p r 11 hydro p.o. box 12400. st. john's. ni hydro place ... · newfoundland and labrador hydro...

Ai\p r I newfoundland labrador

11 hydro a nalcor energy company

Hydro Place. 500 Columbus Drive.

P.O. Box 12400. St. John's. NI

Canada A1B 4K7

t. 709.737.1400 f. 709.737.1800

www.n1h.nl.ca

December 17, 2014

The Board of Commissioners of Public Utilities

Prince Charles Building

120 Torbay Road, P.O. Box 21040

St. John's, Newfoundland & Labrador

A1A 5B2

Attention: Ms. Cheryl Blundon

Director Corporate Services & Board Secretary

Dear Ms. Blundon:

Re: Newfoundland and Labrador Hydro - the Board's Investigation and Hearing into Supply Issues and Power Outages on the Island Interconnected System: Supplementary Response in Relation to PUB-NLH-457 and PUB-NLH-458

In its responses to the Board's RFIs PUB-NLH-457 and PUB-NLH-458, Hydro indicated that additional

documentation would be supplied to the Board when the related work was completed. In this regard,

please find enclosed the original and 12 copies of the following:

a) Hydro's analysis of the impact of transmission line contingencies on system losses related to

alternate generation dispatches (re: PUB-NLH-457); and,

b) Two reports by Trans Grid Solutions Inc. related to the simulation of the Sunnyside Ti failure and

an investigation of the Western Avalon T5 transformer failure concerning whether or not

harmonics or system resonance were contributing factors to the system events of January, 2014

(re: PUB-NLH-458).

We trust the foregoing is satisfactory. If you have any questions or comments, please contact the

undersigned.

Yours truly,

NEWFOUNDLAND AND LABRADOR HYDRO

Tracey UPennell

Legal Counsel

TLP/jc

cc: Gerard Hayes — Newfoundland Power

Thomas Johnson — Consumer Advocate Paul Coxworthy — Stewart McKelvey Stirling Scales

Danny Dumaresque

ecc: Roberta Frampton Benefiel — Grand Riverkeeper Labrador

Engineering Support Services for:

Western Avalon Transformer T5 Tap Failure

Newfoundland and Labrador Hydro

Attention:

Mr. Peter Thomas

Report R1335.02.01

PSCAD Investigation of the Western Avalon T5 Transformer Failure

Prepared by:

TransGrid Solutions Inc. 200 – 150 Innovation Dr.

Winnipeg, MB R3T 2E1

CANADA

www.transgridsolutions.com

Nov.26th, 2014

Reference: PUB-NLH-458 Atth 1

Newfoundland and Labrador Hydro Western Avalon Transformer T5 Tap Failure


Disclaimer

This report was prepared by TransGrid Solutions Inc. (“TGS”), whose responsibility is limited to the scope of work

as shown herein. TGS disclaims responsibility for the work of others incorporated or referenced herein. This report

has been prepared exclusively for Newfoundland and Labrador Hydro and the project identified herein and must

not be reused or modified without the prior written authorization of TGS.

Revisions

Project Name: Western Avalon Transformer T5 Tap Failure

Document

Title:


Document

Type:

Draft report

Document No.: R1335.02.01

Last Action

Date:

Nov.26th, 2014

Rev.

No.

Status Prepared By Checked By Date Comments

00 IFC D. Kell 20/11/14

01 IFA D. Kell 26/11/14 Add harmonic analysis

02 ABC D. Kell 05/12/14

Legend of Document Status:

Approved by Client ABC

Draft for Comments DFC

Issued for Comments IFC

Issued for Approval IFA

Issued for Information IFI

Returned for Correction RFC

Approval not Required ANR




R1335.02.01 2 Nov.26th, 2014

© TransGrid Solutions Inc, 2014

Table of Contents 1. Introduction .................................................................................................................................................. 3 2. PSCAD Model Development .......................................................................................................................... 5

2.1 AC System model ................................................................................................................................... 5

2.2 AC Transmission Lines ............................................................................................................................ 7

2.3 Power Transformers .............................................................................................................................. 9

2.4 T5 Transformer at Sunny Side and T1 Transformer at Western Avalon.................................................... 9

2.5 PSCAD Model Benchmarking Against PSS/E .......................................................................................... 11

3. Modeling of Fault Sequence ........................................................................................................................ 12 4. Results......................................................................................................................................................... 13

4.1 Time Domain Simulation ...................................................................................................................... 13

4.2 Harmonic Impedance Scan ................................................................................................................... 20

5. Conclusions and Recommendations ............................................................................................................ 21

List of Figures Figure 1-1 Example of Low resolution Waveforms ................................................................................................... 3

Figure 2-1 SLD of Newfoundland System ................................................................................................................. 6

Figure 2-2 Example Line representation .................................................................................................................. 8

Figure 2-3 Detailed Line Model ............................................................................................................................... 9

Figure 2-4 T1 - SSD T5 – WAV ................................................................................................................................ 10

Figure 4-1 PSCAD Simulation of remnant flux ........................................................................................................ 13

Figure 4-2 Energization of Western Avalon transformer T5 via line TL237 ............................................................. 14

Figure 4-3 SSD DFR traces for T5 Failure ................................................................................................................ 15

Figure 4-4 TL207 Values ........................................................................................................................................ 16

Figure 4-5 TL203 Values ........................................................................................................................................ 16

Figure 4-6 T5 Voltages .......................................................................................................................................... 17

Figure 4-7 T5 Currents and Flux ............................................................................................................................ 18

Figure 4-8 FFT of 230kV ac voltage ........................................................................................................................ 19

Figure 4-9 Harmonic Impedances as seen from WAV bus ...................................................................................... 20

List of Tables

Table 2-1 Frequency Dependent AC lines ................................................................................................................ 7

Table 2-2 SSD T5 - WAV ........................................................................................................................................ 10

Table 2-3 SCL Comparison .................................................................................................................................... 11

Table 2-4 Loadflow Comparison ............................................................................................................................ 11

Table 3-1 AC System Fault Sequence ..................................................................................................................... 12




R1335.02.01 3 Nov.26th, 2014


1. Introduction

This report presents the findings of the PSCAD simulation to recreate the Newfoundland T5 transformer tap

changer fault that occurred at 12:22pm on January 4th, 2014.

The original scope of the project was to model the fault based on delivered Comtrade data which records the

pertinent waveforms. Upon receipt of the data, the following was noticed:

1. The Comtrade data delivered forthe events at :

04/01/2014,13:13:09.000260

04/01/2014,10:05:34.000260

Both of which did not match the time of the fault. The Comtrade data which had the same time stamp as the

fault was delivered as a pdf file, which is shown in Figure 4-3.

2. The Comtrade data had a very low resolution as shown in Figure 1-1, which made it very hard to match to the

PSCAD simulation.

Figure 1-1 Example of Low resolution Waveforms

A PSCAD model was created on the following information in order to recreate the fault based on the pdf Comtrade

file.

Jan 4 1200.sav

PSS/E data file which represents the Newfoundland System conditions prior to the fault

Peak Case.dyr

PSS/E dynamics file

WAVTS T5 transformer nameplate, test data and transformer data (delivered as pdf files)

- 1 0 0

- 5 0

0

5 0

1 0 0

1 0 1 5 2 0 2 5 3 0

S S T S C F 1 C > 1 _ A -2 3 0 K V T L 2 0 2 A P h a s e V o l t a g e (V )

E le c tr o te k C o n ce p ts ® T O P , T h e O u tp u t P r o c e sso r ®

Ma

gn

itu

de

(

Ma

g)

T im e ( m s )




R1335.02.01 4 Nov.26th, 2014


AC Line geometric data based on past work performed by TGS. This included mutual coupling of

the lines that shared the same right-of-way.

AMEC report “Newfoundland and Labrador Hydro Transmission Availability” March 21, 2014.

This report gave the details of the sequence of events during the fault of January 4th.




R1335.02.01 5 Nov.26th, 2014


2. PSCAD Model Development

2.1 AC System model

The following files provided by Newfoundland and Labrador Hydro (NLH) and were used as the base case to

develop the equivalent PSCAD model:

Jan 4 1200.sav

Jan 4 1200.dyr

The equivalent ac system in PSSE was converted to PSCAD. Since this study deals with a transient phenomenon as

a result of transformer failure at Sunnyside, the representation of the power system components with more

accuracy at higher frequencies than the fundamental frequency is crucial for a better analysis of the event.

Therefore the following changes were made to the converted case:

All the generators were represented with their associated dynamic data, with the exception of

the BDP G7 unit as there was some stability issues associated with this machine. Due to the short

duration of the simulation, this will not affect the accuracy of the simulation.

A majority of the transmission lines surrounding the Sunnyside and Western Avalon were

modeled with frequency dependent line models according to the tower geometries and

conductor configurations

T5 transformer at Western Avalon was represented based on the latest data provided by NLH

Figure 2-1 shows the single line diagram of the ac system used in this study. Please note that the complete case is

not shown in the diagram and only the main portion of the ac system surrounding Sunnyside and Western Avalon

is shown.




R1335.02.01 6 Nov.26th, 2014


Figure 2-1 SLD of Newfoundland System




R1335.02.01 7 Nov.26th, 2014


2.2 AC Transmission Lines

The major transmission line models in the PSCAD case which when originally converted from PSS/E, were

represented as pi-sections and Bergeron line models

The Bergeron and pi-section line representations are only adequate for studies that essentially require the correct

fundamental frequency impedance. However, in order to get a higher degree of accuracy, some of the lines near

the affected buses were replaced with frequency dependent line models.

Table 2-1 lists of the lines that were modeled as frequency dependent lines.

Table 2-1 Frequency Dependent AC lines

From Bus# To Bus# From Bus# To Bus#

195229 195230 195167 195169

195222 195227 195169 195171

195221 195216 195171 195173

195216 195215 195173 195175

195221 195220 195222 195229

195220 195218 195227 195229

195215 195208 195229 195236

195215 195205 195229 195234

195205 195536 195234 195236

195205 195208 195236 195238

195622 195625 195221 195222

195625 195627 195152 195153

195124 195122 195153 195154

195122 195120 195155 195157

195120 195115 195157 195159

195115 195112 195152 195159

195112 195111 195620 195620

195209 195210 195621 195622

195208 195209 195112 195113

195205 195206 195600 196500

195500 195620 - -




R1335.02.01 8 Nov.26th, 2014


Figure 2-2 shows an example of one of the more detailed line representation, which includes the mutual coupling

of lines TL201, TL242 and TL218 and the mutual coupling of lines TL201 and TL217. Figure 2-3 shows the detailed

line model for the lines TL201, TL242 and TL218.

Figure 2-2 Example Line representation

N1

N2

N3

236

229

N4

234

238

SECT

ION

PI

2 C

IRCU

IT1.0

e-3

[ohm

]

TL_201_218_242_1

TL201

TL242

TL_201_218_242_1

1

TL218

TL201

TL242

TL_201_218_242_1

1

TL218

1.0

e-3

[ohm

]

1.0

e-3

[ohm

]T

L_2490_229_1_1

1.0

e-3

[ohm

]

1.0

e-3

[ohm

]

1.0

e-3

[ohm

]

1.0

e-3

[ohm

]

TL201/T

L217

CO

UPLED

PI

SECT

ION

CO

UPLED

PI

SECT

ION

CO

UPLED

PI

SECT

ION

1.0

e-3

[ohm

]

P =

260.6

Q =

12.7

9

V A

P =

166.1

Q =

-33.0

9

V A

P =

113.5

Q =

-59.9

8

V A

P =

-185.2

Q =

-61.5

1

V A

P =

-205.7

Q =

-68.4

5

V A

P =

111.8

Q =

-59.0

1

V A

P =

105.2

Q =

5.4

69

V A

P =

183.8

Q =

55.2

1

V A

==

=T

l218

==

=T

l242

==

=T

l217

P =

-162

Q =

43.2

8

V A

==

=SZ

==

=SZ

==

=SZ




R1335.02.01 9 Nov.26th, 2014


Figure 2-3 Detailed Line Model

2.3 Power Transformers

The power transformer models T5 at Sunny Side are updated based on the following data files received from NLH:

“WAVTS T5 nameplate.pdf”

“WAVTS T5 Test Report.pdf”

“Western Avalon-T5.pdf”

2.4 T5 Transformer at Sunny Side and T1 Transformer at Western Avalon

The T5 transformer at Sunnyside is modeled based on the data provided by NLH. Figure 2-4 shows the PSCAD

model and Table 2-2 shows the data used to model the transformers.

2000.0 [ohm*m]Resistivity:

Analytical Approximation (Wedepohl)Analytical Approximation (Deri-Semlyen)Aerial:

Underground:

Mutual: Analytical Approximation (LUCCA)

20Max. Order per Delay Grp. for Prop. Func.:

20Maximum Order of Fitting for Yc:

1.0E6 [Hz]Curve Fitting End Frequency:

Curve Fitting Starting Frequency: 0.5 [Hz]

Frequency Dependent (Phase) Model Options

Maximum Fitting Error for Yc: 1.0 [%]

1.0 [%]Maximum Fitting Error for Prop. Func.:

Travel Time Interpolation:

100Total Number of Frequency Increments:

On

DisabledDC Correction:

Passivity Checking: Disabled

12.57 [m]

6.7056 [m]

C1 C2 C3

Conductors: Grosbeak

Tower: TL201

0 [m]

G1

5 [m]

Ground_Wires: 1/2"HighStrengthSteel

21.6834 [m]

6.858 [m]

C4 C5 C6

Conductors: 804

Tower: TL242

-35.847 [m]

G1

5 [m]

Ground_Wires: 1/2"HighStrengthSteel 13.53 [m]

5.18 [m]

C7 C8 C9

Conductors: Plover

Tower: TL218

31.27 [m]

G1

5 [m]

Ground_Wires: 1/2"HighStrengthSteel




R1335.02.01 10 Nov.26th, 2014


Figure 2-4 T1 - SSD T5 – WAV

Table 2-2 SSD T5 - WAV

Name Tmva f YD1 YD2 YD3

T5 75 [MVA] 60.0 [Hz] Y Delta Y

Name Lead Xl12 Xl13 Xl23 Ideal

T5 Lags 0.1796 [pu] 0.0563 [pu] 0.1347[pu] Yes

Name NLL CuL Tap V1 V2

T5 0.000533 [pu] 0.00123335[pu] #1 230.0 [kV] 6.9

[kV]

Name V3 Enab Sat Hys

T5 138.0 [kV] Yes Middle Jiles_Atherton

1.0

#1#3

#2

1e-3 [ohm]

1.0e6 [ohm]

1.0e6 [ohm]

N195813

WAV MID5 I_T5n




R1335.02.01 11 Nov.26th, 2014


2.5 PSCAD Model Benchmarking Against PSS/E

After development of the ac network in PSCAD based on the equivalent developed model in PSS/E, it was

necessary to validate the PSCAD case against the equivalent case in PSSE. The validation was performed through

comparison of the short circuit levels at certain buses and the results of the load flow and power transfer at some

selected transmission lines.

The results of the comparison are shown in Table 2-3 and Table 2-4. The results compare very well

Table 2-3 SCL Comparison

Bus

Bus

Number

SCL (MVA)

% difference PSS/E Equivalent PSCAD

Western Avalon 195229 678 702 -3.54%

Sunny side 195222 999 1027 -2.80%

Bay D Espoir 195221 2913 2904 0.31%

Hollyrood 195234 536 570 -6.34%

Stony Brook 195216 1615 1633 -1.11%

Oxen Pond 195238 523 551.44 -5.44%

Table 2-4 Loadflow Comparison

From

Bus

Number

To Bus

Number

Line Real Power (MW) Reactive Power (MVAR)

PSS/E

Equivalent

PSCAD %

difference

PSS/E

Equivalent

PSCAD %

difference

195229 195234 TL217 81.7 84.35 -3.24% -4 -4.53 -13.25%

195236 TL201 84.9 82.95 2.30% -7.1 -6.67 6.06%

195227 TL237 -166.6 -167.6 -0.60% 13.8 13.86 -0.43%

195221 195216 TL204 207 207.45 -0.22% 20.6 20.95 -1.70%

195216 TL231 207 207.45 -0.22% 20.6 20.95 -1.70%

195220 TL234 -103.5 -103.6 -0.10% 3.8 0.5903 84.47%

195222 93.7 94.11 -0.44% -12.7 -12.7 0.00%

In the actual fault, the transformer T5 was energized by closing the breaker L01L03 at Western Avalon and T5

failed 22 seconds later. In the above load flow, the breaker L01L03 (and therefore line TL237) were already

connected. In order to accurately simulate the fault, the line TL237 was opened at Western Avalon prior to the

fault being applied.




R1335.02.01 12 Nov.26th, 2014


3. Modeling of Fault Sequence

The modeling of the fault sequence was based on the AMEC report “Newfoundland and Labrador Hydro

Transmission Availability” March 21, 2014. The supplied load flow was configured to represent the ac system just

prior to the T5 transformer being energized. The detailed sequence of events is as shown Table 3-1 below:

Table 3-1 AC System Fault Sequence

Time event 2

Close Breaker L01L03 and energize T5 via TL237

5.07 Apply phase to phase fault

5.15333 Open ac breaker and trip transformer




R1335.02.01 13 Nov.26th, 2014


4. Results

4.1 Time Domain Simulation

As described in the AMEC report, during the system restoration, one of the first steps taken was to try and

energize the Western Avalon B1 and B3 buses from Come-by-Chance via line TL237 using breaker B1L37, which

failed. During the subsequent incident investigation, it was found that only two phases of the breaker B1L37

closed, which means transformer T5 was energized via two phases. Figure 4-1 shows an example of energizing the

transformer 2 times, with only phases A and B closing. Assuming typical parameters for core characteristics, the

remnant flux density is approximately -0.064 on phase A , -1.335 on phase B and 1.131 on phase C. These flux

densities will be included in the transformer model.

Figure 4-1 PSCAD Simulation of remnant flux

Figure 4-2 shows the enenergization of transformer T5 via line TL237. There is a peak current of about 250A in the

neutral (I_T5n).




R1335.02.01 14 Nov.26th, 2014


Figure 4-2 Energization of Western Avalon transformer T5 via line TL237




R1335.02.01 15 Nov.26th, 2014


Figure 4-3 shows the actual measurements taken at Sunnyside for the T5 tapchanger failure

Figure 4-3 SSD DFR traces for T5 Failure




R1335.02.01 16 Nov.26th, 2014


Figure 4-4 TL207 Values

Figure 4-5 TL203 Values

x 0.00 0.10 0.20 0.30 0.40 0.50 ...

...

...

-200

-150

-100

-50

0

50

100

150

200

kV

V_207:1

-200

-150

-100

-50

0

50

100

150

200

kV

V_207:2

-200

-150

-100

-50

0

50

100

150

200

kV

V_207:3

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

kA

I_207:1

-2.00

-1.50

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

kA

I_207:2

-0.80

-0.60

-0.40

-0.20

0.00

0.20

0.40

0.60

kA

I_207:3

x 0.00 0.10 0.20 0.30 0.40 0.50 ...

...

...

-200

-150

-100

-50

0

50

100

150

200

kV

V_203:A

-200

-150

-100

-50

0

50

100

150

200

kV

V_203:B

-200

-150

-100

-50

0

50

100

150

200

kV

V_203:C

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0.75

1.00

kA

I_203:A

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0.75

1.00

kA

I_203:B

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0.75

1.00

kA

I_203:C




R1335.02.01 17 Nov.26th, 2014


Figure 4-6 T5 Voltages

x 0.00 0.10 0.20 0.30 0.40 0.50 ...

...

...

-200

-150 -100

-50 0

50 100

150 200

kV

V_230_A

-200

-150 -100

-50 0

50 100

150 200

kV

V_230_B

-200

-150 -100

-50 0

50 100

150 200

kV

V_230_C

-0.100 -0.075

-0.050 -0.025

0.000

0.025 0.050

0.075

kV

V_6.9_A

-10.0

-7.5 -5.0

-2.5 0.0

2.5 5.0

7.5 10.0

kV

V_6.9_B

-10.0

-7.5 -5.0

-2.5 0.0

2.5 5.0

7.5 10.0

kV

V_6.9_C

-150

-100

-50

0

50

100

150

kV

V_138_A

-150

-100

-50

0

50

100

150

kV

V_138_B

-150

-100

-50

0

50

100

150

kV

V_138_C




R1335.02.01 18 Nov.26th, 2014


Figure 4-7 T5 Currents and Flux

Referring to the above figures, as the PSCAD model cannot represent a fault on the tap-changer, the fault was

applied externally to the transformer as a phase to phase fault. This did show that 2 KA fault current in a and b

phase windings.




R1335.02.01 19 Nov.26th, 2014


A harmonic analysis was performed on the 230kV ac voltage just prior to the trip and there was approximately a

2.5% 2nd harmonic voltage distortion present, which is due to the energizing of the transformer. No other major

harmonics were present.

Figure 4-8 FFT of 230kV ac voltage




R1335.02.01 20 Nov.26th, 2014


4.2 Harmonic Impedance Scan

In order to determine if there was a potential for a harmonic impedance resonance, a harmonic impedance scan

was performed at three intervals:

1. At the Western Avalon bus just after connecting line TL237

2. At the Western Avalon bus just after connecting line TL237 and T5 transformer

3. Impedance as seen from the 138kV side of the T5 transformer looking towards the Western Avalon 230kV

bus.

Figure 4-9 Harmonic Impedances as seen from WAV bus

Referring to figure 4-9, there is negligible differences in the impedance when both line TL237 and T5 are connected

compared to when just line TL237 is connected. For these cases, there are some resonant cases around 189Hz and

383Hz, but are away from the areas of concern. If we look at the impedance through the T5 unit, we see there are

no resonances for this case.

x 0.0 0.3k 0.5k 0.8k 1.0k 1.3k 1.5k 1.8k 2.0k ...

...

...

0

100

200

300

400

500

600

700

800

ohm

s

|Z+|(ohms) TL237 Closed T5 open |Z+|(ohms) TL237 Closed T5 Closed |Z+|(ohms) TL237 Closed T5 Closed (Through T5)

-80

-60

-40

-20

0

20

40

60

80

100

deg

PHASE(Z+)(Deg) TL237 Closed T5 open PHASE(Z+)(Deg) TL237 Closed T5 Closed PHASE(Z+) TL237 Closed T5 Closed (Through T5)




R1335.02.01 21 Nov.26th, 2014


5. Conclusions and Recommendations

Although the simulation did not reproduce the exact fault, a comparable one was simulated. The simulations do

not point to an exact failure cause, but the simulations have shown that energizing 2 phases on the transformer

numerous times before energizing all three phases does produce some remnant flux and an unbalanced loading on

the transformer that could have contributed to the failure.

TGS recommended that once more details are available about the actual fault of the transformer, that the model

be updated and simulations rerun to recreate what happened. This new information will give further confidence in

the developed model and allow for future studies to be done either as a system strengthening exercise or for a

post-mortem on future faults.


p r 11 hydro p.o. box 12400. st. john's. ni hydro place ... · newfoundland and labrador hydro...

Documents