Post on 09-Mar-2021
transcript
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
ASME TURBO EXPO 2001 - Land, Sea, & Air, The 46th ASME International Gas Turbine & Aeroengine Technical Congress New Orleans, Louisiana, USA, June 4-7, 2001
2001-GT-010 1
UNCERTAINTY REDUCTION IN GAS TURBINE PERFORMANCE DIAGNOSTICS
BY ACCOUNTING FOR HUMIDITY EFFECTS
K. Mathioudakis, A. Tsalavoutas,
Laboratory of Thermal Turbomachines National Technical University of Athens
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 2 of 22
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for
Humidity Effects
• Modelling humidity effects on gas turbine performance
• Impact on Monitoring
o Effects on Measurements
o Effects on "Health Indices"
• Demonstration on Test Data
• Summary-Conclusions
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 3 of 22
Modeling Humidity effects on gas turbine performance
Layout of twin shaft gas turbine, used for constituting a component based performance model.
1 41
3
2 5 4
C PTCT
Working medium: mixture of air and water vapor for calculations of thermodynamic processes
water content: ad
w
mmwar
,
=
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 4 of 22
Modeling Humidity effects on gas turbine performance
Component performance parameters transformation
Compressor
ddwwdw
RRT
NT
Nγγ ⋅=
⋅
dwwd
dw
RRP
TWP
TWγγ ⋅⋅=
ddwwdw
RRTh
Th
γγ ⋅⋅=
∆
∆
Turbines
( ) )1(1 wdddwwdw
RRT
NT
Nγγγγ +⋅⋅+⋅⋅=
⋅
( ) ( )12
112
1
21
12 −⋅
+−⋅
+
⋅
+⋅
+⋅
⋅⋅
=
d
d
w
w
d
wdw
wd
dwRR
PTW
PTW γ
γγ
γ
γγγ
γ
( ) ( )wdddwwdw
RRTh
Th
γγγγ +⋅⋅+⋅⋅=
∆
∆
11
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 5 of 22
Modeling Humidity effects on gas turbine performance
Referred ("Corrected") quantities
Rotational speed: θγ
⋅×⋅
=2874.1RNN cor
Mass flow rate: θ
γδ⋅×=
4.1287RWWcor
Temperature: θTTcor =
Pressure:
5,31
14,011
−
−
+=
−
γ
γ
δγ ppcor
Power: θγ
δγ
⋅⋅⋅=2874.14.1RPPcor
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 6 of 22
Effect of Humidity on Quantities used for Monitoring
Effect On Measured Quantities
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04war
-2
-1
0
1
2
Load
Wf
CDT
W
GG Speed
GGET
CDP
-1
0
1
2
3
4
Perc
enta
ge D
evia
tion
(%)
LoadWf
CDT
W
GG SpeedGGETEGTCDP
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04Water Air Ratio
-2
-1
0
1
2
0 20 40 60 80 100
GG SpeedWf
EGT
WCDT
CDP
GGET
0 20 40 60 80 100
0 20 40 60 80 100
Relative Humidity
15 C
25 C
35 C
(a) P=ct
(b)TIT=ct
(c)EGT=ct
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 7 of 22
Effect of Humidity on Quantities used for Monitoring
Percentage change of referred quantities for constant referred load.
-5
-4
-3
-2
-1
0
1
0 0.01 0.02 0.03 0.04war
Δ %
CDPcorr
CDTcor
EGTcor
Wcor
N1cor
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 8 of 22
Compressor delivery temperature - power output
Different relative humidity levels, related uncertainty bands
-15
-10
-5
0
5
88 90 92 94 96 98 100 102
Load %
CD
T-C
DT n
om (
o C)
0 %
30%
60%
90%
. If humidity is not accounted for: -ability for diagnosing malfunction is reduced, when based on performance quantity interrelations - changes caused by humidity may be large enough to be misinterpreted as the occurrence of a malfunction, and thus have a false alarm.
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 9 of 22
Effect of Humidity on Quantities used for Monitoring
Effect on Measurement Deviation Patterns
-3
-2
-1
0
P W1 CDP CDT(C) EGT(C)
(a)Δ %
-3
-2
-1
0
P W1 CDP CDT(C) EGT(C)
(b)Δ %
(a) war=0.009 (b) IGV fault. Signatures for N1 constant
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 10 of 22
Effects On Monitored Parameters Condition assessed using “health indices” Technique used "ADAPTIVE MODELLING" Compressor
( ) ( )refPTWPTWf 2222221 //)/= ,
refCCf ,2 /ηη= Gas Generator Turbine
( ) ( )refPTWPTWf 4444445 ///= ,
refGGTGGTf ,6 /ηη= Power Turbine
( ) ( )ref
PTWPTWf 1.41.41.41.41.41.47 /// ⋅⋅=
refPTPTf ,8 /ηη= Measurements collected from the engine, fed to the adaptive model, without properly accounting for humidity, produce health parameters different form their nominal values.
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 11 of 22
Humidity Effect on Component Health Indices
Modification factors deviation, for humid air data processed by dry-air adaptive-model
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04war
-6
-5
-4
-3
-2
-1
0
1
2
Perc
enta
ge D
evia
tion
(%)
f8
f2
f1
f4f6
f5f7
.
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 12 of 22
Measurement data from an industrial gas turbine
Layout of gas turbine and quantities measured for monitoring
Inlet pressure, Temperature
CDP,CDT
EGT N1
Airflow
Load
.
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 13 of 22
Daily variation of ambient conditions
15
20
25
30
35
40
45
0 2 4 6 8 10 12 14 16 18 20 22 24
Hour
Ambi
ent T
empe
ratu
re
Day 1 Day 2 Day 3 Day 4Day 5 Day 6 Day 7 Day 8
20
30
40
50
60
70
80
0 2 4 6 8 10 12 14 16 18 20 22 24
Rela
tive H
umid
ity
0.008
0.01
0.012
0.014
0.016
0 2 4 6 8 10 12 14 16 18 20 22 24Time of Day (h)
war
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 14 of 22
Variation of ambient conditions over a week
70
75
80
85
90
95
100
1 2 3 4 5 6 7 8Time (Days)
Load
(%)
15
20
25
30
35
40
45
1 2 3 4 5 6 7 8Time (Days)
Tam
b (o
C)
0.008
0.01
0.012
0.014
0.016
1 2 3 4 5 6 7 8Time (Days)
war
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 15 of 22
Referred measured quantities for low and high humidity level
96.0
97.0
98.0
99.0
100.0
90.00 91.00 92.00 93.00 94.00 95.00 96.00Load/(δ θ) %
CD
Pco
r %
99.0
99.5
100.0
100.5
90.00 91.00 92.00 93.00 94.00 95.00 96.00Load/(δ θ) %
EG
Tcor
99.0
99.2
99.4
99.6
99.8
90.00 91.00 92.00 93.00 94.00 95.00 96.00Load/(δ θ) %
N1c
or %
98
99
100
101
90.50 91.00 91.50 92.00 92.50 93.00 93.50 94.00 94.50 95.00 95.50Load/(δ θ) %
CDTc
or
94
95
96
97
90.00 91.00 92.00 93.00 94.00 95.00 96.00Pcor %
W1c
or
Points around Instants A and B
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 16 of 22
Estimation of Measurement Deviations
Predicted and measured differences
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
CDT CDP W NGG EGT
Δ %
Predicted Measured
Comparison of Operating conditions of low(war~0.008) and high ambient humidity ( war~0.014)
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 17 of 22
.
Health indices deviation for different levels of humidity
.
Sensitivity of the individual indices to humidity
0 0.01 0.02 0.03 0.04war
-5
-4
-3
-2
-1
0
1
Perc
enta
ge D
evia
tion
(%)
f2
f1
f6
f5
f7
f1,f2,f5,f6,f7 estimated from N1, CDP, CDT, EGT, W2 Indices mostly influenced: f1 , f7
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 18 of 22
Time evolution of Health-Indices
Compressor flow capacity evolution
-3-2-10123
0 1 2 3 4 5 6 7 8
df1(
%)
-3-2-10123
0 1 2 3 4 5 6 7 8Time (days)
df1(
%)
(a)
(b)
(a) standard humidity of 60%, (b) actual measured humidity
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 19 of 22
Humidity induced variation compared to actually expected variations
-3-2-10123
0 1 2 3 4 5 6 7 8
df1(
%)
-3-2-10123
0 1 2 3 4 5 6 7 8Time (days)
df1(
%)
(a)
(b)
Impact of fouling on Compressor flow capacity, and the effect of compressor wash
-3
-2
-1
0
1
2
3
0 10 20 30 40 50Time (Days)
Df1
%
Compressor Wash
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 20 of 22
Time evolution of Health-Indices
Power turbine swallowing capacity evolution
-3
-2
-1
0
1
2
3
0 1 2 3 4 5 6 7 8Time (Days)
df7
(%)
-3-2-10123
0 1 2 3 4 5 6 7 8Time (Days)
df7
(%)
(b)
(a)
(a) standard humidity of 60%, (b) measured humidity
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 21 of 22
Some Practical Aspects
• Humidity measurement essential for accurate monitoring
• Humidity influence more important when large
variations in ambient absolute humidity are expected.
o (This would be the case more for hot and humid climates)
• Referred variables to be used for comparing data from different humidity levels.
o The reference condition can be chosen to be representative of an average day, but it does not practically influence the magnitude of observed changes, if it is consistently used.
• It is preferable to use the standard absolute humidity
for calculations instead of keeping the relative humidity constant or assuming dry air (if a measurement is not available)
• Monitoring turbines with inlet humidification
requires the support by models incorporating humidity.
LABORATORY OF THERMAL TURBOMACHINES NATIONAL TECHNICAL UNIVERSITY OF ATHENS
Uncertainty Reduction in Gas Turbine Performance Diagnostics by Accounting for Humidity Effects 01-021
Page 22 of 22
SUMMARY-CONCLUSIONS
• Effect of humidity on quantities related to gas turbine performance monitoring analyzed. A twin shaft industrial gas turbine was used as a test case.
• A method of incorporating humidity effects in an engine performance model described, as well as a method to "refer" measured performance variables
• The model was used to demonstrate how measured quantities deviate from their values in operation with dry air.
• If humidity is not taken into account, uncertainty bands in measured quantities may be too broad reducing the discriminative ability for diagnostics.
• Measurement deviations caused by humidity changes could be of magnitude similar to the magnitude of deviation caused by faults
• "Health” parameters exhibit apparent deviations if humidity is not correctly accounted for
• Incorporating a measurement of humidity in calculations eliminates apparent trends while it reduces the scattering of “health” parameter values.