Making Dynamic Simulations Output Comparable to Synchrophasor

Measurements of PMUsA. Srivastava†, H. Lee†, M. Zhou†, P. Banerjee†, E. Farantatos‡

and M. Patel‡

†Smart Grid Demonstration and Research Investigation Lab (SGDRIL)Energy System Innovation center, Washington State University

‡Electric Power Research InstituteContact: asrivast@eecs.wsu.edu

NASPI March 2016

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

Number of applications for PMU based model validation:• Validation of Power Plant Models (PPPD),

Renewable Energy Models (REMV), Static Var System Models (SVSMV) - (EPRI)

• Validating Generator model using PPMV (BPA)

• Model validation of HVDC and Nuclear Unit (ISO-NE)

• Tuning Wind Turbine model (ERCOT)• Load Model validation (WECC)

Better model leads to better analysis, Understanding interaction of components, Improve situational awareness and Higher resource utilization

Typical Schematic for PMU based model Validation

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

The phasor values obtained from dynamic simulationtools and PMUs in the field may differ due to:• The filters used in the PMU may introduce

magnitude attenuation and phase offset• The results from phasor domain simulations do not

result from estimation algorithm from a point ofwave signal

• PMU may use several cycle data to obtain the phasorat any instant. Hence, estimated phasor by PMU isrepresentative of several preceding cycles.

• Test system benchmarking by matching response fromRTDS and Electromechanical Dynamics SimulationPlatform

• Analyzing effect of PMU filtering and estimation bycomparing RTDS+PMU with Dynamics simulationplatform

Real Time Simulator PDC

PMU A

PMU B

PMU C

PMU D

Power Amplifier

Power Amplifier

...

Ethernet Switch

GPS

Ethernet LAN Connection

Hardwired Connection

GPS Clock Connection

Test Bed @ Washington State University PMU Performance Analyzer

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

G1

1

2

3

4

5

7

8

9

10

11

1413

12

6

G2

Test System Model: Kundur System

Test System Model: IEEE 14 System

• The voltage magnitude and angle at each bus are obtained from the power flow solution of the test systems in both PSS/E and RTDS.

• Total Vector Error (TVE) gives a measure of the mismatch between the voltage phasors obtained from the RTDS and PSS/E, by considering the phasors generated from RTDS as the true value.

where,

: denotes the voltage phasors obtained from PSS/E power flow solution: denotes the voltage phasors obtained from RTDS power flow solution

Comparison between RTDS and PSS/E

• According to the IEEE ICAP Synchrophasor Measurement Test SuiteSpecifications [1], the following two criteria are used as metrics tocompare the response between the two platforms:

– Error at the Peak Point– Error at Settling Point

[1] “IEEE Synchrophasor Measurement Test Suite Specification," in IEEE Synchrophasor Measurement Test Suite Specification”, vol., no., 2014

N = Number of peak points and valley points

where,

PRTDS, PPSSE=values at peak points and valley points

e = mismatch at settling point

where,

N= Number of peak points and valley pointsTRTDS, TPSSE=time at peak points and valley points

1 2 3 4 5 6 7 8 9 10 11 12 13 140

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.01Total Vector Error (IEEE 14 Bus System)

BUS

Tota

l Vec

tor

Err

or (%

)

1 2 3 4 5 6 7 8 9 10 110

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.01

BUS

Tota

l Vec

tor

Err

or (%

)

Total Vector Error (Kundur System)

Steady State Response Results and Benchmarking

1 2 3 4 5 6 7 8 9 10 11 12 13 140

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.01Total Vector Error (IEEE 14 Bus System)

BUS

Tota

l Vec

tor

Err

or (%

)

1 2 3 4 5 6 7 8 9 10 110

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.01Total Vector Error (Kundur Bus System)

BUS

Tota

l Vec

tor

Err

or (%

)

Operation Condition A Operation Condition B

%TVE (IEEE 14)

%TVE(Kundur)

Exciter Step Change Response of Generator 1

(IEEE 14)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 51

1.01

1.02

1.03

1.04

1.05

1.06

1.07

1.08

1.09

1.1

time (s)

Vte

rmin

al1

(p

.u.)

PSSERTDS

Terminal Voltage Magnitude of Generator 1

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5-2

-1

0

1

2

3

4

5

time (s)

Efd

1 (

p.u

.)

PSSERTDS

Field Voltage Magnitude of Generator 1

Exciter Step Change Response of Generator 1

(Kundur)

Terminal Voltage Magnitude of Generator 1

0 1 2 3 4 5 6 71

1.01

1.02

1.03

1.04

1.05

1.06

1.07

Time (s)

Term

inal V

olt

ag

e (

p.u

.)

RTDSPSSE

Magnitude Error : 0.1106 %Time Error : 0.03565 sec

Field Voltage Magnitude of Generator 1

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50.8

1

1.2

1.4

1.6

1.8

2

2.2

Time (s)

Efd

(p

.u.)

RTDSPSSE

Magnitude Error: 0.3162 %Time Error : 0.01416 sec

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

• The Absolute Vector Error (AVE) was used as a metric to quantify thecomparison between the PSS/E and RTDS simulation results and betweenthe PSS/E simulation results and the measured responses from the PMUs.

• The Total Vector Error comparing PMU ‘x’ with PSS/E and RTDS with PSSEis also evaluated for the best and the worst PMU during the initial 0.2secof the transient for the PMUs located at the generator bus for both thetest system.

Four PMUs have been used in the experiments:• PMU A Vendor 1 (P-Class) • PMU B Vendor 2 (M-Class) • PMU C Vendor 2 (P-Class) • PMU D Vendor 3 (DFR)

IEEE 14 Bus system: Load Shedding at Bus 3

0.95 1 1.05 1.11.04

1.045

1.05

1.055

1.06

1.065

1.07

1.075

Time (sec)

Vo

ltag

e (

pu

)

Voltage Magnitude Response

PSSERTDSPMU BPMU D

Voltage Magnitude Response from 0.95sec to 1.1sec

– Load Shed – PMU installed at Bus 2

Phase Angle Response from 0.95sec to 1.1sec

– Load Shed – PMU installed at Bus 2

0.95 1 1.05 1.1-6

-5.5

-5

-4.5

-4

-3.5

-3

Time (sec)

An

gle

(d

eg

ree

)

Phase Angle

PSSERTDSPMU BPMU D

IEEE 14 Bus system: Load Shedding at Bus 3

AVE from 0.95sec to 1.1sec – Load Shed – PMU installed at Bus 2

0.95 1 1.05 1.10

0.005

0.01

0.015

0.02

0.025

0.03

0.035Absolute Vector Error

Time (sec)

|VE

|

AVE RTDS - PSSEAVE PMU B - PSSEAVE PMU D - PSSE

TVE from 0.95sec to 1.1sec – Load Shed – PMU installed at Bus 2

0.95 1 1.05 1.10

0.5

1

1.5

2

2.5

3

3.5

Time (sec)

TV

E (

%)

Total Vector Error

TVE RTDS - PSSETVE PMU B - PSSETVE PMU D - PSSE

Kundur system: Bus Fault at Bus 6

Voltage Magnitude Response from 0.95sec to 1.2sec

– Bus Fault – PMU installed at Bus 3

0.95 1 1.05 1.1 1.15 1.20.9

0.92

0.94

0.96

0.98

1

1.02

1.04

Time (sec)

Vo

ltag

e (

pu

)

Voltage Magnitude Response

PSSERTDSPMU CPMU D

Phase Angle from 0.95sec to 1.2sec– Bus Fault – PMU installed at Bus 3

0.95 1 1.05 1.1 1.15 1.2-100

-90

-80

-70

-60

-50

-40

-30

-20Phase Angle

Time (sec)

An

gle

(d

eg

ree

)

PSSERTDSPMU CPMU D

Kundur system: Bus Fault at Bus 6

AVE from 0.95sec to 1.2sec– Bus Fault – PMU installed at Bus 3

0.95 1 1.05 1.1 1.15 1.20

0.02

0.04

0.06

0.08

0.1

0.12

0.14Absolute Vector Error

Time (sec)

|VE

|

AVE RTDS - PSSEAVE PMU C - PSSEAVE PMU D - PSSE

TVE from 0.95sec to 1.2sec– Bus Fault – PMU installed at Bus 3

0.95 1 1.05 1.1 1.15 1.20

2

4

6

8

10

12

14Total Vector Error

Time (sec)

TV

E (

%)

TVE RTDS - PSSETVE PMU C - PSSETVE PMU D - PSSE

0 1 2 3 4 5 60

2

4

6

8

10

12

14Total Vector Error

Modulation Frequency (Hz)

TV

E (

%)

PMU APMU BPMU CPMU D

PMUs Total Vector Error for Joint Amplitude and Phase Modulated Signal

0 1 2 3 4 5 6 7 8 9 100

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8Total Vector Error

Time (sec)

TV

E (

%)

PMU APMU BPMU CPMU D

PMUs Total Vector Error for Frequency Ramp

OutlineIntroduction to PMU based model validation

Motivation and goals for this study

• Test system model• Steady state response• Dynamic response

Test system benchmarking in RTDS and Dynamic Simulators

• Disturbance test scenarios and results• Dynamic compliance validation and results

Hardware in the loop benchmarking and PMU filtering effect

Summary

Summary

25

The steady state power flow matches 0.005% of the TVE for both. The step response of the excitation system of individual generators resulted in

magnitude error and the time error which are less than 0.5% and 0.04sec, respectively.

The AVE of the PMU D, when installed at Bus 2 is 0.005 for around 3sec after the occurrence of the disturbance, then decreased to 0.002. The AVE of other three PMUs are equal to the AVE of RTDS and PSS/E.

The AVE of PMU D is more than the AVE of other PMUs for 1 sec after the fault is cleared. The AVE of other PMUs are nearly same as the AVE of the RTDS and PSS/E.

The TVE of RTDS meters for Joint AM-PM is observed to be less than 0.001% The TVE of PMU D is more than 10% for most of the modulation frequency. The TVE of the RTDS meters for the frequency ramp test is less than 0.15% The TVE of PMU D is more than 1% for frequency ranging from 55Hz to 58Hz

Conclusions

26

The response of the DFR is observed to be different as compared to that of other PMUs

The TVE of P class PMUs is also more than 1% for off nominal frequency more than 62Hz and lower than 58Hz

M class PMUs accurately report the voltage phasor during slow to moderate disturbances.

The application of PMU data just after the occurrence of the disturbance for modeling systems of very small time constants is a challenging task. Hence, suitable filters may be designed for PSS/E to match the system responses with P class and M class PMUs separately.