Upstream of the flue gas scrubber▸ T = 165°C (330°F)▸ Inner diameter 6200mm (20.34ft)▸ approx. 5D downstream of a 90°-bend with guiding plates
Validation by▸ an extended measurement traverse at real conditions ▸ Comparison with thermo-dynamic model calculation
FLOW MEASUREMENTPROJECT DIRECT CO2 MONITORING
15Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
ADJUSTMENT TIME MEASUREMENTPROJECT DIRECT CO2 MONITORING
Zero flow check and SOS-checkEach device passed a zero-flow and a speed of sound check to reduce the manufacturing uncertainty
U (k=2)Time difference / µs 0.5Time absolute / µs 2.8
16Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
3D laser scanner on site for precise measurement of▸ Diameter. ▸ Path length and ▸ Path angle
0.1° error @ 60° path angle 0.3% velocity error
GEOMETRY PARAMETERPROJECT DIRECT CO2 MONITORING
Parameter (N2) value uRadius / mm 3102 7Path length 1 / mm 6135 5
Path length 2 / mm 6177 5
Path angle 1 / ° 57.66 0.1Path angle 2 / ° 57.63 0.1
17Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
UNCERTAINTY CONTRIBUTIONSPROJECT DIRECT CO2 MONITORING
0.00%
0.02%
0.04%
0.06%
0.08%
0.10%
0.12%
0.14%
0.16%
0.18%
2 5 10 15 20 25
rel.
stan
dard
unc
erta
inty
/ %
gas velocity / m/sradius path angle path length time abs time diff
18Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
Virtual adjustment: theoretically defined calibration function based on CFD calculations
„Virtual calibration“ of the flow meter · „Worst Case analysis“ – rotation of the flow profile ±0.6%
COMPUTATIONAL FLUID DYNAMICSPROJECT DIRECT CO2 MONITORING
19Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
UNCERTAINTY CONTRIBUTIONPROJECT DIRECT CO2 MONITORING
0.00%
0.05%
0.10%
0.15%
0.20%
0.25%
0.30%
0.35%
0.40%
2 5 10 15 20 25
rel.
Stan
dard
unc
erta
inty
/ %
gas velocity / m/s
flow profile process values disturbance probeintruding probe path position integration
20Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
-1.5%
-1.0%
-0.5%
0.0%
0.5%
1.0%
1.5%
0 5 10 15 20 25 30
Unc
erta
inty
(k=2
)
Gas Velocity / m/s
EXPECTED UNCERTAINTY FLOW MEASUREMENTPROJECT DIRECT CO2 MONITORING
21Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
: (1) continuous thermo-dynamic calculation- Process measurement- Model of the thermodynamic cycle
: (2) Mass balance analysis- Mass of burned coal- Chemical analysis the coal
: (3) extended traverse test measurement- Acceptance inspection
PROJECT DIRECT CO2 MONITORING
MessgerätVolumenstrom
MessgerätCO2-Konzentration
RG-Kanal
Emissions-Rechner
CO2-Frachtgemessen
Online-Kontrollrechn.
Online-Kreisprozess-berechnung
CO2-Frachtberechnet
Offline-Kontrollrechn.
Blockwirkungs-gradmessungmittelskalibrierterPräzisions-messtechnik(Basis Abnahme-regelwerk)
Werks-/ DKD-Kalibrierungen
MessgerätVolumenstrom
MessgerätCO2-Konzentration
RG-Kanal
Emissions-Rechner
CO2-Frachtgemessen
Online-Kontrollrechn.
Online-Kreisprozess-berechnung
Online-Kreisprozess-berechnung
Online-Kreisprozess-berechnung
CO2-Frachtberechnet
Offline-Kontrollrechn.
Blockwirkungs-gradmessungmittelskalibrierterPräzisions-messtechnik(Basis Abnahme-regelwerk)
Werks-/ DKD-Kalibrierungen
22Toralf Dietz | NIST Workshop Smoke Stack Measurement
Average deviation = 0.1% sigma: = ±1.4%
UNIT N1THERMODYNAMIC MODEL CALCULATION
20./21. April 2015
23Toralf Dietz | NIST Workshop Smoke Stack Measurement
Average deviation: = -1.85% Sigma: = ±1.1%
UNIT N2THERMODYNAMIC MODEL CALCULATION
20./21. April 2015
24Toralf Dietz | NIST Workshop Smoke Stack Measurement
Average deviation: = 0.81% Sigma: = ±1.0%
UNIT P1THERMODYNAMIC MODEL CALCULATION
20./21. April 2015
25Toralf Dietz | NIST Workshop Smoke Stack Measurement
Average deviation: = 0.49% Sigma: = ±1.0%
UNIT P2THERMODYNAMIC MODEL CALCULATION
20./21. April 2015
26Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
method Period Unit N1 Unit N2 Unit P1 Unit P2
Total mass balance(U: 1.5%)
2 month 0.80%
Thermo dynamic model calculation(U: 1.5%)
1st month -0.10% 1.85% 0.81% 0.49%
2nd month 0.50% 2.50% 1.20% 0.50%
Extended Traverse(U: 1.3 .. 2.2%)
Single test -0.14% 1.86% 0.75% 0.47%
TOTAL CO2 MASS DIFFERENCESRESULTS
27Toralf Dietz | NIST Workshop Smoke Stack Measurement20./21. April 2015
CEMS:▸ Measurement uncertainty of better than 1.5% is realistic for direct CO2 monitoring▸ Verification uncertainty is at the same level!
Recommendations for the Ultrasonic flow meter▸ Install with max. possible straight upstream length▸ Reduce the uncertainty by multi path layout (≥ 2path)▸ Use CFD analysis
― to find an optimized path layout (if you have the freedom)― And/or calculate a “dry” calibration function
▸ Do precise geometry measurement, especially― Path angle― Diameter
SUMMARY
MANY THANKS FOR YOUR ATTENTION.