You are on page 1of 100

# Power System Studies Best Practices & International Experiences

## Speaker : Dr.Dharshana Muthumuni, PhD, P.Eng Manitoba HVDC Research Centre

1

There are a variety of simulation tools used in power system studies to predict system behaviour. - Load flow - Transient stability - Fault studies - Harmonic analysis - Electromagnetic transient simulations. The fundamental difference between an electromagnetic transient (EMT) type simulation and other types of simulations is in the way the electrical circuit response is mathematically represented. - EMT programs represents the power system response by the respective differential equations. - Other programs represent the power system response by reducing the differential equations to the respective phasor form.

## Power System Studies Best Practices & International Experiences

A typical EMT simulation results showing the voltage across the tree phases of a circuit breaker is shown below. This is an illustration of the well known breaker Transient recovery Voltage (TRV). Note the high frequency compnents on the voltage immediately after the current is interrupted. Such details analysis can be done only on EMT platforms.

## Load Flow / Transient Stability

Each solution based on phasor calculations

Electro-Magnetic Transients
Direct time domain solution of Differential Equations

V ( ) = R I ( ) + j ( L ) I ( )

v(t ) = R i (t ) + L

d i (t ) dt

## Power System Studies Best Practices & International Experiences

In addition to solving the complete set of differential equations, the EMT solution allows to include other power system details such as - Non linear effects (iron core saturation for example) - Frequency dependent effects (skin effect in conductors). - Response of associated mechanical systems. - Mechanical equations of rotational inertias of a turbine generator system - Rotational dynamics of wind turbines - Inclusion of the response of complex control systems associated with the power system equipment - Generators controls - Complex controls functions of HVDC schemes - Complex control functions of wind integrations
5

## Power System Studies Best Practices & International Experiences

All the above features make EMT the most accurate and detailed simulation method available to power system engineers.
- It should be noted that such level of details is not required for every analysis. As an example, the load flow solution is essentially a 50 Hz, steady state situation. The power system elements can be represented by the corresponding phasor equations to obtain the correct solution.

## Power System Studies Best Practices & International Experiences

RRL BRK

Another example of a fast (involves high frequency oscillations) electromagnetic transient event is capacitor switching. Note the very short (relatively) transient period. The short duration over voltage and current could be influenced by one of more of the following: Natural response of the system LRC elements (those near the cap. Bank has the largest impact) Travelling waves in lines/cables near the cap. Bank. Non linear effects such as transformer saturation.

BRK

## 0.1 [H] 1.0 [uF]

C apbank : Graphs 300 200 100 kV 0 -100 -200 -300 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 -2.00 0.100 R LIa R LIb R LIc Vcap

kA

0.150

0.200

0.250

0.300

0.350

0.400

0.450

0.500

## Why Do We Study the short Transient period?

Mainly to protect major equipment from insulation failure Ensure that main station equipment are protected from lightning induced voltage surges Ensure satisfactory operation of circuit breakers Design surge arresters Design current/voltage limiting devices (inrush reactors, grading capacitors) Identify the worst case (magnitude and duration of the transient) Identify and design mitigation methods

## Power System Studies Best Practices & International Experiences

EMT programs are by no means limited to studding the fast transients associated with switching As described previously, the EMT program represents the power system equipment in accurate detail. Also, we mentioned that complex control functions can be incorporated into the simulation. This is a key factor in HVDC and FACTS simulations. The response of these devices are tightly coupled to how they are controlled (and protected). As such, EMT simulations are readily applicable to the design and integration of HVDC, FACTS and other power electronic based power system devices. In fact, such simulations play a very important role in the design and integration process.

## Start-up of an induction motor

Slow Transient (electro-mechanical)

Main : Graphs

## Motor starting - With a mehanical load connected to the shaft.

pu

1.0000 0.9960 0.9920 0.9880 0.9840 4.0 2.0 0.0 -2.0 -4.0 0.20

Vrms

BRK 0.0 0 W S TL Tm IM

## Breaker is closed at 0.5s to start the motor.

Ea

Ia BRK Ea

RRL

#1

#2

A V

kV

Ia

RL

kA

0.00 -0.20

2 X

* 0.8

Tm

W
Mechanical load model: Load torque is modeled as being proportiobnal to w^2.

1.00 0.80 0.60 0.40 0.20 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2.20 ... ... ...

pu

## Power System Studies Best Practices & International Experiences

10

EMT programs and their importance in HVDC and other Power electronic based systems.
Design of complex control strategies to meet the system performance objectives. Design protection functions associated with the HVDC/FACTS controls to limit over voltages and other undesirable conditions. Design equipment insulation ratings Study HVDC/FACTS control interactions with other control systems (eg. Excitation control of a nearby thermal generation) and to find solutions to such issues. Sub synchronous torsional interactions involving nearby thermal generators that could damage generator shafts. Other types of sub synchronous (below 60 Hz) resonance issues (becoming a series concern with power electronic controls of wind systems)

## Power System Studies Best Practices & International Experiences

11

EMT programs and their importance in HVDC and other Power electronic based systems.
Wind farm fault ride through studies where the fault response has to consider the ratings as well as the protection and control aspects of the wind turbine power electronic converters (Type 3 and Type 4). Research and development related - Study new concepts - Multi terminal HVDC schemes Others..

## Power System Studies Best Practices & International Experiences

12

PSCAD implementation of a conventional HVDC inverter valves, the converter transformers and the A side filters.
DCP A V
Ibus

DCIC

AMID GMID

AM GM

Com . Bus

0.0606 [H] #2 AO KB 6 Puls e Bridge #1 167.2 [uF] 15.04 [uF] 116.38 [ohm ] 37.03 [ohm ] 15.04 [uF] 13.23 [ohm ]

A O I

0.0061 [H]

## 7.522 [uF] Com . Bus

AMIS GMIS

AM GM

#2

#1

1 0 0 0 0 0 0 .0[o h m ]

## Power System Studies Best Practices & International Experiences

13

PSCAD implementation of a conventional HVDC inverter controls for thyristor firing pulses.
INVERTER CONTROLS
input current order (pu)
CO

Inverter
current order for rectifier

## INVERTER CURRENT CONTROLS

DC Voltage (compensated) G 1 + sT VDCL + F MPVS

1.50 1.00

AC Volts (RMS)

AC Voltage

VDCI

* 0 .0 1

POWER Power F

CMARG

0.1

(pu)
P I BETAIC

## dc voltage measured at inverter

voltage compounding

0.50
Voltage Dep. Current Limit Rectifier Current Order

## 0.00 -0.50 -1.00 -1.50

F CERRI D + CERRIM

CMIC

G 1 + sT

## dc current measured at inverter Angle Order for inverter

B + D Alpha Order

Pi

(pu)

DC Volts

DC Current

D Max

BETAI

AOI

## DGEI Delta Gamma Error

F GMES Min in 1 Cycle Gamma GMESS D + + B -0.544 GERRI P D Max GNLG I E BETAIG

0.50 0.00 -0.50 -1.00 x 0.250 0.300 0.350 0.400 0.450 0.500 0.550 0.600

0.2618

GMIN

14

## Selected Applications in Power Systems

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Transformer energizing Transmission line and equipment switching Capacitor bank switching Circuit breaker Transient Recovery Voltage (TRV) Overvoltages caused by lightning strikes Motor starting Protection Equipment failure Post event investigations Network resonance and Ferro resonance problems Distributed generation studies Power quality

Many more..

## Power System Studies Best Practices & International Experiences

15

Transformer Energizing
7000 6000 5000 4000 3000 2000 1000 0

380 Hz

5.0 4.0

Ia(kA)

|Z+|(ohms)

240 Hz

x

Frequency [Hz]

- Voltage dips

Ia
1.0

## Network characteristics frequency scans

- Over voltages due to harmonic resonance conditions
0.0

Ia (kA)

[6] 0.00628253

16

## Transmission line and Equipment Switching

E1 Minimum: Maximum: Mean: Std Dev: 2% Level: 98% Level: 364.6304 395.8886 379.7837 7.564519 364.248 395.3193 E2 280.1849 352.6374 320.0234 17.03806 285.0315 355.0153 E3 332.9529 413.6732 375.2662 18.64495 336.9741 413.5582 E4 353.4259 383.773 369.2719 7.544283 353.7778 384.7659

## Overvoltage magnitudes and equipment insulation levels

Surge arresters

Statistical distribution of overvoltage magnitude Transmission line flash-over rates Investigation of overvoltage mitigation methods
Power System Studies Best Practices & International Experiences 17

## Back to back switching

Resonance concerns

- Inrush reactors

Outrush reactors

## Capacitor Bank Switching

Cable energizing

Faults

2 40 k 2 60 A = 21 . 326 6 s 10

Inrush Reactor

B4 0.000277 [H]

3.76 [uF]

Inrush Reactor

B3

0.000277 [H]

3.76 [uF]

Inrush Reactor

B2

0.000277 [H]

3.76 [uF]

Inrush Reactor

B1

0.000277 [H]

3.76 [uF]

= di dt
allow

18

## Circuit breaker Transient Recovery Voltage (TRV)

TRV is the voltage developed across the breaker poles immediately after current interruption
- Fast event - Simulation circuit should consider details of station equipment - Breaker TRV withstand capability limits
TRV 300 250 200 150 100
kV
LEGEND: DS DS1 BRK BRK1
BAY A b1 50MVAr Cap.Bank

sysTRV
Station X2 Station W Station Y2

## Station Z2 CAP BANK 1 b1 CAP BANK 2 7500 pF CCVT 50MVAr Cap.Bank

a b1 b1

a b1

a b1

a b1

b1

b1

BAY B

BAY C

BAY D

BAY E

BAY F

REACTOR 1

50 0 -50 -100
a1 a1 b b a1 b a1 b a1 b a1 b 7500 pF CCVT

## 50MVAr Reactor 50MVAr Cap.Bank

CAP BANK 3

b1

b1

b1

-150 time(s) 0.2450 0.2500 0.2550 0.2600 0.2650 0.2700 0.2750 0.2800 0.2850

REACTOR 2

REACTOR 3

50MVAr Reactor

## Power System Studies Best Practices & International Experiences

19

Lightning Overvoltages

Lightning overvoltage studies are required to: Determine the required insulation levels of equipment (BIL).
Surge arresters size and location Determine transmission line flashover rates - Very Fast event - Simulation circuit should consider details of station equipment
Lightning overvoltage waveforms 60 Lightning Current

40

## How do we represent system equipment

Vstring_1

20

0 1.0k 0.8k 0.5k 0.3k 0.0 x 0.015m 0.020m 0.025m 0.030m 0.035m 0.040m 0.045m

## Power System Studies Best Practices & International Experiences

20

Motor Starting
Voltage dips and flicker caused by frequent starting of large motors at industrial plants is a power quality concern for utilities - Model data to match manufactures T-S and I-S curves - Impact of rotating inertia - Power System impedance characteristics near the interconnection point - Representation of load characteristics and the overall impact.
M otor starting transients 1.010 1.000 0.990 0.980 0.970 0.960 0.950 0.940 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 5.0 4.0 3.0 2.0 1.0 0.0 -1.0 1.00 0.80 0.60 0.40 0.20 0.00 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 Vrm s

pu

I a

kA

pu

W s

pu
tim e(s)

21

## Few Interesting (recent) applications:

.

1) 2) 3) 4)

Transformer Bushing failure in a power plant. System black start restoration studies Fault current limiting devices Protection

22

## Transformer Bushing failure in a sub station.

Gas insulated bus-bars connect the GIS to the generator transformers. Number of transformers failed over a period of about 2 years. Perform apparatus inspections and perform engineering studies (simulations) to investigate the causes for the 380 kV transformer bushing failures.

23

## Transformer Bushing failure in a Station.

a1 80 [pF] PT E A R T H I N G S W I T C H 25 [pF] 25 [pF] 25 [pF]

D3R_

D3Y_

## CB1R_ CB1Y_ CB1B_

Electrical studies: Switching, and temporary over voltages Ferro resonance Harmonic impacts GIS Very Fast Transients.

D I S C O N N E C T O R E A R T H I N G S W I T C H B R E A K E R E A R T H I N G S W I T C H

D1R_

D1Y_

D1B_

2 . 4[ p F ]

2 . 4[ p F ] b1

2 . 4[ p F ] b2

24

## Transformer Bushing failure in a Station.

Switching transients
1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 Tim e ... 0.000 Esov - PP9 Bushing

PU

0.010

0.020

0.030

0.040

0.050

0.060

0.070

0.080

## GIS switching related VFT - MHz range

4.0 3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -4.0 E_ST3_DER
3.00 2.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00 E_GT1_2

25

## Transformer Related Issues

Ferro Resonance
Many factors can lead to ferro-resonance situations. Transients simulations are necessary to identify possible problems.
F e r r o r e s o n a n c e C a s e S tu d y 1 .2 5 1 .0 0 0 .7 5 0 .5 0 0 .2 5 0 .0 0 - 0 .2 5 - 0 .5 0 - 0 .7 5 - 1 .0 0 - 1 .2 5 5 .0 4 .0 3 .0 2 .0 1 .0 0 .0 - 1 .0 - 2 .0 - 3 .0 - 4 .0 0 .0 0 V busA V busB V busC

V P r iB

Voltage

0 .1 0

0 .2 0

0 .3 0

0 .4 0

0 .5 0

0 .6 0

0 .7 0

0 .8 0

0 .9 0

## Power System Studies Best Practices & International Experiences

26

CT Saturation
Mal-operation of an earth fault relay during transformer energising.
Is

Inrush current caused unequal saturation of the 3 CTs, resulting in a burden current.
Main,CT1 : Graphs 12.0 Ib (A) Ib
PI SECTION

## -2.0 2.00 B (T) B1

Timed Fault Logic Ea
ct_3_new _.f

## -0.25 120 Ia (Amps) Ia

Is #1 #2

-20 0.00 0.20 0.40 0.60 0.80 1.00 ... ... ...

27

RL RRL Ib

## M ain : Graphs 600 TR V

400
T _l i ne

200
Iflt Tim ed Fault Logic V a

ABC

TRV B R E A 1

0
30 [pF]

3 0[p F ]

I_ C B 1

## Tim ed Breaker Logic Clos ed@t0

3 0[p F ]

BREA1

-200

V b

-400
+ C_LIM 1 3 3 .5[p F ] 2 0 0 0 0 0 .1 0 8 650 [pF] Lim iting Reactor

M ain : Graphs

0.220

0.230

0.240

30 [pF]

TRV_env_A30

TRV_env_A30

TRV

TRV_env_A60

TRV_env_A60

TRV_env_A100

TRV_env_A100

3 0[p F ]

B R E A 2

## 3 0[p F ] DS DS1 50 [pF] DS DS2 50 [pF] DS DS3 50 [pF] 50 [pF] SA

-400 -600 -800 800 600 400 200 0 -200 -400 -600 -800 800 TRV_env_C30 TRV_env_C30 TRV TRV_env_C60 TRV_env_C60 TRV_env_C100 TRV_env_C100 TRV_env_B30 TRV_env_B30 TRV TRV_env_B60 TRV_env_B60 TRV_env_B100 TRV_env_B100

16000 [pF]

## 600 400 200 0

# 1 # 2 1 2 0 0 0[p F ]

250 [nF]

-200
Sub Trans ient Reactance

0 .0 0 0 8 5

R L

## Power System Studies Best Practices & International Experiences

3 .5 9 9 e -4

-400 -600 -800 x 0.1980 0.2000 0.2020 0.2040 0.2060 0.2080 0.2100 0.2120 ... ... ...

28

## Fault current limiting devices

PSCAD implementation of a Super Conducting Fault Current Limiter
9 4 10 5 11
DC circuit

8 3

DC2

DC1

A1

9
2.5

10

11
2.046 [mH]

A2

B1 B2

1.5 B [T]

C1
1

C2

0.5
B USED

P+jQ

29

## System black start restoration studies

Getting a power system back into operation after black-out- is not an easy task and should be carefully planned (Black start restoration plans) Some issues: -Can we energize lines , transformers and loads through available generating units? - System has very little damping due to low load. Resonance issues.

30

T Line_01_01

## 26.326 km Eng #1 #2 P+jQ #1 #2 P+jQ

QILWAH

MIKHWAH

Bil Jurs hi E1 E3

## Timed Breaker Logic Open@t0

38 km BRKG 53.618 km
TLine_01_02 TLi ne_01_03

TLine_01_a

#1 #1 #2

#2
T Li ne_01_b

#1

#2 P+jQ

Restoration steps are determined and documented step by step. System single line drawings are used to illustrate each step

## P+jQ P+jQ BRKG E2

T Li ne_01_04 T Line_02_01 T Li ne_02_02 TLine_01_06 TLine_01_07 T Line_02_03 row1 TLi ne_01_05 row1

Iqunt

NAMERA

QUNFUDAH TOWN

Iqunt

Etcps 67 km E1 E2 9 km 95 km 36 km

E3

## QUNFUDAH CPS Etcps TCPS

THORYBAN

49 km
TLine_01_08 TLine_01_09 TLine_02_05 T Line_02_05_2

86 km
T Line_02_04

#2 #2 #1 SUFFAH P+jQ

#1 47 km
T Li ne_01_10

Electrical studies are necessary to verify that the selected restoration actions (steps) can be implemented without damaging equipment.

P+jQ

SABTAL JARAJ

68 km
TLi ne_01_12 TLine_01_11 T Line_02_06

72 km

#1

#2 P+jQ

#2 P+jQ

#1

AL BIRK

T Li ne_01_13

TLine_02_07

MAJARDAH 94 km

100 km 82 km
T Li ne_01_15 TLi ne_01_14

TLine_02_08

#2 P+jQ

#1

MUHAIL NORTH

T Li ne_01_17

67 km

44 km

T Line_01_16

T Li ne_02_09

#2 P+jQ

#1

MUHAIL

31

## System black start restoration studies

Transform er E nergizing 1.50 1.00 E _49900

0.50

PU

## 0.00 -0.50 -1.00 -1.50 2.00 1.50 1.00 0.50 Itf

kA
x

0.00 -0.50 -1.00 -1.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00 ... ... ...

PU

## 0.00 -0.50 -1.00 -1.50 2.00 1.50 1.00 0.50 Itf

kA
x

0.00 -0.50 -1.00 -1.50 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 ... ... ...

32

## System black start restoration studies

1.20 1.00 0.80
PU

Vrms_machine

Long line energizing through small generating units Self excitation issues
S/H in out hold S2M Vref0 Vref Vs PSS2B P1 Pw W Exciter (ST4B) VT Ef0 IT 3 Ef If Ef0

0.60 0.40 0.20 0.00 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 1.50 E sov - PP9 Bushing E a- M achine terminal

3 Ef0 Ef If S Ea Tm w Tm Tm0 TM0 W w Cv Cv w Tm1 B + + F TIME L2N TIME S2M Timed Breaker Logic Closed@t0 Timed Breaker Logic Open@t0 A V #1 #2 BRKG Esov Ia2 A V Ia1 BRKL
TL_SUD_P...

Sudair Bus
Ib1

PU

Te

1.00 0.50
PU

## Steam_Tur_1 Wref Tm2

0.99982 x 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 ... ... ...

33

## Power System Studies Best Practices & International Experiences

34

CT Saturation
Apply faults at different inception angles and observe the initial DC exponential In the fault current.
200 km long 230 kV transmission line simulated in two segments to facilitate application of faults at different point on the line.
P + jQ

RRL

Iabc Vabc

T TLine1

T TLine2

## 75 km long second 230 kV transmission line

Timed Fault Logic Fault at 0.4876s and 0.49167s
T TLine3

RRL

A->G

## Power System Studies Best Practices & International Experiences

35

CT Saturation
Discuss the CT model. The dc exponential in the fault current will force the CT core flux to move in one direction. This can lead to saturation and a resultant loss of secondary current to the relay. CT secondary burden influences the saturation of the CT core.

## Power System Studies Best Practices & International Experiences

36

CT Saturation
Main : Graphs 200 Vabc

Loss (reduction) of secondary current due to CT core saturation. Saturation is a result of the dc component in the fault current.

V -200 15.0 I sec Ia_sec -30.0 3.0 Iabc I pri -5.0 0.25 Flux den B -2.00 0.400 0.450 0.500 0.550 0.600 0.650 0.700 ... ... ...

## Power System Studies Best Practices & International Experiences

37

CT Saturation
CT of phase A saturated during energising of a single phase transformer in a distribution feeder.
Is BRK1 COUPLED PI SECTION

ct_3_new _.f

Ea

1 2 3

## Ia BRK2 #1 0.01 [ohm]

Is

-20 0.00 0.20 0.40 0.60 0.80 1.00 ... ... ...

#2

38

## Series compensated line

60.0 V freq RRL A V F Ph P = -103.6 Q = 541.8 Vcua P_CUA Q_CUA P = 269.7 Q = 323 A V Ph freq F V RRL RL freq RL

Station A

Vcua Station E

L1C

L2C ITYL1

System model
T T T PAL2 PAL3 PAL1 T TYL1_a T TYL1_b T YP

L1T

L2T

Bypass breakers Arresters Bypass breaker logic based on fault current level and arrester (parallel with capacitor) energy

ITYL1

T TYL2_a

T TYL2_b T YA

Ph

T TYL3_a

T TYL3_b

BT3

RL

freq

freq

freq

Station C Ph

BT3

BY3

RRL RL

A->G RRL RL

Ph F V

39

## Discuss protection complexities.

300 E 1 E 1 -300 400 E 2 E 2

## -300 25.0 ITY L1 ITYL1

Complex voltage and current waveforms the relays must deal with during a faults close to the capacitors

## -20.0 1.00 B a1 signal Ba1

0.0 1.150 1.200 1.250 1.300 1.350 1.400 1.450 ... ... ...

40

## Fault current contribution of machines

Detailed emt type simulation is the only way to accurately models the nature of fault current waveforms of different machines. With the increased popularity of Induction generator based wind farms, the accurate estimate of fault contribution of complex induction machine generation concepts has become very important. e.g. Fault contribution of a DFIG is complex to analyse on simple fault analysis programs.

41

## Fault current contribution of machines

Induction Machine 80 60 40 20 0 -20 -40 -60 2.40 Isc 2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00 4.20 ... ... ... I1

Induction Machine - Current decays to zero as the stored energy in windings is dissipated.
Main : Graphs

Fault current

## 50 40 30 20 y 10 0 -10 -20 0.0

Synchronous Machine -Sustained fault current - Affect of dampers and the field winding transients.
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 ... ... ...

42

## Realistic waveforms for Relay testing

200 km long 230 kV transmission line simulated in two segments to facilitate application of faults at different point on the line.
P+jQ

RRL V1

I1

T TLine1

T TLine2

T TLine3

RRL

## Type2 Timed Fault Logic Type3 Timed Fault Logic

Generate and record waveforms under different conditions batch mode simulation Different load conditions (not included in the example case relay_test.psc) Point on wave Location Fault type
Power System Studies Best Practices & International Experiences 43

Some other instances where EMT type simulations maybe necessary in relay studies
Unbalanced operation of the system and the effect on relay settings (e.g. Unequal units in a three phase transformer bank, open conductor)
Typical fault analysis programs are not able to handle such events.

Harmonics due to non linear loads and the effect on relays. Integration of new technology
Distributed generation Islanding protection Complex wind farm protection requirements

## Power System Studies Best Practices & International Experiences

44

Some other instances where EMT type simulations maybe necessary in relay studies
Investigate new relay concepts Complex networks Mechanical resonance issues (SSR)

## Power System Studies Best Practices & International Experiences

45

Voltage Flicker Due To Arc Furnace and Motor Loads and Mitigation Strategies

46

## Arc Furnace Model of PSCAD

The developed EAF model is based on the nonlinear differential equations as outlined in [1], which models the non-linear characteristics of the electric arc as pictured below. The equations representing the arc voltage (v) to arc current (i) are shown below, where r is the arc radius:

k r
1

dr 2 3 + k2r = k m+2 i dt r

47

## Modeling and Simulation of Harmonic and Flicker

Harmonic and Flicker studies involving arc furnace installations is challenging
- The power system feeding the industrial plant (with arc furnace) must be modeled in adequate detail in an EMT platform. - Correct power flows and fault levels - Detailed representation of lines, cables, transformers and other equipment - Model FACTS devices in the vicinity

48

## Network Model Developed For a Recent Study

OPICO ETESAL RRL 1 PI Section RL

MHSAJAPA BRK46_1 3 #1 #2 MHSAJAPA 11 PI Section 20 PI Section 5 10 PI Section PI Section 7 PI Section 8 SITIO NINO 23 - 46kV #1 #2 5 BRK46_2

4 PI Section

PI Section

## Arc furnace facility

6 PI Section 15 14 PI Section

BRK Eln

EAF Facility

## QUEZALTEPEQUE 46kV #1 #2 CARGA HORNO

Network model developed in PSCAD for a recent eaf flicker mitigation study performed by MHRC engineers.

PI Section

10

PI Section

RRL

RL

PI Section

49

## Study Model Must Be Validated Prior To Detailed Investigations

2

MHSAJAPA BRK46_1 3 #1 #2 MHSAJAPA 11 P I S e c ti o n 20 P I S e c ti o n 12 P I S e c ti o n 14 20 SITIO NINO 23 - 23kV 5 10 PI Section P I S e c ti o n 7 P I S e c ti o n 8 SITIO NINO 23 - 46kV #1 #2 5 BRK46_2

The arc furnace absorbs about 25MW and 8.5MVAr from the grid. The voltage at bus 5 drops to 0.98p.u. (45.08kV).
PI Section

P I S e c ti o n 3

RL

PI Section

## Arc furnace facility

6 PI Section 15 14 PI Section

BRK Eln

EAF Facili ty

## QUEZALTEPEQUE 46kV #1 #2 CARGA HORNO

The arc furnace also causes flicker with an instantaneous flicker level (IFL) of about 16.6 at the POI. The simulated voltage waveform at bus this matches field recordings and measured flicker levels. This validates the study model.
40 30 Vbus 20 10

PI Section

12 QUEZALTEPEQUE 23kV 17 18 P I S e c ti o n 9

10

PI Section

RRL

RL

kV

P I S e c ti o n

50

## Arc Furnace and Filters Modeled in PSCAD

Carga Carga Horno Horno 25 46 [kV] / 0.58 [kV] A V #1 0.0348 13.14 Ohm s - 0.0348 H #2 Varc 1 . 0 e 9[o h m ] Ia Ext. 1 . 0 Ig Ib Ic

4HP

3P

1 1 0 . 5[ m H ]

1 4 8 . 7[ m H ]

## Power System Studies Best Practices & International Experiences

1 4 8 8

The electrical network of the industrial plant must also be modeled in adequate detail - EAF characteristics are represented by the detailed model available in PSCAD - Filters - SVC and FACTS devices if present - Transformers, lines, cables, other loads - Motor loads with appropriate load characteristic representation to capture their contribution to voltage fluctuations

4 .1 2

5 .4 4

51

Model Validation
Model validation and setting up the arc furnace model to replicate the field measurements in terms of the follows is very important before mitigation methods can be studied - Flicker levels should closely follow field measurements - Power and reactive power should closely follow field measurements - Voltage profiles should closely follow field measurements Achieving the above is challenging and requires special expertise. This is due to the highly non linear and sporadic nature of the eaf load. - It is important to note that it is impossible to get a simulation case to match the observed results perfectly due to the nature of the problem. The important thing is to capture the essential features and the trends of a practical arc furnace.

52

## STATCOM as a Possible Mitigation Option

igrid iPCS

ifurnace

There are a number of possible solution to the harmonic and flicker problems encountered at industrial plants with arc furnace loads. - Passive filters - SVC - STATCOM - Equipment upgrades

STATCOM with special auxiliary control loops targeting flicker mitigation is a viable mitigation method. This principle was used in a recent study for the arc furnace flicker mitigation performed by MHRC.

53

## STATCOM as a Possible Mitigation Option

igrid

iPCS

ifurnace

Simulation studies are necessary to develop the special control techniques that should be introduced in addition to standard FACTS controls. Simulation studies can determine the ratings of equipment required to lower the flicker levels to acceptable range The equipment ratings depend on (among other factors) - Existing flicker levels - The target flicker level (need only be lowered to acceptable levels as specified in relevant standards).

54

## STATCOM as a possible Mitigation Option

DCP

Q GG1 2

Q GG3 2

Q GG5 2

D 6000

P = -0.5274 P = -0.5281 2 [MVA] Q = 0.9126 Q = 0.9862 V = 45.86 46 [kV] / 0.48 [kV] V = 0.4574 + SrcLV Is_LV A A #1 #2 ABC V ELV V Lf F = 4 5 0 [H z ]

V_dc

6000

Q GG4 2

Q GG6 2

Q GG2 2

DCM

55

## Typical STATCOM Control Functions

iq B iqref D + In id B PI Antiwind-up Rst Rst A MinVq + F wL id * wLf iq vq1 Tid1 wL * wLf Kpd1 Kpd Tid vd D + F Out idref D + In PI Antiwind-up Rs t Rs t A vd1 vd1 Out 0.395 MaxVd [kV] -0.395 MinVd [kV] MaxVd MinVd MaxVq

## Kpq Kpq1 Tiq1 Tiq

Modulation references

Refa3 B D + Refa

## Edc X vd1 vq1 Y Y P

* 0.5 M X P phase M N

## D N/D M MagInv PhaseInv * 57.2958 P X Y P X Y 0.0 M D Q 0 A B C

+ B Refa3

Refb

+ B Refa3

Refc

MagInv * 3

* 0.16

M Magh3 P Phaseh3 Y P

X M X 0.0 Y

D Q 0

A B C

PhaseInv

Theta

* 3

56

## Special STATCOM Control Functions to Address Voltage Flicker

id B D 0 .0 + B idref Err_d * Ifurd_err vd wL iq * 377 * L
iq B D + Ifurq_err * B Err_q vq D Vref P + B VmagPU Vpcc I id * -1 * 377 wL * L D + F vq1 vq1 iqref D + I P q PI out

D + I

P d PI out

id Regulator
A D ++ vd1 vd1

D Q 0

id id iq iq

## A B Ifurb C Ifurc Ifura

D Ifurd Q 0 Ifurq

X1

## id furnace Ifurd iq furnace

X2 Ifurq

Mag2 D (7) + Ph1 Ifurd (7) B Ifurd_dc Ph2 D (7) Ifurq + dc2 B Ifurq_dc Ifurq_dc

Ifurd_err

F = 10 [Hz] dc1

Ifurq_err

Ifurd_dc

iq regulator

## Power System Studies Best Practices & International Experiences

57

Simulation Results

E AF_Facility 40 30 20 10 0 -10 -20 -30 -40 1.0000 0.9950 0.9900 0.9850 0.9800 0.9750 0.9700 0.9650 0.9600 tim e (s) Vbus_rm s Vbus

Vbus

kV

kV

3.900

3.950

4.000

4.050

4.100

kV

58

59

## Flicker Due to Cyclic Motor Loads

Variable motor (cyclic) loads such as those of large compressors can give rise to flicker problems at weak interconnection points. MHRC has provided FACTS based solutions (SVC etc) through emt simulation studies.

60

## Bus 1 Bus 2 P = 24.65 Q = -10.58 V = 118.1 A V

#1

#2

T TLine_02

T TLine_01

Ia 45 km line

#1

#2

Compressor Motors

RRL

100 km line

## 345 kv 'Weak system

RL

Ea

System Model and the Synchronous motors driving the compressor load.

61

## Flicker Due To Cyclic Motor Loads

pu Tm1 Torque1 1 sT Clear 1 Average Torque = 0.82 pu (approx) 6 units are pulsating as one Zero Detector F 360.0 + D Degrees Torque1.dat * -1.0 6 - 6000 HP Compressor Motors 327.27273 rpm - 22 Pole machine G 1 + sT

## The compressor characteristics are stored in a data file.

Actual load torque characteristics (torque vs rotor angular position) is stored in a look- up table and fed into the study model to get realistic simulation waveforms, replicating flicker measured in the field.

62

## Flicker Due To Cyclic Motor Loads

M ain : Graphs 98.0 97.0 96.0 95.0 94.0 93.0 92.0 91.0 90.0 0.30 0.20 0.10 0.00 -0.10 -0.20 -0.30 120.0 119.0 118.0 117.0 116.0 115.0 114.0 113.0 30 20 10 0 -10 -20 -30 6.50 E a

Ia

Simulation results compared closely with field measurements, thus validating the study model for investigating mitigation options. After investigating a number of options (system and equipment upgrades), a SVC based solution was recommended and eventually implemented.
Q1

Vrm s

P 1

7.00

7.50

8.00

8.50

9.00

9.50

10.00

63

## Power System Studies Best Practices & International Experiences

64

The single line diagram show a transmission system with significant wind interconnection. The network is characterized by a number of long over-head 345 kV lines. The system is relatively weak.
Gray Whitedeer 64 km 176 km Tesla a1 147.04 km Whitedeer
CTT _WD_... CT T _GR_...

TLine_11

4.81 7.40 a2

GEN
TLine_8

A V

Hereford

68 km
CT T_ST_T...

Series Caps b a

68 km
CT T_ST _T...

Bus 1
52.8 km
T Line_12

121.6 km 76.8 km
Sh_WD_ST

Rieley 102.4 km
ET T_T S_...

T Line_9

## P = 805.3 Q = -366.3 V = 0.9888

67.2 km a1 Silverton A V a2 a3

ETT _T S_...

Thermal Gen

99101 NAZARETH

7.42 b 70.2 km
ET T_RL_...

Ea 250

## a SVC Edith Clarke 100.8 km

Sh_ST _CW

7.41

121.2 km
ET T_EC_...

SVC
Cottonwood P = 798.5 Q = -369.3 V = 1.055 RRL A V 3.16 [H]

## Power System Studies Best Practices & International Experiences

65

Response of the system following a fault (cleared in normal time) is shown below for the two scenarios that were studied. In the strong connection case, the system seems better damped than in the weak connection case. The SVC keeps oscillating between its minimum and maximum Q limits. This is unacceptable control performance. Such issues should be identified at the design stage. Control parameters may be re- tuned to mitigate the issues.
SVC_Page : Graphs 1.20 Vpu

1.00

1.00
0.80

0.80

y (pu)

y (pu)

0.60

0.40

0.20

## Weak system response

y (MVAR)

50 0 -50 -100 -150 -200 x 0.0 2.0 4.0 6.0 8.0 10.0 12.0

y (MVAR)

## Power System Studies Best Practices & International Experiences

66

Distributed Generation Applications Control interaction issues E - Control and protection - Energy storage - Power quality

67

68

## Power System Studies Best Practices & International Experiences

69

Introduction
There are a number of different HVDC converter technology that is in use world wide Line commutated (or conventional) HVDC (LCC)
Based on Thyristors

## Capacitor commutated converter based HVDC (CCC)

Based on Thyristors

## Voltage source converters (VSC)

Based on IGBTs

LCC is the most mature as well as the widely used technology for large scale HVDC power transmission.

## Power System Studies Best Practices & International Experiences

70

Introduction

A typical LCC - HVDC Scheme Power flow is from the rectifier to the Inverter side
0.5968 [H] 2.5 [ohm] 2.5 [ohm] 0.5968 [H]
R bus Ibus

Inverter_AC

Rectifier_AC

26.0 [uF]

A V

A V

Rectifier

Inverter

## Rectifier 2.00 1.50

V o lt a g e (p u )

Rectifier

DC Volts

DC Current

## 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50

AC Voltage

0.50 0.00 -0.50 -1.00 -1.50 -2.00 1.660 1.680 1.700 1.720 1.740 1.760 1.780 1.800 ... ... ...

(p u )

1.00 0.50 0.00 1.660 1.680 1.700 1.720 1.740 1.760 1.780 1.800 ... ... ...

-2.00 1.660 1.680 1.700 1.720 1.740 1.760 1.780 1.800 ... ... ...

## Power System Studies Best Practices & International Experiences

71

Introduction

Rectifier

Id 1

Power

Inverter

+
Vd or Ed

7 3 4

2
1 Valve group 2 AC Filter 3 Converter transformer 4 DC filter 5 Ground/sea electrode or metallic 6 Receiving ac system 7 Sending ac system

Id

## Power System Studies Best Practices & International Experiences

72

Introduction
The basic building block of a LCC converter a six pulse Thyristor bridge.
It is important to note that the thyristors (the switch) can be turned ON with a firing pulse but cannot be forced to turn OFF. The turn off should happen naturally when certain ac system conditions are satisfied (more details on this later !)

## Think of the LCC - HVDC bridge as a frequency converter F => +/- 50 Hz

Power System Studies Best Practices & International Experiences 73

Introduction

12 pulse bridge The harmonics are reduced compared to the 6 pulse bridge.
Power System Studies Best Practices & International Experiences 74

Introduction

75

Introduction

## Power System Studies Best Practices & International Experiences

76

Introduction
Apart from the converter itself, there are a number of other aspects that should be given due consideration when designing HVDC systems.

## Nature of the AC system feeding the HVDC system at both ends.

LCC HVDC absolutely needs a AC system. Passive loads cannot be supplied through LCC HVDC The strength (short circuit capacity or the system impedance as seen from the converter bus) impacts HVDC operation The AC system characteristic under credible operating contingency conditions should be considered to avoid operational problems.

## Power System Studies Best Practices & International Experiences

77

Introduction
The AC system characteristic is its (steady state) impedance vs frequency plot. This is an important consideration.
Gray Whitedeer 64 km 176 km Tesla a1 147.04 km
CTT_WD_... CTT_GR_...

W h i te d e e r

TLine_11

P = 6 3 2 .8 Q = -2 1 9 .9 V = 0 .9 4 8 5 A V

4.81 7.40

Hereford

a2

GEN
TLine_8

E1 3 .1 6 [H ] ABC->G

68 km
CTT_ST_T...

Series Caps b a

68 km
CTT_ST_T...

## 121.6 km 52.8 km 76.8 km

TLine_12 ETT_TS_... Sh_WD_ST

Rieley 102.4 km

## P = 2190 Q = 796.8 V = 1.067 A RRL V RL

TLine_9

P = 8 0 5 .3 Q = - 3 6 6 .3 V = 0 .9 8 8 8 A V

67.2 km a1 Silverton a2 a3

ETT_TS_...

99101 NAZARETH

7.42 b 70.2 km
ETT_RL_...

Ea 250 a SVC

Sh_ST_CW

7.41

121.2 km
ETT_EC_...

3 .1 6 [H ]

## Power System Studies Best Practices & International Experiences

78

Introduction
- HVDC Controls.
-The converter valve firing controls -Protection -Measurements (current, voltage, firing angles .)
INVERTER CONTROLS
input current order (pu)

CO

## INVERTER CURRENT CONTROLS

DC Voltage (compensated) G 1 + sT VDCL MPVS D

## current order for rectifier

E Min CORD CORDER

VDCI

D + + F

* 0.01

## dc voltage measured at inverter

voltage compounding
* POWER Power

CMARG

0.1

## F CMIC G 1 + sT D CMIS Current (filtered) + CERRI D + -

F CERRIM I P BETAIC

## dc current measured at inverter

PSCAD implementation of a conventional HVDC inverter controls for thyristor firing pulses.
Pi B + D -

Max E

BETAI

AOI

## DGEI Delta Gamma Error

Alpha Order

F GMES Min in 1 Cycle Gamma GMESS D + + B -0.544 GERRI D E P Max GNLG I BETAIG

0.2618

GMIN

## Power System Studies Best Practices & International Experiences

79

Line Commutated HVDC systems Basic concepts Commutation Harmonics Reactive power and filtering Basic HVDC equation Control concepts Impact of AC systems strength Impact of AC System Characteristics

## Power System Studies Best Practices & International Experiences

80

Ia R=0
Main : Graphs 10.0 7.5 5.0 2.5 0.0 -2.5 -5.0 -7.5 -10.0 4.0 3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -4.0 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 Vas

Vas FP T Ea T FP

Voltage (kV)

0.00393 [H]

33 MVAR (Max)

C urrent (kA)

Ia_
4.0

FP1

F iring pulse

0.1700

0.1750

0.1800

0.1850

0.1900

0.1950

0.2000

0.2050

0.2100

0.0

[1] 2.30468

## Power System Studies Best Practices & International Experiences

81

Thyristor Characteristics
M ain : Graphs
Ia Ea

Ia 1.00

T1
R=0 E1

g1 T

T3

g1 T

y
g1 T T g1

0.50

T4

T2

## Firing angle (alpha) = 30 deg

-150 -200 2.0550 2.0600 2.0650 2.0700 2.0750 2.0800 2.0850 2.0900 ... ... ...

## Power System Studies Best Practices & International Experiences

82

Thyristor Characteristics
M ain : G raphs
Ia Ea

Ia 1.00

g1 T T

g1

R=0

E1

y
g1 T T g1

0.50

## 0.00 200 150 E 1 E a

Firing angle (alpha) = 90 deg Average dc voltage is zero Half cycle for thyristor to turn OFF
y

100 50 0 -50 -100 -150 -200 2.0550 2.0600 2.0650 2.0700 2.0750 2.0800 2.0850 2.0900 ... ... ...

83

Inverter:
Ia Ea

g1 T T

g1

y
R=0 E1 g1 T T g1

0.50

0.00 200 E 1 E a

y

## Very small time for thyristor to turn OFF

-100 -150 -200 2.0550 2.0600 2.0650 2.0700 2.0750 2.0800 2.0850 2.0900 ... ... ...

84

## Single Phase Bridge

M ain : G raphs
Ia Ea IT1 IT2

Ia 1.00

g1 T T

g1

0.50

0.05 1e6

E1

y
g1 T T g1

0.00 200 150 100 50 0 -50 -100 -150 -200 1.20 1.00 0.80 0.60 0.40 0.20 0.00 -0.20 E 1 E a

IT 1

IT 2

2.0550

2.0600

2.0650

2.0700

2.0750

2.0800

2.0850

2.0900

85

DC Equations

Rectifier

Inverter

86

87

## Reactive Power Requirements

Both rectifier and inverter consumes reactive power.
Sources of reactive power
AC Filters Shunt Capacitors Synchronous Condensers Static Var Compensators (SVC) STATCOM

Filters and shunt devices must be carefully designed to allow for successful HVDC system operation under all credible operating conditions over the planned life time of the system.
Power System Studies Best Practices & International Experiences 88

## Harmonics and Filtering AC Current Harmonics

kP+/-1 k=1,2 Magnitude 1/k

DC Harmonics
kP k=1,2,3

08/12/2013 Professional Power System Simulation Power System Studies Best Practices & International Experiences

89 89

AC system impact
Short Circuit Ratio Short Circuit Level MVA/ DC Power ESCR Short Circuit Level (Shunt Caps + AC Filters) DC Power

2 < Extremely Low ESCR 2< Low ESCR < 3 3 > High (Strong) SCR

90

## Problems with weak systems

1. Decreasing control system stability, sluggish control performance, slower fault recovery rate 2. Increased capacitive compensation to support system voltage 3. Expensive SVC may need to be applied 4. Resonance concerns. 5. During dc or ac fault, excessive reactive power can result in temporary overvoltages

08/12/2013 Professional Power System Simulation Power System Studies Best Practices & International Experiences

91 91

HVDC integration and design study requirements Basic load flow studies and fault studies Transient Stability studies Frequency scans Detailed Electromagnetic Transient Studies Temporary over voltages Communication failure Harmonic resonance problems Insulation coordination Sub synchronous resonance screening Control interaction & related Problems Other

## Power System Studies Best Practices & International Experiences

92

HVDC integration and design study requirements Basic load flow studies and fault studies Load flow studies to determine - ac system upgrade requirements - Critical contingencies for voltage and thermal violations - Must consider different operating conditions - Low and high load conditions - Consideration of future load/generation scenarios - Credible contingencies Fault studies - Determine fault levels - Determine SCR and ESCR at the HVDC bus - Screening for SSTI concerns with thermal generators in the ac system.

## Power System Studies Best Practices & International Experiences

93

HVDC integration and design study requirements Transient Stability studies - How do we represent the HVDC system ? - Typical representation Vs specific models with realistic control system representation (obtained from vendor)

## Power System Studies Best Practices & International Experiences

94

HVDC integration and design study requirements Frequency scans - AC system must be adequately represented in order to get accurate results - HVDC vendor will use the information to design filters (as well as fine tune control systems as necessary ?). - Both positive sequence and negative sequence impedances should be analyzed. - Note that under unbalanced voltage conditions, the HVDC converters will produce non characteristic harmonics (other than k.p +/- 1)

## Power System Studies Best Practices & International Experiences

95

HVDC integration and design study requirements Detailed Electromagnetic Transient Studies Temporary over voltages
Due to DC load rejection, faults Filters and capacitors must be adequately staged to avoid excessive TOV that could damage equipment.

Communication failure
Recovery from various ac system faults, particularly if they occur when the ac system is operating at contingencies resulting in very low short circuit capacity at the connecting ac bus-bar of the dc link

## Power System Studies Best Practices & International Experiences

96

HVDC integration and design study requirements Detailed Electromagnetic Transient Studies Harmonic resonance problems Due to transformer saturation and related harmonics Due to harmonic from the converter resulting from unbalanced ac voltage at the ac bus AC-DC coupling effects due to lines on the same right of way. Due to interactions with the exponential (dc) current components during ac faults, reactor switching etc Switching of filters

## Power System Studies Best Practices & International Experiences

97

HVDC integration and design study requirements Detailed Electromagnetic Transient Studies Insulation coordination
System level insulation coordination studies Valve level studies are typically done by the HVDC vendor

## Sub synchronous resonance screening

Any steam turbine and generator will have mechanical torsional oscillations on the shaft, particularly because a nearby ac system fault and its clearing. A nearby dc link and its controls may add some positive or negative damping to those oscillations. If the damping is negative, the oscillations will take longer to fade away. This can reduce the lifetime of the turbine shaft. It is best if the dc link controls are designed to provide positive damping to the shaft torsional oscillations. Unit Interaction factor as a screening method

## Control interaction & related Problems

Control instability and control interactions with other ac voltage controllers such as FACTS, other dc links and nearby generators with high gain static exciters

Other Issues

## Power System Studies Best Practices & International Experiences

98

HVDC integration and design study requirements Detailed Electromagnetic Transient Studies Other Issues/investigations
If the ac frequencies each side of the dc link are of similar but different frequencies, frequency interactions may transfer from one side to the other and result in oscillations. Performance of control system during faults Discharge of trapped charge in the DC line during ac side faults.

99

100