Towards Cost-Effective Maintenance of Power Transformer by Accurately Predicting Its Insulation Condition

8/12/2019 Towards Cost-Effective Maintenance of Power Transformer by Accurately Predicting Its Insulation Condition

1/17

TOWARDS COST-EFFECTIVE MAINTENANCE

OF POWER TRANSFORMER BYACCURATELY

PREDICTING ITS INSULATION CONDITION

Refat Atef Ghunem1, Khaled Bashir Shaban2, Ayman Hassan El-Hag3, and Khaled Assaleh3

1 Electrical and Computer Engineering Department, University of Waterloo

2 Computer Science and Engineering Department, Qatar University

3 Electrical Engineering Department, University of University of Sharjah


2/17

INTRODUCTION

Transformer asset management has been a priority

to power utilities: Improving condition monitoring and diagnosis

methods

Preparing solid asset maintenance plans

Maintenance costs are found to be heavy burden

on utilities

Developing a well constructed determining process

for transformer health index


3/17

MEGGER TEST

A classical time-based preventive maintenance test

Easy, fast, and inexpensive test to be conducted

Indicates insulation deterioration

Efficient tool for developing trend analysis

A diagnostic tool in unplanned outages


4/17

MEGGER TEST DRAWBACKS

Temperature effect

Rule in practice: factor of two is attributed tomegger test readings for every 10 degrees

change

Moisture Content effects

Accelerating deterioration

Reducing the dielectric strength

Faults are not detected by the insulation

resistance parameter


5/17

TRADITIONAL MAINTENANCE PRACTICES

Reducing the unplanned outage cost and period for

tripped transformer insulation inspectionRule-of-thumb: if the transformer trips due to the

operation of differential, restricted earth fault and

Buchholz relays , insulation inspection must be

conducted before reconnecting to the network

Megger test clearance of abnormal situation is not

reliable for reconnecting the transformer

Supportive tests should be conducted


6/17

SUPPORTIVE TESTS

Other tests are conducted to enhance the diagnosis reliabilityof the megger test:

Oil Breakdown Voltage (BDV)

Oil water content Temperature correction should be considered

Due to equilibrium process in the oil-paper system

Dissolved-gas analysis (DGA):

CO2/CO ratio

Total Dissolved Combustible Gases (TDCG)

BUT, costly ($200/sample) and time consuming(few days)


7/17

TOWARDS COST-EFFECTIVE MAINTENANCE

Optimizing the number of time- based maintenance

tests

Predicting oil quality and dissolved gases

parameters informs about the critical tests that

should be taken

Developing well constructed asset maintenance

plans with optimized number of tests and periodsbetween them


8/17

THE LITERATURE

8

Authors Model Predictors Output Accuracy Comments

2009&

2008

Khaled Shaban, A. H.

El-Hag, and Andrei

Matveev

&

Khaled Assaleh &

Ayman El- Hag

Artificial

Neural Network

(ANN)

&

Polynomial

Network

Megger

Breakdown

voltage(BDV),oil

water content,

interfacial tension(IF)

and acidity

&BDV

Water content

IF

BDV:84%

Water content:60%

IF:95%

Acidity:75%

&

Water content: could not be

estimated

IF:93%

BDV:84%

BDV and IF are indication for

oil contamination only

2004

&

1999

Mohamed Ahmed Abdel

Wahab

&

Mohamed A.A. Wahab,

M.M. Hamada, andA.G. Zeitoun, G. Ismail

ANN

&

Polynomial

Regression models

Oil acidity , oil water

content and transformer

age

&

Service period

BDV

&

BDV, oil water

content and acidity

More than 90% for both

BDV is not comprehensive and

&

No. of data is very small

2009

&

2008

Sheng-wei Fei, Cheng

Liang liu, and Yu bin

Miao

&

Sheng-Wei Fei, and Yu

Sun

Support vector

machine with

genetic algorithm

DGA DGA trend

Less than 90%

&

More than 90%

It is not practical

2000

&

2003

I N da Silva, A N de

Souza, R M C Hossri,and J H C Hossri

&

J.S.Navamany, and P.S.

Ghosh

ANN

furan, volume of carbon

monoxide (CO),Interfacial tension and

acidity

&

CO ,CO2 and furfural

Aging Degree

N/A

&

Around 90%

It is not cost-effective

2009

Ali Naderian Jahromi,

Ray Piercy, Stephen

Cress, Jim R.R. Service

andWang Fan

Weighted averageAll standard transformer

tests

Transformer health

indexN/A

The weights are not assigned

based on quantities correlations

&

Relatively hard to be practiced


9/17

ANN FOR PREDICTION

Artificial neural network (ANN) is used as a

modeling technique Multi-layer feed forward ANN with Back

propagation algorithm


10/17

THE EXPERIMENT

Data from transformer maintenance records of 19 power transformers in Abu DhabiTransmission and Despatch Company (TRANSCO) are used

Voltage rating of 220/33/11KV

Rating is in the range of 50-140MVA

Range of 7-15 years into service

54 samples are used to predict oil BDV

31 samples are used to predict oil water content

32 samples are used to predict TCG

33 samples are used to predict CO2/CO ratio

All the samples are used for training and testing to predict all the available data


11/17

SAMPLE OF THE DATASET

Insulation Resistance

(M) Parameters

HV/E LV/E TV/E

BDV

(kV)

Water

Content

(ppm)

TDCG

(ppm) CO2/CO

1 300 240 400 71.4 26.8 497 8.25

2 820 880 1100 75 30.57 737 7.98

3 280 280 380 74 16.46 672 6.46

4 560 580 620 67 22.62 665 6.23

5 270 220 390 75 17.63 894 4.85


12/17

RESULTS

4 ANNs:

3 neurons in the input layer (HV/E, LV/E and TV/E) 2 hidden layers with three neurons at the first

hidden layer and 10 neurons at the second hiddenlayer

1 neuron at the output layer (BDV, water content,

TDCG and CO2/CO)

Prediction rate:

Oil BDV - 96% Oil water content - 84%

CO2/CO ratio - 91%

TDCG - 88%


13/17

RESULTS

Sample NO.

0 5 10 15 20 25 30 35 40 45 50 55 60

BDVvalue(KV)

62

64

66

68

70

72

74

76

78

80

Actual oil BDV

Predicted oil BDV

Size of testing set

(samples)

Prediction

Accuracy

1 96.66

2 92.79

3 92.534 93.61

5 92.75

6 92.93

7 93.61

8 93.52

9 94.27

10 93.99

11 92.29

12 95.25

Average

Prediction

Accuracy

93.68

This result agrees with the results of the models proposed in [1] and [2]. Transformer oil

breakdown voltage gives an indication about the dielectric capability of transformer oil to

withstand high electrical stresses. Also, BDV can be an indication about the deterioration state

of the insulation system in the transformer. This is because the contamination particles

resulted from the insulation deterioration in the transformer oil decrease the BDV value. Thisexplains the high correlation presented between IR and oil BDV.


14/17

RESULTS

Sample NO.

0 5 10 15 20 25 30 35

Watercontentvalue(ppm)

16

18

20

22

24

26

28

30

32

34

Actual oil water content

Predicted oil water content

Size of testing

set (samples)

Prediction

Accuracy

1 84.73

2 80.613 80.52

4 80.13

5 78.88

6 80.16

7 79.36

8 74.39

9 81.3310 74.74

11 76.57

12 82.91

Average

Prediction

Accuracy 79.53

compared to BDV the prediction accuracy has been reduced between 10-15%. This can be attributed to

the relatively wider scatter of the data of water content (17-32 ppm) compared to BDV (63.6-78.8 kV)

and the relatively small size of the training set (31 vs. 54). Moreover, the oil temperature effect in

altering the value of the measured oil water content was not completely considered as the oil sample

was taken between 60-65 C. Such narrow band of temperature variation may still influence the

accuracy of the water content measurement. The proposed model to predict oil water content in

this study can be considered superior to other proposed models [1-2] where the effect oftemperature variation on measured samples was normalized.


15/17

RESULTS

Sample NO.

0 5 10 15 20 25 30 35

CO2

/COratio

3

4

5

6

7

8

9

10

11

Actual CO2/CO ratio

Predicted CO2/CO ratio

Size of testing set

(samples)

Prediction

Accuracy

1 91.02

2 82.313 75.60

4 73.84

5 77.27

6 83.76

7 75.84

8 67.29

9 80.4610 82.03

11 79.79

12 74.68

Average

Prediction

Accuracy 78.66

Although CO2/CO ratio is more effective in diagnosing cellulose involvement in incipient

faults rather than indicating the deterioration of cellulose, the proposed model shows an

acceptable reliability. This agrees with the practice in utilities to use CO2/CO ratio

with other insulation quality parameters as indicators for cellulose thermaldegradation.


16/17

RESULTS

Sample NO.

0 5 10 15 20 25 30 35

TDCGvalue(ppm)

200

400

600

800

1000

1200

1400

Actual TDCG

Predicted TDCG

Size of testing

set (samples)

Prediction

Accuracy

1 88.37

2 73.693 75.29

4 74.25

5 68.59

6 69.22

7 75.89

8 75.70

9 74.1710 70.59

11 73.82

12 70.11

Average

Prediction

Accuracy

74.14

Insulation resistance is affected by the deterioration and the abnormal excess of

degradation, thus, IR can be an indicative parameter to the TDCG.Although TDCG as a

parameter is a reflection of incipient faults rather than a parameter for aging

index, an acceptable and reliable correlation is verified in the proposed model.


17/17

CONCLUSIONS

Prediction techniques can improve transformerinsulation diagnosis and reduce the predictive and

corrective maintenance costs.

The proposed models suggests a decent potential forpredicting oil BDV, water content, TDCG and CO2/COratio.

It can be more efficient to test the correlations of thetransformer insulation parameters in differentconditions (wide temperature ranges, service ages and

voltage levels, etc.)

Increasing data size can enhance the generalizationcapability of the model

Documents

Towards Cost-Effective Maintenance of Power Transformer by Accurately Predicting Its Insulation Condition