© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m

Tom Key, EPRI, tkey@epri.com

February 6, 2020

Neutral-Grounding for Inverter-Connected DERFor NY ITWG

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m2

NYSERDA Effective Grounding Project

TASK 1. Assess Current Utility Practice To Inform Objectives

– Determine Current Utility Practice (Guided interviews/Survey)

– Clarify Grounding Objectives and Specific Requirements, Dec 4,2019

TASK 2. Develop Relevant Scenarios and Models to be Analyzed (includes Laboratory and HIL Test, and Modeling Objectives)

TASK 3. Coordinated and Concurrent Analysis, Simulation, and Testing

TASK 4. Develop Guidelines for Inverter Effective Grounding by July 19

TASK 5. Deliver Final Report, Outreach and Education (proposed ITWG workshop in 2020)

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m3

Q3 – Do you require additional ground sources for DER (either by a grounding transformer or a grounded-wye/delta transformer)?

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m4

Element Pickup Range Time Delay

Undervoltage (27) 0.5 pu 1 secUndervoltage (27) 0.7 pu 10 secUndervoltage (27) 0.88 pu 20 secOvervoltage (59) 1.1 pu 12 secOvervoltage (59) 1.2 pu 0.16 secUnderfrequency (81u) 56.5 Hz 0.16 secUnderfrequency (81u) 58.8 Hz 299 secOverfrequency (81o) 61.2 Hz 299 secOverfrequency (81o) 62.5 Hz 0.16 sec

Element Pickup Time Delay(cumulative)

TOV 2 pu 0.0016 sec

TOV 1.7 pu 0.003 sec

TOV 1.4 pu 0.016 sec

TOV 1.3 pu 0.166 sec

Inverter overvoltage (TOV) trip settings

ABC

120kV 2500MVA

ABC

abc YgYg

12.47 kV / 120 kV47 MVA

abc

ABC

Feeder Section25 miles

A B C

a b cYg

Yg480V / 12.47 kV 1 MVA

ABC

abc

Breaker

ABC ABCSLG fault

abc

ABC

Feeder Section11 miles

abc

ABC

Feeder Section31 miles

A B CLoad1

3PH INV

Preliminary Simulation Study – Overview

Trip settings in IEEE 1547-2018, Category III

1 MVA

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m5

Overvoltage Scenarios and Grounding 1. Load Rejection Overvoltage (LRO) in balanced 3-phase system

– Yg/Yg DER transformer (effective grounding does not matter)– Main breaker opens (no ground fault)– Load varies from 0 pu -1 pu (Δ or Y connected load)

2. Combined LRO and ground fault overvoltage (GFO)– Yg/Yg and Δ/Yg DER transformers (with and w/o supplemental grounding)– Permanent ground fault and main breaker opens– Load varies from 0 pu -1 pu (ungrounded load, spot check with 30% grounded)

3. GFO with varying amount of grounded load– Yg/Yg and Δ/Yg DER transformer (no supplemental grounding)

– Permanent ground fault occurs and main breaker opens– Load to generation ratio = 1, grounded load % varies from 0% –100%

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m6

7.4 Limitation of Overvoltage Contribution (1547-2018)

7.4.1 Limitation of overvoltage over one fundamental frequency period For Ground Fault Overvoltage (GFO) - on any portion of the Area EPS that is designed to

operate effectively grounded to exceed 138% of its nominal line-to-ground fundamental frequency voltage for a duration exceeding one fundamental frequency period.

7.4.2 Limitation of cumulative instantaneous overvoltage

For Load Rejection Overvoltage (LRO)

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m7

PV System LRO Events two different inverter simulations.Pass 1547.1 LROV Test Fail 1547.1 LROV Test

0

1

2

3

VRM

S-Se

c (p

u)

A

B

C

-2

-1

0

1

2

Vsec

(pu)

3.97 3.98 3.99 4 4.01 4.02 4.03 4.04 4.05

time (s)

0

0.2

0.4

0.6

0.8

1

Trip

sig

nal

OUFP

OUVP

TOVP

SW opens

Inv trips

18 ms

1.38 pu@ 4.1 s

0

1

2

3

VRM

S-Se

c (p

u)

-2

-1

0

1

2

Vsec

(pu)

4 4.05 4.1 4.15 4.2

0

0.2

0.4

0.6

0.8

1

Trip

sig

nal

1.38 pu@ 4.1 s

SW opens

Inv trips

164 ms

• With transient overvoltage protection (TOVP)• Inverter is able to trip within the required time

• TOVP disabled• Inverter fails to trip within the required time

time (s)

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m8

Scenario 1 – LRO

3.9 3.95 4 4.05 4.1 4.15 4.2 4.25 4.3

t/s

-2

-1

0

1

2

Vinv

(pu)

A

B

C

X 4.02

Y 1.729

3.9 3.95 4 4.05 4.1 4.15 4.2 4.25 4.3

t/s

-2

-1

0

1

2

Vinv

(pu)

X 4.02

Y 1.7268

Certified inverter

Uncertified inverter

25.7 ms by TOVP

167 ms by OVP

2.00

1.73 1.69 1.66 1.62 1.591.52

1.341.21

1.101.02

0.9

1.1

1.3

1.5

1.7

1.9

2.1

0 0.2 0.4 0.6 0.8 1

Peak

pha

se v

olta

ge (p

u)

load/Gen ratio

Uncertified

Certified

166 167 167 168 172 174 175 182

267 267 267

19.6 25.7 28.3 29.6 33.6 3953.2

182

267 267 268

0

50

100

150

200

250

300

0 0.2 0.4 0.6 0.8 1Tr

ip ti

me

(ms)

load/Gen ratio

Uncertified

Certified

TOVP

TOVP - Transient overvoltage protectionOVP - Overvoltage protectionOFP - Overfrequency protection

OVP

OFP

• Peak LRO decreases with higher load/gen ratio• Uncertified inverter causes longer trip time, but similar LRO

G3BK1

Variable load, 0 – 1 pu

Test condition: No grounding transformer (GT), delta connected load, No ground fault

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m9

G3BK1

Variable load, 0 – 1 pu

Scenario 2 – LRO + GFOYg/∆ for grounding

Test condition: w/wo grounding transformer (GT), delta connected load, permanent ground fault

3.37

2.95 2.892.79 2.77

2.562.38

2.22 2.171.99

1.85

2.72

2.29

2.001.90

1.631.49

1

1.5

2

2.5

3

3.5

0 0.2 0.4 0.6 0.8 1

Peak

pha

se v

olta

ge (p

u)

load/Gen ratio

Uncertified inverter

No GT With GT

3.27

2.692.50

2.392.26

2.132.01

1.88 1.86 1.84 1.76

2.52

2.25

2.01

1.7031.518

1.407

1

1.5

2

2.5

3

3.5

0 0.2 0.4 0.6 0.8 1

Peak

pha

se v

olta

ge (p

u)

load/Gen ratio

Certified inverter

No GT With GT

• Uncertified inverter suffers higher overvoltage in this case due to the longer trip time • Grounding transformer helps to mitigate the overvoltage

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m10

G3BK1

1pu load

Scenario 3 – GFO

• Overvoltage caused by ground fault (GF) tends to reduce with higher portion of ground load (Yg connected)

• Grounding transformer (GT) helps with overvoltage mitigation when grounded load is light, but this effects drops with increasing grounded load. Specially, GT slightly increase GFO when 60% loads are grounded

• Further investigation is needed to study the different overvoltage change trend in uncertified inverter, when 40%-80% grounded load and no GT exist

Test condition: w/wo grounding transformer (GT), load/generation ratio = 1, delta ang Yg connected load, permanent ground fault

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2

0% 20% 40% 60% 80% 100%

Peak

pha

se v

olta

ge (p

u)

Percentage of grounded load

Uncertified inverter, no GT Certified inverter, no GT

Uncertified inverter, with GT Certified inverter, with GT

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m11

Comparison between Scenario 1 and 2Scenario 1 (S1): No grounding transformer (GT), delta connected load, No ground fault

Scenario 2 (S2) : w/wo grounding transformer (GT), delta connected load, permanent ground fault

3.27

2.692.50

2.392.26

2.132.01

1.88 1.86 1.84 1.76

2.52

2.252.01

1.701.52 1.41

2.00

1.73 1.69 1.66 1.62 1.59 1.521.34

1.211.10 1.02

0

0.5

1

1.5

2

2.5

3

3.5

0 0.2 0.4 0.6 0.8 1

Peak

pha

se v

olta

ge (p

u)

load/Gen ratio

Certified inverter

S2- No GT S2 - With GT S1 - No GT

• Overvoltage caused by ground fault and load rejection cannot be simply superposed

• Even with grounding transformer, the ground fault overvoltage (GFO) is higher than the load rejection overvoltage (LRO)

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m12

Test Plan Objective

– Characterize the inverters’ dynamic response to voltage changes– Estimate equivalent negative sequence impedance for theoretical

analysis– Identify impact of volt-var function on inverter response

Test scenarios (with and without volt-var) Response to step change of positive sequence grid voltage Response to negative sequence voltage variation

– Magnitude variation– Phase angle variation

Test approach – Lab test of commercial inverters– Controller hardware-in-the-loop (CHIL) with different negative

sequence current control schemes– Will be performed in parallel

Expected outcomes– Validated inverter simulation model with accurate dynamic response– Characteristics of inverter negative sequence impedance when

different control schemes are implemented

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m13

Future Work

Perform lab and CHIL tests per designed test plan Simulation model refinement

– Arrester– Capacitor bank– …. Simulation of other scenarios

– Yg/∆ transformer for grounding– Varying generation before island forms– Inverter with grid-support functions– ….

© 2020 Electric Power Research Institute, Inc. All rights reserved.w w w . e p r i . c o m14

Together…Shaping the Future of Electricity