Click here to load reader

View

17Download

1

Embed Size (px)

Wavelet and ANN Based Detection ofInrush and Internal Faults in Power Transformers

Dr. C. VenkateshProf. and HoDDept. of EEE

SR Engineering CollegeWarangal, AP

Introduction Transformer Faults Differential Protection Inrush Current Fault Current and Inrush Current Detection Wavelet Transforms Artificial Neural Networks Conclusion

Presentation Sequence

Power transformers are one of the most

important components of power system

network.

Their protection is of utmost importance.

Any maloperation can lead to heavy power

losses and monetary losses.

Faults in Power Transformer

INTERNAL FAULTS

Terminal to ground fault

Turn to turn fault

Core insulation failure

Phase to ground fault

Phase to phase fault

Turn to turn fault

Core insulation failure

EXTERNAL FAULTS

Open circuit fault

External system short circuits

Line-to-ground fault (LG)

Line-to-line fault (LL)

Line-to-line to ground fault (LLG)

3-phase fault (LLL)

3-phase to ground fault (LLLG)

DIFFERENTIAL PROTECTIONThe electric protective relaying of power transformersis based on percentage differential relaying technique.

These compare currents from all the terminals to a predetermined threshold .

In case of an internal fault the circuit breaker isolates the transformer from the system

INRUSH CURRENT

When the transformer is suddenly connected to an ACvoltage source, there may be a substantial increase in theprimary winding current, known as inrush current.

The phase relation is that flux and current are in phase andlag voltage by 90 degrees.

Voltage at zero positionVoltage at its positive peak

Continuous Operation of Power Transformer

Cold Start of Power Transformer

Inrush Current and Resulting Differential Current

Transformer Equivalent Circuit

Differential protection discriminates between internal and external faultcurrents

But fails to discriminate between fault and inrush current.

The magnitude of magnetization inrush current is similar to internal faultcurrent and inrush current.

Inrush is a transient phenomenon and doesnt require relay operation.

Relay to be operated only for internal faults and not for inrush

Many relays are of slow acting type so that they are not operated for inrush.

A method to discriminate between inrush and fault current, to prevent anymaloperation of the relay is required.

EARLIER ALGORITHMS USED FOR DISCRIMINATION AND THEIR

DISADVANTAGES The first method was by using the magnitude of second

harmonic component.

However during extensive fault conditions the secondharmonic of fault current is greater than that of inrush.

Then there was the algorithm using wavelet transforms withfuzzy systems

However the major drawback of this was the need of fuzzylaws which require extensive training patterns

Then there was the algorithm using wavelet transform usingfeed forward neural networks

Then came mathematical models like support vector machineand Gaussian mixture model

These suffered the drawback of requirement of large datawindow for inputs and were comparatively less effective thanartificial neural networks.

PROPOSED ALGORITHM Capture one cycle of primary current Obtain differential current Id=Ip-Is Calculate DWT of primary current Obtain standard deviation of decomposed levels d1,d2,d3 . These parameters are given to ANN as input data to

discriminate the faults and inrush that is healthy condition.

If ANN output is discriminated as fault, then issue trip signalotherwise proceed further i.e. monitor the differentialcurrent.

SINGLE PHASE TRANSFORMER

Single phase transformer simulation circuit

INRUSH CURRENT ANALYSIS

Inrush current at t=0.04sec

Inrush current at t=0.045sec

Inrush current at t=0.052ec.

INTERNAL FAULT CURRENT ANALYSIS

Primary fault current at t=0.04sec Primary fault current at t=0.045sec

WAVELET ANALYSIS OF PRIMARY CURRENTS

Coefficients obtained from the inrush current waveform

time d1 d2 d30.04 1.055 5.926 41.54

0.041 0.8699 7.486 52.370.042 0.7731 9.461 62.060.043 0.7869 9.199 66.640.044 0.8488 10.55 80.810.045 0.8818 10.4 93.540.046 0.8047 9.529 76.870.047 0.7619 8.604 65.310.048 0.7993 8.743 51.830.049 0.9599 6.434 40.07

0.05 1.049 5.304 43.350.051 0.938 7.287 49.590.052 0.7419 9.636 60.090.053 0.7927 9.05 67.880.054 0.8511 10.36 80.790.055 0.8684 10.55 95.440.056 0.7953 9.52 67.840.057 0.7331 8.835 62.440.058 0.8514 8.836 46.060.059 0.989 6.193 41.54

0.06 1.159 5.963 43.49

standard deviation values of inrush current detail coefficients

WAVELET ANALYSIS OF PRIMARY CURRENTS

Coefficients obtained from internal fault current waveform(terminal to ground fault)

time d1 d2 d30.04 83.95 199.7 2565

0.041 198.4 408.4 19320.042 478.8 902.5 51460.043 499 1631 73730.044 179.5 288.6 16340.045 161.8 220.4 14300.046 166.1 172.3 16280.047 484.2 2574 41370.048 175.8 740.7 22340.049 190.1 436.4 17870.05 15.21 46.98 77.02

0.051 160.6 573.5 15720.052 296.9 1262 23840.053 70.65 166.6 12700.054 43.58 142.6 945.90.055 54.41 196.5 11390.056 39.14 168.5 11670.057 112.9 216.3 12510.058 49.1 133.3 11790.059 135.9 401.7 730.10.06 70.45 131 916.6

standard deviation values of primary current for internal fault on primary

time d1 d2 d30.04 579.8 1252 2735

0.041 376.3 1138 17800.042 291.9 928.6 527.40.043 282.9 962.6 15400.044 287.1 469.4 557.40.045 226.7 751.4 725.90.046 222.4 203.1 886.80.047 213.5 562.1 531.10.048 303.1 1283 1.36E+040.049 275.8 1628 1.53E+04

0.05 357.9 2481 1.66E+040.051 573.7 3468 2.01E+040.052 666.4 3586 7.65E+030.053 362.2 3677 1.28E+040.054 764.6 4033 1.16E+040.055 596.9 3103 1.21E+040.056 561.5 3189 2.25E+040.057 680.3 3602 1.87E+040.058 359.2 3653 1.60E+040.059 530 3756 2.13E+04

0.06 680.1 3892 3.98E+03

WAVELET ANALYSIS OF PRIMARY CURRENTS

standard deviation values of primary current for internal fault on secondary

Structure of ANN

Data set One input layer with three inputnodes,One hidden layer with four nodesandan output layer with two outputnodes

45 values are given for training and 18 values are used to test the ANN

ANN Training:

Training output Testing output

THREE PHASE TRANSFORMER

INRUSH CURRENT ANALYSIS

t = 0.04 sInrush waveform of phase A

Inrush waveform of phase B

Inrush waveform of phase C

INTERNAL FAULT CURRENT ANALYSIS ABG fault on primary side of transformer at t = 0.04 s

Phase A current

Phase B current

Phase C current

INTERNAL FAULT CURRENT ANALYSIS ABG fault on secondary side of transformer at t = 0.04 s

Phase A current

Phase B current

Phase C current

Structure of ANN

Data set

One input layer with threeinput neurons,One hidden layer with fourneurons andan output layer with twooutput neurons.

63 values are given for training and 42 values are used to test the ANN

ANN for each phase current analysis

Training output

Network A

Network B

Network C

Test output

Network A

Network B

Network C

LOAD FLOW ANALYSIS OF A IEEE 14 BUS SYSTEM USING BACK PROPAGATION ALGORITHM

Technique Employed: Artificial Neural Network.Algorithm: Back Propagation Algorithm.

BUS NO.1 IS TAKEN AS SLACK BUSNO. OF PQ BUSES = 9NO. OF PV BUSES = 4

At each bus Pgen, Qgen, Pload and Qload are givenPinj = Pgen Pload andQinj = Qgen - Qload

are calculated and are inputsHENCE, NO. OF INPUT NODES = 13*2 = 26.

For pq buses we need to calculate |V| and

output:

ETA= 0.50

Iterations= 4534

Error Rate= 0.000100

output:ETA= 0.500000 Iterations= 3714 Error Rate= 0.000100ETA= 0.550000 Iterations= 3895 Error Rate= 0.000100ETA= 0.600000 Iterations= 3427 Error Rate= 0.000100ETA= 0.650000 Iterations= 4308 Error Rate= 0.000100ETA= 0.700000 Iterations= 3449 Error Rate= 0.000100ETA= 0.750000 Iterations= 3965 Error Rate= 0.000100

ETA= 0.800000 Iterations= 4420 Error Rate= 0.000100

CONCLUSION

Efficient algorithm has been developed for discrimination of inrushcurrent and internal fault current in power transformers

Wavelet-ANN detection technique demonstrated.

ANN application for Load Flow Analysis discussed.

Thankyou