Laboratory Scale FACTS Controller Development - sandia.gov · Simulated Power Transmission System....

Laboratory Scale FACTS Controller Development

Mariesa CrowUniversity of Missouri-Rolla

Funded in part by the Energy Storage Systems Program of the U.S. Department Of Energy (DOE/ESS) through Sandia National Laboratories (SNL

Issues• Hardware-software co-design• Device placement and control

– Decentralized– Steady-state– Dynamic

• Cyber security• Reliability

33

vv

Transmission LineGenerationFACTS

Wind Power

Energy Storage

Solar Power

Energy Storage

FACTS Device

Distributed Decisions

Power Electronics

Communications

Sensing and monitoring Inputs

Power Electronics

Power Electronics

Distributed controland fault/attack detection

FACTS Interaction Laboratory

HIL Line

UPFC

Simulation

Engine

FACTS Power System Model

FACTS Power System

num_generatorsnum_lines

num_FACTSnum_busestotal_power

supply_power()reconfigure()

run_FACTS_placement()

Voltage stability, no overloads, flow balance,

availability

Service Provider(Utility)

providesinitiates

Contin-gency

typenamecause

affects : Event

Eve

nt R

ecep

tor

controls : Event

Attributes

Methods

Constraints

First Decomposition

FACTS Device

manipulates

setpoint

change_setpoint(newSetpoint)control_power_line()reconfigure()

flow balance

Neighbors limits, monitors

places

Placement

locations

compute_locations()

Placement is optimized

Power Transmission System

flowscapacitiesgenerationsloads

AG[For each line-capacity <= flow <= capacity{checked by FACTS}]------------------------------AG[For each bussum of lines.flow is 0.{checked by FACTS}]------------------------------AG[For each loadload is greater than or equal to 0.{checked by FACTS}]

supply_power()

affects : Event

uses

provides

initiates

affects :Event

senses

Eve

nt R

ecep

tor

FACTS Device

Power Transmission System

Placement AlgorithmG

G

ys ca p e

Hardware: physical buses & lines

SimulatedPowersystem

Unified Power Flow Controller (UPFC) FACTS

DSP Board

UPFC Power Electronics

change_switches()inject_voltage()inject_current()read_sensors()

line_powerline_voltage

power < max_limit

manipulates, sensesreads

controls

limits, monitors

Embedded Computer

CAN Bus

Interface board

controlsreads

Interfacesetpoint

capture_sensor_data()synchronize()get_line_power()change_setpoint(newSetpoint)compute_next_level_setpoint()DSP Board

EmbeddedComputer

UPFC PowerElectronics

Power Supply Sensor Board Current Sensors

DSP Board(Under the data cable)

Interface board

Simulated Power Transmission System

Simulation Engine(Load Flow) HIL Line

sensor_datapower_system_config

configure_simulation()accept_contingency()send_sensor_data()

set_HIL_line()compute_power()

generation_setpoint

change_HIL_Line_flow()

sets

affects : Event

sensesuses

manipulateslimits, monitors

Simulation Engine Hardware in the LoopLine

3332

31 30

35

80

78

747966 7

5

77

7672

8281

8683

8485

156 157 161162

vv

167165

15815915544

45160

166

163

5 11

6

8

9

1817

43

7

14

12 13

138139

15

19

16

112

114

115

118

119

103

107

108

110

102

104

109

142

376463

56153 145151

15213649

4847

146154

150149

143

4243

141140

50

57

230 kV345 kV500 kV

36

69

Simulation Engine

(multiprocessor)

UPFC

UPFCUPFC

Machine 1

D/A output

A/D input

UPFC 1

Programmable load

Machine 2

Machine 3

Power System Simulation Engine

Programmable load

Programmable load

UPFC 2

UPFC 3

D/A output

A/D input

D/A output

A/D input

Manual power flow control

Manual power flow control

Placement for steady-state performance

Placement for dynamic performance

Long term control (MF & SQP)

agent based long term control

Line overload cascading failure

scenarios

dynamic cascading failure scenarios

Closed-loop long term control

Cascading failure scenarios

Closed loop dynamic control

Closed loop multi-device

Dynamic nonlinear control

Visualization

prev

ious

wor

k

IEEE 118 Bus Test System

Manual Power Flow Control

4 5 6 7 8 9 10-460

-440

-420

-400

-380

-360

-340

-320

-300

-280

-260

time (seconds)

P1 a

nd P

2 (W)

10 11 12 13 14 15-0.75

-0.7

-0.65

-0.6

-0.55

-0.5

-0.45

time (secods)

P1 a

nd P

2 (pu)

10 11 12 13 14 150.9595

0.96

0.9605

0.961

0.9615

0.962

0.9625

0.963

0.9635

0.964

time (seconds)

bus

volta

ge (p

u)

10 11 12 13 14 15-0.08

-0.075

-0.07

-0.065

-0.06

-0.055

-0.05

time (seconds)

bus

angl

e (ra

d)

actual UPFC power flows

measured andfiltered intosimulation

simulatedbus voltages

& angles

Note induced low frequency oscillations

Manual power flow control

Placement for steady-state performance

Placement for dynamic performance

Long term control (MF & SQP)

agent based long term control

Line overload cascading failure

scenarios

dynamic cascading failure scenarios

Closed-loop long term control

Cascading failure scenarios

Closed loop dynamic control

Closed loop multi-device

Dynamic nonlinear control

Visualization

prev

ious

wor

k

Closed-loop long term control

• Which placements and settings yield the lowest PI over all possible contingencies?

Performance Index

∑ ∑

=

SLC Linesall i

i

SSPI

2

max

Si – Power flow on line (MVA)

Simax – Rating of the line

SLC – Single Line Contingency

PI distributes line loadings as higher loadings incur heavier penalties than lower loadings

-2-1012

-10

1

55

60

65

70

75

Facts 1 SettingsFacts 2 Settings

PI F

itnes

sOptimal Setting for a Single Contingency

Degrees of Freedom:

• Number of FACTS devices• Settings• Placements

across the set of all contingencies

24 6

24.8

25

25.2

25.4

25.6

25.8

26

Tota

l Ove

rload

ed P

ower

EA + SQPH + SQP

Evolutionary Algorithm vs.Pruned Brute Force (Heuristic)

Manual power flow control

Placement for steady-state performance

Placement for dynamic performance

Long term control (MF & SQP)

agent based long term control

Line overload cascading failure

scenarios

dynamic cascading failure scenarios

Closed-loop long term control

Cascading failure scenarios

Closed loop dynamic control

Closed loop multi-device

Dynamic nonlinear control

Visualization

prev

ious

wor

k

Cascading ScenarioOutage 48-49

45

46

W.Lancst Crooksvl

47

G69

68

G

G

49

66

65

Zanesvll48

50

Philo

MuskngumN

MuskngumS

67

G

Natrium

Kammer

44WMVernon

N.Newark

SpornW

Summerfl

62

64

37

34

36

NwLibrty39

40 41 42

43

S.Kenton

38

S.TiffinWest End Howard

Rockhill

EastLima

Sterling

Portsmth

Bellefnt 74 75SthPoint

CollCrnr23

GTannrsCk

Trenton

24

Portsmth

Hillsbro72

70

71

Sargents73

NPortsmt

WLima35

SpornE

54

51

Cascading ScenarioOutage 48-49

45

46

W.Lancst Crooksvl

47

G69

68

G

G

49

66

65

Zanesvll48

50

Philo

MuskngumN

MuskngumS

67

G

Natrium

Kammer

44WMVernon

N.Newark

SpornW

Summerfl

62

64

37

34

36

NwLibrty39

40 41 42

43

S.Kenton

38

S.TiffinWest End Howard

Rockhill

EastLima

Sterling

Portsmth

Bellefnt 74 75SthPoint

CollCrnr23

GTannrsCk

Trenton

24

Portsmth

Hillsbro72

70

71

Sargents73

NPortsmt

WLima35

SpornE

54

51

Cascading ScenarioOutage 48-49

45

46

W.Lancst Crooksvl

47

G69

68

G

G

49

66

65

Zanesvll48

50

Philo

MuskngumN

MuskngumS

67

G

Natrium

Kammer

44WMVernon

N.Newark

SpornW

Summerfl

62

64

37

34

36

NwLibrty39

40 41 42

43

S.Kenton

38

S.TiffinWest End Howard

Rockhill

EastLima

Sterling

Portsmth

Bellefnt 74 75SthPoint

CollCrnr23

GTannrsCk

Trenton

24

Portsmth

Hillsbro72

70

71

Sargents73

NPortsmt

WLima35

SpornE

54

51

Cascading ScenarioOutage 48-49

45

46

W.Lancst Crooksvl

47

G69

68

G

G

49

66

65

Zanesvll48

50

Philo

MuskngumN

MuskngumS

67

G

Natrium

Kammer

44WMVernon

N.Newark

SpornW

Summerfl

62

64

37

34

36

NwLibrty39

40 41 42

43

S.Kenton

38

S.TiffinWest End Howard

Rockhill

EastLima

Sterling

Portsmth

Bellefnt 74 75SthPoint

CollCrnr23

GTannrsCk

Trenton

24

Portsmth

Hillsbro72

70

71

Sargents73

NPortsmt

WLima35

SpornE

54

51

Manual power flow control

Placement for steady-state performance

Placement for dynamic performance

Long term control (MF & SQP)

agent based long term control

Line overload cascading failure

scenarios

dynamic cascading failure scenarios

Closed-loop long term control

Cascading failure scenarios

Closed loop dynamic control

Closed loop multi-device

Dynamic nonlinear control

Visualization

prev

ious

wor

k

0 5 10376

376.5

377

377.5ω 5

0 5 10376.4376.6376.8

377377.2377.4

ω 6

0 5 10376.4376.6376.8

377377.2377.4

ω 7

0 5 10376

376.5

377

377.5

378

ω 8

time (sec)time (sec)

(rad

/sec

)(r

ad/s

ec)

(rad

/sec

)(r

ad/s

ec)

cntrduncntrd

0 5 10376.4376.6376.8

377377.2377.4

cntrduncntrd

0 5 10376.4376.6376.8

377377.2377.4

376 4376.6376.8

377377.2377.4

376 4376.6376.8

377377.2377.4

ω 1(r

ad/s

ec)

ω 3(r

ad/s

ec)

ω 2(r

ad/s

ec)

ω 4(r

ad/s

ec)

Two devices – uncoordinated control design – local information only

Two devices – coordinated control design – local information only

0 5 10376.4

376.6

376.8

377

377.2

377.4

ω 1 (rad/s

ec)

cntrduncntrd

0 5 10376.4

376.6

376.8

377

377.2

377.4

ω 2 (rad/s

ec)0 5 10

376.4

376.6

376.8

377

377.2

377.4

ω 3 (rad/s

ec)

tiime (sec)0 5 10

376.4

376.6

376.8

377

377.2

377.4

ω 4 (rad/s

ec)time (sec)

Two devices – H∞ uncoordinated control design – tie line information only

Manual power flow control

Placement for steady-state performance

Placement for dynamic performance

Long term control (MF & SQP)

agent based long term control

Line overload cascading failure

scenarios

dynamic cascading failure scenarios

Closed-loop long term control

Cascading failure scenarios

Closed loop dynamic control

Closed loop multi-device

Dynamic nonlinear control

Visualization

prev

ious

wor

k

Seed Physical and Logical Intrusions

• Assertions describe the correctness of the control algorithms

• Software and hardware errors will be seeded into the FACTS network and the fault tolerance will be reported

• This behavior will be used to develop security policies for FACTS power systems

Visualization

Special Thanks

• Imre Gyuk - DOE Energy Storage Program• Stan Atcitty - Sandia National Laboratories• John Boyes - Sandia National Laboratories