SIEMENS - PTD Power Transmission and Distribution
1e
Seminario Tecnológico en Sistemas Eléctricos de Potencia
r
©
Siemens AG 2006
TMDS TRANSFORMER MONITORING AND DIAGNOSTIC SYSTEM
Page 2
Abr-07
CONSTRUCTION OF TRANSFORMERS
0,5-800 TONS 0,5-200 M3 Cu Fe Wood Insulation Paper Insulation Liquid
Page 3
Abr-07
ANSI/IEEE Std C57.91-1995, IEEE Guide for loading Mineral-Oil Immersed Transformers Part 5 Life
Expectations. IEEE Std C57.91TM-1995/Corrigendum 1 ANSI/IEEE C57.92 1981, Guide for loading mineral oilimmersed power transformers up to 100 MVA, With 55°C or 65°C Average Winding Rise.
Page 4
Abr-07
Life Expectancy
?
Basis
Normal Insulation Life Hours
Years
65 000
7.42
25% retained tensile strength of insulation
135 000
15.41
200 retained degree of Polimerization in insulation
150 000
17.12
180 000
20.55
50% Retained tensile strength of insulation (Former IEEE Std C57.92-1981 criterion)
Interpretation of Distribution Transformer functional life test data (Former IEEE C57.91-1981 criterion)
1- Tensile strength or degree of Polimerization (D.P.) Retention Values were determined by sealed TUBE AGING ON Well-Dried Insulation Samples In Oxigen Free Oil. 2- Refer to I.2. in annex I for discussion of the effect of higher values of water and Page 5
Abr-07
180.000 hours/(24 hours/day) = 7500 days Loss of Life per day “Average Expected” = 1/7500* 100% = 0.0133%
Page 6
Abr-07
s
Particles
Water
Free
Bubbles due to
Vapor
Major Insulation
fibers
Turns/ Coils
Oil aging products
metals
Bubble evolution
Oversaturation Cavitation
Increasing percent Saturation
Overheating
Decreasing the dielectric strenght
Accelerating the rate of aging
Aging Model CIGRE WG A2.18 Life Management Techniques for Power Transformers Page 7
Abr-07
After 25 years the failure rate increases exponentially
Page 8
Abr-07
H2 EMPIRICAL PLOTTED
Page 9
Abr-07
However…. Major Losses involving large Oil-Cooled transformers continue to occur on a frequent basis
Page 10
Abr-07
Analysis of transformers Failures ( over 5 years 1997-2001 Worlwide Transformer fleet.)
William H Bartley P.E. (Hartford Steam Boiler, Inspection & Insurance Co) Presentated at International Association of Engineering Insurers 36th Annual Conference – Stockholm, 2003. Period of study
1997-2001
Sampling data obtained, (n)
94
Exchange rates; (USD)
€ 1.0;
Page 11
Abr-07
Table 1- Number and Amounts of Losses by Year
Year
Total # of Losses
Total Loss
Total Property Damage
Total Business Interruption
1997 19
$ 40,779.507 $ 25,036,673 $ 15,742,834
1998 25
$ 24,932,235 $ 24,897,114 $
35,121
1999 15
$ 37,391,591 $ 36,994,202 $
397,389
2000 20
$ 150,181,779
2001 15
$ 33,343,700 $ 19,453,016 $ 13,890,684
Total
$ 286,628,811
94
$ 56,858,084 $ 93,323,695
$ 163,239,089
$ 123,389,722
* Total Looses in 2000 includes one claim with a business Interruption portion of over $86 million US
Page 12
Abr-07
Table 1A- Number and Amounts of Losses by MVA and Year
Year
Total # of Losses
Losses Total MVA w/data Reported
Total PD (w/ size data)
Cost/MVA
1997 19
9
2567
$ 20,456,741 $ 7,969
1998 25
25
5685
$ 24,897,114 $ 4,379
1999 15
13
2433
$ 36,415,806 $ 14,967
2000 20
19
4386
$ 56,354,689 $ 12,849
2001 15
12
2128
$ 16,487,058 $ 7,748
Total
78
17.199 MVA $ 154,611,408
94
* During this five year period the average cost is $USD 8,990 per MVA , or about $USD 9 per KVA Page 13
Abr-07
Table 2 – Losses By Application
Y
Generator
Industrial
Utility
Step Up 1997
$ 29,201,329
Unknown
Substations 3
$ 2,239,393
4
$
11
5,243,075 1998
$ 15,800,148
Annual Totals
8
$ 3,995,229
6
$
$
1
4.095,710
$
3,031,433
4
$ 24,922,958
4
$
11
$
$123,417,78
1
$ 24,724,182
4
$
6
$ 3,320,665
6
1
$
Abr-076
19,797,476
$
7,416,375
15
37,391,591
8 0 2,039,810 150,181,779 The largest number o claims (38) occurred in the Utility Substation Sector, But the highest paid category 2001 $ 1 $ was Generator Step Up transformers, with a total of over $ $200 million. If the4extraordinary Business32,082,501 Interruption loss is 1 ignored, the generator step up1,261,199 transformers is still significantly than other 33,343,700 cathegory. (This is to$203,533,19 be expected due to3the very large size of these Total $ 55,881,762 18 transformers) $ 38 $ 2 $ s Page 14 9
25
24,939,235
6,116,535 2000
19
40,779,507
5,136,858 1999
$
286,628,811
20
15
94
Table-3 Cause of Failures Cause of Failure
Number
Insulation Failure (pyrolisis, oxidation, acidity 18 years old is the
Total Paid
24
$ 149,967.277
Design /Material / Workmanship
22
$
64,696,051
Unknown
15
$
29,776,245
Oil Contamination
4
$
11,836,367
Overloading
5
$
8,568,768
Fire/Explosion (outside the transformer)
3
$
8,045,771
Line Surge (Switching, Spikes, short circuit strength)
4
$
4,959,691
Improper Maint/Operation
5
$
3,518,783
Flood
2
$
2,240,198
Loose Connection
6
$
2,186,725
Lightning
3
$
657,935
Moisture
1
$
157,000
94
$
average age of these transformers)
Page 15
Abr-07
Total
286,628,811
Figure 1 –Frequency -Severity of Transformer Failures
Page 16
Abr-07
Table 4 Distribution of losses by age of Transformer.
Age of failure
Numbe Cost of r of failures failure s
0 to 5 Years
9
$ 11,246,360
6 to 10….
6
$ 22,465,881
11 to 15…
9
$ 3,179,291
16 to 20…
10
$ 10,518,283
Over 25...
16
$ 16,441,930
35
$ 207,734,306
Age Unknown *
Average age at failure was 18 years
* This line includes the one claim with a business interruption element of $86 million USD Page 17
Abr-07
Figure 2 Base GVA ADDITIONS (U.S.A. COMM DPT)
U.S. Commerce Department data; The electric utility Industry reached a peak in new installations in the U.S. around 1973-74, In those two years, the U.S.A. added about 185 GVA, of Power transformers. Page 18
Abr-07
Risk model of future transformers failures
A risk model of future transformers failures, based on aging, Developed by HSB, and published in 2000. the model is based on mortality models that were first proposed in the 19th Century Behaviour of Human mortality By Benjamin Gompertz in 1825 For the instantaneous failure rate
f(t) = α
e
βt
[α] = is a constant [β] = is a time constant [t ] = time in years Page 19
Abr-07
Risk model of future transformers failures
1860 W.M Makeham, Modified the Gompertz equation adding a constant term in order to correct for including mortality due to accidental death f(t) = A + α
A+α
e e
βt
βt
f(t) =---------------1+μ [μ] = Modifier
βt
to allow the curve more closely approximate the slower rate of increasing
mortality at older ages (1932) by Page 20
e
Abr-07
W, PERKS, R.E. Berhards and Others
Figure 3 Transformer Failure Rate HSB 2000
Number of Failures = [Failure Rate] x [Population that is surviving]
The instantaneous failure rate for transformers in a given year is the proba bility of failure per unit time for the population of transformers that has survived up until time t Page 21
Abr-07
Using population profile From figure # 2 the predicted failures can be Plotted for all U.S. utility Transformers, built Between 1964 & 1962 X –axis year of predicted Failures Y-axis population of failures (GVA) ! Prediction intended to show And illustrate the magnitude of The problem ! Page 22
Abr-07
ACTION PLAN SUGGESTED BY W,H Bartley
. One conservative strategy suggests that the industry start a massive capital replacement program that duplicates the construction profile of the 60´s and 70´s But this would cause many transformers to be replaced needlessly and cost the utility industry billions of US dollars. The ideal strategy is a life assesment or life cycle management program, that sets loading priorities, and provides direction to identify a) transformers defects that can be corrected; b ) transfomers that can be modified or refurbished; c) transformers that should be re-located and d) transformers that should be retired. The insurance industry should be aware that both IEEE, and CIGRE are developing guidelines for aging transformers.
Page 23
Abr-07
REFERENCES William H. Bartley, HSB, Analysis of Transformer Failures, Proceedings of the sixty – Seventh Annual International Doble Client Conference , Boston MA, 2000. William H. Bartley, HSB, Failure History of Transformers –Theoretical Projections for Random Failures, Proceedings of the TJH2B TechCon Mesa AZ, 2001. Tim Higgins, Mathematical Models of Mortality, presented at the Workshop on Mortality Modeling and Forecasting, Australian National University, February 2003. IEEE C57.140 Draft 9 March, 2003, IEEE Guide for the Evaluation and Reconditioning of Liquid Immersed Power Transformers , Rowland James & William Bartley Co-Chair CIGRE 12-20 Guide on Economics of Transformer Management ( draft 23.7.02) CIGRE A2-18 Life Management Techniques for Power Transformers Page 24
Abr-07
Aging Factor Calculation (ANSI IEEE) Transformer Insulation Life (Part 5 IEEE Std C57.91 -1995) Experimental evidence indicates that the relation of insulation deterioration to time and temperature follows and adaptation of Arrhenius reaction rate theory that has the following form:
FAA EXP
15000 15000 383 273 H
Per unit life ( Calculated but not actual )
For WINDING HOTTEST SPOT Temperature of 110 ºC, FAA = 1
273º C 110 º C 383º C Page 25
Abr-07
Equivalent aging factor for the total time period FEQA F
AA
may be used to calculate equivalent aging of the transformer .
The equivalent life ( in hours or days ) at the reference temperature that will be consumed in a given time period for the given temperature is the following
F t t N
FEQA
n 1 AAn N n 1
n
N
Is total number of time intervals .
FAAn ∆tn
Is aging aceleration factor for the temperature wich exists during the time Is the time interval…
Page 26
Abr-07
n
∆tn interval.
Example calculations of aging Factors
Pn=43 MVA
39,5 44 27,5 30 MVA MVA 96ºC 120ºC 49ºC 55ºC Faa=0,23 Faa=2,0 Faa=0,005 Faa=0,05
©
Siemens AG 2006
Day load profile calculations
Page 28
Abr-07
Simulación Ciclo de carga de un día
En un día típico, la pérdida de vida respecto a la pérdida de vida nominal es de:
15.3% Page 29
Abr-07
Transformer Monitoring and Diagnostics System ©
Siemens AG 2006
Presentation Agenda Product Origin Introduction to TMDS System Architecture Technical Overview Actionable Information Value Proposition
Page 31
Abr-07
Product Lifecycle Early
Siemens Introduces First Monitoring Product Based on customer demand Siemens Brazil begins to investigate moving away from the “data logger” monitoring system.
Mid 2003
Development of TMDS begins
Mid 2004
TMDS Introduced
Early 2005
TMDS 3000 Introduced Adaptation of product based on requirements for application on legacy transformers begins.
February 2007 Page 32
Launch of TMDS L Abr-07
Experience List Reference List for Transformer Monitoring and Diagnostic System Rev. Feb/07
YEAR
CUSTOMER
QUANTITY
RATINGS RATED (MVA) VOLTAGE (kV)
TYPE
2007*
Furnas Viana
Brazil
3
75
345
TMDS 2000
2007*
CESP Jupiá
Brazil
4
134
440
TMDS 2000
2007*
Entergy Amite
USA
1
400
230
TMDS 3000
2007*
ENAP
Chile
2
80
110
TMDS 2000
2007*
Enelven
Colombia
4
83
138
TMDS 3000
2007*
Enelven
Colombia
6
42
138
TMDS 3000
2007*
Enelco
Colombia
8
40
115
TMDS 3000
2007*
UHE Lajeado
Brazil
4
320
230
TMDS 2000
2007
Braskem
Brazil
1
50
69
TMDS 2000
2006
Entergy Sterlington
USA
1
616
525
TMDS 3000
2006
Entergy Indian Point
USA
2
629
345
TMDS 3000
2006
Entergy Dell
USA
1
672
525
TMDS 3000
2006
Furnas Rio Verde
Brazil
4
15
230
TMDS 2000
2006
UHE Tucuruí
Brazil
12
405
550
TMDS 2000
2006
Furnas Serra da Mesa
Brazil
7
80
525
TMDS 2000
2006
Eletronuclear Angra dos Reis
Brazil
4
490
525
TMDS 2000
2005
Cemig Vespasiano
Brazil
2
300
525
TDMS 2000
2005
Furnas Vitória
Brazil
4
75
345
TMDS 2000
2005 2005
Itaipú Furnas São José
Brazil Brazil
6 7
275 200
525 525
SITRAM SITRAM
* On-going project
Page 33
COUNTRY
Abr-07
Product Definition
TMDS is an automated engineering tool that collects operational data and analyzes it using industry accepted practices, thus creating actionable information that can be applied to better manage the subject transformer. Page 34
Abr-07
Product Benefits Actionable information generated by TMDS can be used to:
• Reduce the frequency and duration of unscheduled transformer related outages thereby reducing O&M costs. • Provide forewarning of impending failure, thus leading to corrective action avoid catastrophic failure and reduce asset and collateral damage. • Better operation/utilization of transformer. • Facilitate a shift from a time, to condition based transformer maintenance policy. • More accurately evaluate deferment of unit replacement. • Justifiably seek a reduction in the premium paid for business interruption insurance. Page 35
Abr-07
System Configurations
Page 36
Abr-07
Base System The base system is a Consolidated Monitoring Package (CMP) that includes the GAS-Guard sensors, online monitoring of sensor output, access to Siemens transformer engineers for assistance with analysis, system troubleshooting and maintenance.
Page 37
Abr-07
The GAS-Guard sensors are self contained fully automated closed loop gas chromatographs. The GAS-Guard sensors offer transformer owners advanced field based dissolved gas analysis (DGA) capability.
Advanced System Standard Data Analysis Models Statistical Significance Analysis Module Limit Adaption (Learning System) Oil Level Model Cooling System Effectiveness Model Thermal Model 2 Advanced Winding Hot Spot Thermal Model Load Factor Model DGA Model (3 Gas) Oil Moisture Content Model Ambient Air Temperature: RTD PT100 Tank Oil Level: Analogue Output
Skin Temperature (1): RTD PT100
Top Oil Temperature: RTD PT100
Fan Motor Current (Bank B): Transducer
Fan Motor Current (Bank A): Transducer Oil Pump Motor Current (Bank A): Transducer
Oil Pump Motor Current (Bank B): Transducer Current Load (Single Phase): Transducer
Page 38
Abr-07
Gas in Oil Monitoring: GAS-Guard 3
Bottom Oil Temperature: RTD PT100 Moisture in Oil: Invasive
Premium System Standard Data Analysis Models
Ambient Air Temperature: RTD PT100 Skin Temperature (1): RTD PT100 Fan Motor Current (Bank B): Transducer
Statistical Significance Analysis Module Limit Adaption (Learning System) Oil Level Model Cooling System Effectiveness Model Thermal Model 2 Advanced Winding Hot Spot Thermal Model Load Factor Model Thermal Limit Prediction Model Bushing Monitor: Load-ability Recommendation Model Doble, HSP Oil Vapour Pressure Model Relative Aging Factor Tank Oil Level: DGA Model (8 Gas) Analogue Output Oil Moisture Content Model Top Oil Temperature: Load Shedding Recommendation Model RTD PT100 Bushing Monitoring Model Fan Motor Current (Bank A): Transducer Oil Pump Motor Current (Bank A): Transducer
Oil Pump Motor Current (Bank B): Transducer
Bottom Oil Temperature: RTD PT100 Gas in Oil Monitoring: GAS-Guard 8 , Kelman Moisture in Oil: Invasive Current Load (Single & Three Phase): Skin Temperature (2): Transducer RTD PT100
Page 39
Abr-07
bushings
oil level
top oil
amb temp
bottom oil gas
moisture oil/water flow
Page 40
Abr-07
air flow
LTC
Data Storage Strategy
Temp
Efficient Data Handling for Base Data Captures 3 Values Vs 24
V2 V3 V1
T Only significant Data is Captured
Page 41
Abr-07
Statistical Definition: Used to Trigger Model Calculations Statistical Definition changes are monitored for potential statistical breakouts by analyzing shifts of means and spread of standard deviations
Statistical Definition is created During defined learning period - Mean, Std Dev of Population with 95% confidence interval
-1, -2
+1, +2
-1, -2
+1, +2
This system learns! Based on operating data to avoid nuisance trips, alarms and can detect statistically significant breakouts before hard limits are reached Page 42
Abr-07
Absolute & Rate of Change Metrics: Used to Trigger Model Calculations
T
A
Page 43
These parameters can be established using: 1. Industry accepted guidelines such as IEEE/ ANSI standards 2. Calculations made based on off-line condition assessment or unit history such as past DGA or Doble test results. Abr-07
Standard Architecture
Page 44
Abr-07
E-Mail Messaging
[email protected]
TMDS Message
Siemens Transformer Monitoring and Diagnostics System (TMDS) at “S” Substation has detected an event which demands action. Please check the web address: www.siemens-tmdsl-subs.com
Page 45
Abr-07
Diagnostics
Page 46
Abr-07
Diagnostics Duval Triangle (IEC 60599-1999 Annex B.3)
Page 47
Abr-07
Data Flow Chart
Page 48
Abr-07
Monitoring Condition Operation Information Treeing
Page 49
Abr-07
Monitoring Condition Operation Information Treeing
Page 50
Abr-07
Thank you!
Page 51
Abr-07
