rick meeker, p.e. florida state university, center for ... pv studies of florida ... circuit 345 to...

Solar Forum 2013 High Penetration

F e b 1 3 - 1 4 , S a n D i e g o , C A

HIGH PENETRATION SOLAR PV TEST CASES IN THE FLORIDA GRID

Rick Meeker, P.E. Florida State University, Center for Advanced Power Systems Houtan Moaveni Univ. of Central Florida, Florida Solar Energy Center

The Sunshine State Solar Grid Initiative SUNGRIN

http://www.gosolarcalifornia.org/�


F e b 1 3 - 1 4 , S a n D i e g o , C A 2

Project Sponsors and Partners

Florida State University Center for Advanced Power Systems



F e b 1 3 - 1 4 , S a n D i e g o , C A 3

Team FSU CAPS: Rick Meeker, P.E., Mischa Steurer, Ph.D., Hui (Helen) Li, Ph.D., Omar

Faruque, Ph.D., Karl Schoder, Ph.D., Peter McLaren, Ph.D., Harsha Ravindra*, Liming Liu, Ph.D., Yan Zhou*, Ye Yang*, Ran Mo*, Jianwu Cao*, Isaac Leonard, Krystal Wright, James Langston, Tim Chiocchio, Mike Andrus

• UCF FSEC: Dave Click, Bob Reedy, P.E., Houtan Moaveni

FMPA: Tom Reedy, Ann Beckwith

FPL: Daniel Waugh, John Shaffer

FRCC: Eric Senkowicz, P.E.

GRU: David Kvaltine, Darko Kovac, Hamid Rezaei, Ernie Hodge, Hector Zamot-Ayala

JEA: Jay Worley, Tom Ventresca

Lakeland Electric: Jeff Curry, Phuong Tran, Lien Nguyen, Jackie Hayes, Willie Mathis

OUC: Jennifer Szaro, Eva Reyes, Orangel Garcia, Sam Griffin, Orlando Alancastro

AMEC: Mike Allman

OSISoft: Len Uronis, Candace Keyes

SMA: Elie Nasr

* students



F e b 1 3 - 1 4 , S a n D i e g o , C A 4

Focus Areas

PV and load variability characterization and impact Power system circuit model development

(distribution and transmission; Florida circuits) High-penetration PV impact analysis with FL utility

circuit models and data Development and testing (including HIL) of Power

electronics, storage, and control solutions and strategies

Outreach and engagement



F e b 1 3 - 1 4 , S a n D i e g o , C A 5

PV Variability at High-Pen Sites Data: • Satellite: SolarAnywhere data at 1 km spatial

and 1-min temporal resolution • Ground Measurements: PV site measured

data, typically 1 min., 15 min. Approach: • Classify and compare variability, temporal and

geographic, using Variability Index • Distribution of ramps of irradiance and

modeled power output • Estimate the reduction in variability due to

geographic smoothing • Quantify the amplitude and frequency of

occurrence of fluctuations at each timescale using a stationary wavelet transform

Conclusion: • Geographic smoothing offers a strong benefit

over short timescales • This benefit decreases with longer timescales

and increased spatial correlation



F e b 1 3 - 1 4 , S a n D i e g o , C A 6

High-Penetration PV Studies of Florida Distribution Circuits

1

2

3 4

5

6

7

8 9

JEA Circuit 345 to Jacksonville Solar Plant (Ph. 1-2)

GRU Circuit 435 GRU Circuit 432 (Ph. 2-3)

NASA KSC Circuit (w/FPL) (Ph. 1, 4-5)

Lakeland Electric Lakeland Airport 2 circuits (Ph. 2-4)

OUC Stanton Solar (Ph. 3-4)



F e b 1 3 - 1 4 , S a n D i e g o , C A 7

Modeling Tools Real Time Digital Simulator (RTDS)

• Electromagnetic Transient Program (EMTP) solution • 3-phase time-domain solution • >110,000 MFLOPS; 14 “racks”, parallel processing

• EMTP type simulation covers load-flow, harmonic, dynamic, and transient regime • 50 µs timestep simulation in real-time; <2 µs possible for portions of the simulation • 66 real-time electrical nodes per “rack” x 14 racks = 924 nodes • Extensive digital and analog I/O for interfacing hardware to simulation (>2500 analog,

>200 digital). Can connect in real-time to any electrical node within the simulation.

OpenDSS (EPRI) • Frequency domain solution – power flow, dynamics, harmonics • Running from MATLAB (where other processing, analysis, and plotting occurs)

PSSE (Siemens PTI) • Phasor-based load flow • Primarily used for transmission grid modeling



F e b 1 3 - 1 4 , S a n D i e g o , C A 8

Model Validation Approaches and Challenges Software based validation

• Cross-validation against other model results from SynerGEE and OpenDSS • Both steady-state and fault situations were compared (RTDS vs. OpenDSS

and RTDS vs. Synergee) • Fault currents at different locations provided by the utility are compared

against simulation results

Field measurement based validation • Voltage, current, real and reactive power at substation and along circuit • Synchronized data from 3-5 locations are sought • Imperfect knowledge of load variation along circuit and across phases

• Circuit load at substation and recloser (if present) known from field msmt.’s • Loads distributed based on the transformer’s kVA ratings and utility estimates

of typical transformer loading vs. load type (residential / commercial)



F e b 1 3 - 1 4 , S a n D i e g o , C A 9

JEA – PSEG Jacksonville Solar 15 MW DC; 12.6 MW AC Online Nov. 2009 Owner: PSEG; under PPA

to JEA 100 acres 24kV distribution feeder 230kV substation Feeder length : 9 miles PV location: 4.8 miles Max. ckt. load <12.6 MW Inverters (20):

> SMA Sunny Central 630 HE

Panels (~200,000): > First Solar FS-275



F e b 1 3 - 1 4 , S a n D i e g o , C A 10

JEA Circuit Model Validation Power flow validation against field measurements

For simulation runs at four different time periods, various times of day: Voltage at substation breaker in agreement within

1.97% to 2.21% Voltage at PV plant in agreement within 1.25% to

1.56% Current at substation breaker in agreement within

3.24% to 4.92%



F e b 1 3 - 1 4 , S a n D i e g o , C A 11

JEA Circuit Model – Voltage Profile Analysis • Examined voltage profile for varying X/R ratio,

• Circuit X/R is about 4.5 • At higher X/R (>7), low voltages would occur beyond

~7 km out • Examined voltage for varying load, and PV production,

• With loading greater than 8 MVA, at low PV production, and 11 MVA, at high PV production, it is possible to see low voltages on the feeder

• Examined voltage regulation, • With no regulation (actual circuit), traditional

regulation, and PV participation, single plant vs. multiple plants,

• Without regulation, using severe load swing, slightly low voltage can occur at end of feeder, but overall voltage tends to be quite stable and in limits

• Traditional regulation alleviates somewhat • PV regulation exhibits very good control (in simulation) • Multiple PV plants – concern with control interaction

0 2 4 6 8 10 12 14

0.86

0.88

0.9

0.92

0.94

0.96

0.98

1

1.02

1.04

1.06

Voltage profile for different loadings on the feeder

Length of the feeder in km

2.24.46.68.811.113.315.517.720.0

Service voltage upper limit 1.05pu

Service voltage lower limit 0.96pu

Volta

ge in

pu

PV power at 3 MW

Upper limit of service voltage (1.058pu)

Regulation Upper limit (1.03 pu)

Near Substation

Near PV

End of feeder



F e b 1 3 - 1 4 , S a n D i e g o , C A 12

Protection Impact Analysis for JEA Feeder & Substation Relay settings provided by JEA Substation & protection modeled in RTDS

• 2-Rack real time (60 µs) case • 5-Rack non-real time (1 µs) case with detailed

feeder data PV inverters are current limited to 1.3 p.u. Substation, which has strong source (230kV). Various types of faults applied at different

locations on substation for, • NO PV penetration. • PV penetration on 1 feeder. • PV penetration on many feeders.

Impact of PV location with respect to the fault location also studied

Conclusions: No sympathetic tripping on the PV feeder

breaker from adjacent feeder fault Results specific to this site. Need to examine

others

Phase A

Phase C

Phase B

Phase A

Phase B

Phase C

Current through breaker 344 for fault on FDR 344, Phase A, with 0.1 ohm fault impedance

Fault current from 12.6 MW PV system on FDR 345, due to

fault on FDR 344 Phase A



F e b 1 3 - 1 4 , S a n D i e g o , C A 13

GRU Circuit 435 Feeder (“6th St. Solar)

Substation Voltage Reg. Recloser Capacitors PV (large)

12.47 kV circuit with overhead primary about 4.5 miles.

Peak loading on the feeder 9 MVA.

Peak PV installed capacity 2.6MW distributed:

Recloser at around 2.2 miles from substation.

Four capacitor banks Voltage regulator installed

near substation. A fairly typical mix of

residential and commercial loads



F e b 1 3 - 1 4 , S a n D i e g o , C A 14

RTDS One rack real time GRU Feeder Model

1 Rack Case, reduced feeder model

– Line, transformer, and phasing data – 10 buses, 7 line sections, 7 PQ-

loads, 2 PV locations, 4 capacitors, 1 regulator

Load 2

Load

1

Load 7 Load

5Lo

ad 6

SubstationVoltageregulator

Ia,Ib,IcP,Q

Ia,Ib,IcVa, Vb, Vc

P,Q

Recloser

Cap Bank 1900 kVar

Cap Bank 2900 kVar

Cap Bank 3600 kVar

Cap Bank 4900 kVar

PV System 12.25 MW

PV System 20.35 MW

± 10% 32 steps2 min DB

S1

S2

Srcl

S3

S5

S4

S6

S7

00

01

0203

04

05

0607

08

BusNo

LineSection

• Feeder data was provided by GRU. • To be able to run model in real time,

feeder model was reduced. • Load distribution on feeder assumed

based on aggregated transformer ratings.



F e b 1 3 - 1 4 , S a n D i e g o , C A 15

0 10 20 30 40 50 60200220240260280

Phase A

0 10 20 30 40 50 60220240260280300

Phase B

Cur

rent

in A

MeasuredSimulated

0 10 20 30 40 50 60240

260

280

300

320Phase C

Time in mins

• Compared RTDS power flows against field data at breaker, recloser, and PV plant.

• Results of one hour simulation validation case for 5/14/2012 from 4.00 – 5.00 P.M.

• Average loading 7 MW for duration.

• Figure shows plot of measured vs. simulated breaker currents for 3 phases for one hour.

Field Measurement Based Validation of GRU Circuit 435

• Voltages at recloser location agree within less than 2.5% error • Average error for currents varies between 1.97% and 8.46%, depending upon

phase and location, with maximum error of 16.11%



F e b 1 3 - 1 4 , S a n D i e g o , C A 16

0 10 20 30 40 50 60200

250

300

350

Phase A

0 10 20 30 40 50 60200

250

300

350

Phase B

Bre

aker

Cur

rent

in A

0 10 20 30 40 50 60200

250

300

350

Phase C

Time in mins

No Lag/Lead10 sec Lag1 min Lag3 min Lag10 sec Lead1 min Lead3 min Lead

Examining Model Uncertainty Sensitivity to Different Error Sources Example: Validation Data Time Synchronization Errors



F e b 1 3 - 1 4 , S a n D i e g o , C A 17

PV Penetration Analysis – GRU (with RTDS Model)

High loading scenario – 9 MVA

2/13/2013 17

0 0.5 1 1.5 2 2.5 3 3.5 41

1.01

1.02

1.03

1.04

1.05

1.06

1.07

1.08

distance (mi)

p.u.

vol

tage

PV penetration [MW] at high circuit loading

no PV2.6 MW4 MW6 MW8 MW10 MW

0 0.5 1 1.5 2 2.5 3 3.5 41

1.01

1.02

1.03

1.04

1.05

1.06

1.07

1.08

distance (mi)

p.u.

vol

tage

PV penetration [MW] at low circuit loading

no PV2.6 MW4 MW6 MW8 MW10 MW

Low loading scenario – 4 MVA



F e b 1 3 - 1 4 , S a n D i e g o , C A 18

PV Participation in Voltage Regulation GRU Circuit – High Loading

2/13/2013 18

0 0.5 1 1.5 2 2.5 3 3.5 40.99

1

1.01

1.02

1.03

1.04

1.05

1.06

1.07

1.08

distance (mi)

p.u.

vol

tage

Voltage profile for different regulation schemes - High loading scenario

No RegulationConstant pfConstant QSet Point V primaryGerman LV curve 0.90 pfVolt - Var CurveVolt - pf curve 0.95Volt - pf curve 0.90

0 0.5 1 1.5 2 2.5 3 3.5 40.99

1

1.01

1.02

1.03

1.04

1.05

1.06

1.07

1.08

distance (mi)

p.u.

vol

tage

Voltage profile for different regulation schemes - Low loading scenario

No RegulationConstant pfConstant QSet Point V primaryGerman LV curve 0.90 pfVolt - Var CurveVolt - pf curve 0.95Volt - pf curve 0.90

High loading scenario – 9 MVA Low loading scenario – 4 MVA



F e b 1 3 - 1 4 , S a n D i e g o , C A 19

Dynamic Model of the Florida Transmission Grid PSSE Validation of 154-bus Model Against Very Large Utility Planning and Event Analysis Model • Developing reduced size FL grid model that

captures important behavior, for investigation into grid changes like high-pen. PV effects

• Currently validating 154 bus PSSE model • A model of <300 busses to be built in RTDS



F e b 1 3 - 1 4 , S a n D i e g o , C A 20

Summary Some Observations from Work-to-date - Modeling Models that capture dynamic behavior will be needed for

investigation of control behavior and interactions Distribution circuits often exhibit significant phase imbalance;

(individual) phase analysis will be needed in some cases Validation of current measurements (more sensitive) in

addition to voltage provides increased confidence in model Common sources of modeling and simulation error:

• Field measurement time synchronization • Load estimates (spatial, temporal, and per phase) • Operation and settings of regulation devices



F e b 1 3 - 1 4 , S a n D i e g o , C A 21

Summary Some Observations from Work-to-date - Analysis Percent penetration and 15% screen is a poor metric Seeking a classification of circuit types and more appropriate

metrics than % penetration – must look at many circuits first PV can be very effective (usually exceeding performance using

traditional means) in voltage regulation System dynamics and interaction of regulation equipment can

cause issues Control and protection schemes, algorithms and settings –

challenge and opportunity Hardware-in-the-loop simulation is useful for evaluation and

testing of dynamic behavior in system context



F e b 1 3 - 1 4 , S a n D i e g o , C A 22

Future Work Arrive at meaningful penetration metrics (screens) through

analysis of a large no. of circuits Reduce, refine, and streamline modeling and analysis

processes and tools Develop and demonstrate hardware-in-the-loop evaluation

methods and rationale for improved anti-islanding and islanded operation inverter testing

Examine transmission level benefits (e.g. “230 kV DG”) through integrating distribution and transmission simulation-assisted analysis

Develop a basis for PV capacity credit PV as mitigation of FIDVR



F e b 1 3 - 1 4 , S a n D i e g o , C A 23

Publications Zhou, Y., Li, H., Liu, L., “Integrated Autonomous Voltage Regulation and Islanding Detection for High Penetration PV

Applications”, IEEE Transactions on Power Electronics, Vol 28., No. 6, June 2013. Langston, J., Schoder, K., Steurer, M., Faruque, O., Hauer, J., Bogdan, F., Bravo, R., Mather, B., Katiraei, F., “Power

hardware-in-the-loop testing of a 500 kW photovoltaic array inverter”, IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society, Oct. 2012.

Ravindra, H., Faruque, O., McLaren, P., Schoder, K., Steurer, M., Meeker, R., “Impact of PV on Distribution Protection System”, Proceedings, North American Power Symposium, Sept. 2012.

Liu,. X., Aichhorn, A., Liu, L., Li, H., “Coordinated Control of Distributed Energy Storage System with Tap Changer Transformers for Voltage Rise Mitigation Under High Photovoltaic Penetration”, IEEE Transactions on Smart Grid, Vol. 3., No. 2, June 2012.

Ravidra, H., Faruque, O., Schoder, K., Steurer, M., McLaren, P., Meeker, R., “Dynamic Interactions Between Distribution Network Voltage Regulators for Large and Distributed PV Plants”, Proceedings of the IEEE Power Engineering Society, Transmission and Distribution Conference and Exposition, Orlando, FL, IEEE , May 2012.

Click, D., Moaveni, H., Davis, K., Meeker, R., Reedy, R., Pappalardo, A., Krueger, R., “Effects of Solar Resource Variability on the Future Florida Transmission and Distribution System”, ”, Proceedings of the IEEE Power Engineering Society, Transmission and Distribution Conference and Exposition, Orlando, FL, IEEE , May 2012.

Meeker, R., Domijan, A., Islam, M., Omole, A., Islam, A., Damnjanovic, A., “Characterizing Solar PV Output Variability and Effects on the Electric System in Florida, Initial Results”, Proceedings of the 5th International Conference on Energy Sustainability, ASME, Aug., 2011.

Meeker, “Integrating High-Penetration Levels of Renewables into the Grid – What we Know Now”, accepted, ASME Energy Committee Colloquium, June 2012.

Click, et al, “Effects of Solar Resource Variability on the Future Florida Transmission and Distribution System”, accepted, IEEE PES T&D, May 2012.

Allman, M., Meeker, R., Reedy, B., Senkowicz, E., “Integrating Solar PV into the Grid”, Relay, the quarterly magazine of the Florida Municipal Electric Association (FMEA), Winter 2009.



F e b 1 3 - 1 4 , S a n D i e g o , C A

Q &A AND DISCUSSION

Contact:

R. H. (Rick) Meeker, Jr., P.E. [email protected] 850-645-1711

http://www.caps.fsu.edu/sungrin


mailto:[email protected]�

http://www.caps.fsu.edu/sungrin�

rick meeker, p.e. florida state university, center for ... pv studies of florida ... circuit 345 to...

Documents