Clemson University

TigerPrints

All Theses Theses

8-2023

Adaptive Protection Scheme for Microgrids with Seamless

Transition
Jason Sabet
[email protected]

Follow this and additional works at: https://tigerprints.clemson.edu/all_theses

Recommended Citation
Sabet, Jason, "Adaptive Protection Scheme for Microgrids with Seamless Transition" (2023). All Theses.
4086.
https://tigerprints.clemson.edu/all_theses/4086

This Thesis is brought to you for free and open access by the Theses at TigerPrints. It has been accepted for
inclusion in All Theses by an authorized administrator of TigerPrints. For more information, please contact
[email protected].
Adaptive Protection Scheme for Microgrids
with Seamless Transition

A Thesis
Presented to
the Graduate School of
Clemson University

In Partial Fulfillment
of the Requirements for the Degree
Master of Science
Electrical Engineering

by
Jason Sammy Sabet
August 2023

Accepted by:
Dr. Sukumar Brahma, Committee Chair
Dr. Sheikh Jakir Hossain
Dr. Ramtin Hadidi
Abstract

As traditional, carbon-emitting power generation facilities are being retired,

the future of the electric grid is being powered by renewable energy sources and
supported by energy storage. Unlike centralized power generation, renewable energy
can be installed in a distributed or local manner, leading to a restructuring of the
distribution system towards a more modular design and the growth of microgrid
technology.
Microgrids are small, locally fed power systems that supply electricity to a
specific customer base. Utilities are using this technology to enhance the reliability
and resilience of their distribution systems. These utility microgrids can provide
grid support during normal operations and can also operate as a self-sustained island
during grid outages.
With a growing commitment to carbon-neutrality, common forms of modular
clean generation are solar photovoltaic arrays, battery energy storage systems, and
wind turbines. These sources require an inverter to connect to the AC grid, and hence,
are called inverter-based resources (IBRs). Advanced inverter-based microgrids with
seamless transition can switch between grid-connected and island modes without any
customer blinks, a unique feature not found in conventional inverters.
The addition of inverter-based microgrids to distribution systems introduces
new protection challenges. If microgrid generation is providing grid-support, the

ii
multi-source distribution feeder will have bi-directional power flow and bi-directional
fault current in the event of a short-circuit. The inverter-based resource is current-
limited, providing limited fault current in the case of a short circuit. This requires
the distribution protection design to be capable of handling a wider range of fault
currents with bi-directional flow.
To overcome these challenges, protection designs utilize communication to
interchange between two separate relay settings based on the microgrids mode of op-
eration. However, for microgrids equipped with seamless transition, delays in change
of settings and communication delays introduce risk to reliable protection.
By evaluating a Duke Energy microgrid system, this thesis illustrates the prac-
tical risks of seamless transition while using an inverter and the protection deficiencies
that arise. Although conventional inverter design does not allow for seamless transi-
tion, inverter controls are designed on an actively deployed inverter to offer seamless
transition functionality. Considering cost and the protection risks from seamless tran-
sition, an adaptive protection scheme has been designed without the need for change
of settings or communication. The efficacy of the adaptive protection scheme has
been successfully demonstrated through hardware-in-the-loop simulations and relay
tests, which have validated the robustness of the protection design.

iii
Acknowledgments

I would like to express my gratitude to my Duke Energy mentors, Lead En-

gineer Dr. Sheikh Jakir Hossain and Lead Engineer Mr. Thomas Wilson Edward,
P.E., for teaching me how to approach research and problem-solving with a realistic
perspective. I would also like to thank my manager, Dr. Sherif Abdelrazek, and the
Energy Storage and Microgrids Team at Duke Energy for their continuous support
throughout this research project.
I would also like to express my appreciation to Dr. Sukumar Brahma, Dr.
Ramtin Hadidi, and the CUEPRA organization for their encouragement in pursuing
academic research at Clemson University and their efforts to help me grow within
the power and energy field. I would like to thank PhD student Trupal Patel for
his mentorship since my undergraduate studies and for his guidance throughout my
research.

iv
Table of Contents

Title Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv

List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

List of Listings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

1 Introduction and Literature Review . . . . . . . . . . . . . . . . . . 1

1.1 Duke Energy Microgrids and Decarbonization . . . . . . . . . . . . . 2
1.2 Seamless Transition and Limitations . . . . . . . . . . . . . . . . . . 5
1.3 Inverter Fault Limitations and Standards . . . . . . . . . . . . . . . . 6
1.4 Challenges and Literature Review . . . . . . . . . . . . . . . . . . . . 10
1.5 Objective and Contribution of Thesis . . . . . . . . . . . . . . . . . . 13

2 Distribution System Description and Modeling . . . . . . . . . . . 15

2.1 Feeder Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2 Microgrid Description . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3 System Description and Modeling . . . . . . . . . . . . . . . . . . . . 21

3 Implementing Inverter Controls and Seamless Transition . . . . . 31

3.1 Implementing Inverter operation in Grid-Forming Mode while Grid-
Connected . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2 Seamless Transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 Existing Protection Devices . . . . . . . . . . . . . . . . . . . . . . . 55

5 Proposed Adaptive Protection Scheme . . . . . . . . . . . . . . . . 62

5.1 Protection Design Criteria and Assumptions . . . . . . . . . . . . . . 63
5.2 Voltage-Restrained Time Overcurrent Protection . . . . . . . . . . . . 69
5.3 Protection Design Overview . . . . . . . . . . . . . . . . . . . . . . . 71

v
 5.4 Recloser 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.5 Generation Facility Protection in Reverse Direction . . . . . . . . . . 84
5.6 Generation Facility Protection in Forward Direction . . . . . . . . . . 97
5.7 Islanding Recloser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.8 Recloser 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.9 Recloser 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

6 Testing Adaptive Protection Scheme using HIL Simulations . . . 122

6.1 Testing Background and Parameters . . . . . . . . . . . . . . . . . . 123
6.2 Relay Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
6.3 Feeder Protection Test . . . . . . . . . . . . . . . . . . . . . . . . . . 129
6.4 Microgrid Protection Test . . . . . . . . . . . . . . . . . . . . . . . . 139
6.5 Protection Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
A : MATLAB functions for producing protection settings . . . . . . . . 154
B : Custom Protection Logic for 51V Voltage-Restrained Inverse-Time
Overcurrent Element in SEL Relay . . . . . . . . . . . . . . . . . . . 168

vi
List of Tables

2.1 Generation facility export and import restrictions . . . . . . . . . . . 20

2.2 Substation transformer parameters . . . . . . . . . . . . . . . . . . . 22
2.3 System load data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Voltage regulator parameters . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Generation facility transformer parameters . . . . . . . . . . . . . . . 24
2.6 Simulated steady-state voltage during peak load . . . . . . . . . . . . 25
2.7 Simulated steady-state current during peak load . . . . . . . . . . . . 25
2.8 Simulated steady-state voltage during valley load . . . . . . . . . . . 25
2.9 Simulated steady-state current during valley load . . . . . . . . . . . 26

3.1 Generation facility maximum export with PV and BESS . . . . . . . 37

3.2 Generation facility maximum export with BESS . . . . . . . . . . . . 37
3.3 Generation facility maximum import with BESS . . . . . . . . . . . . 37
3.4 BESS step sizes for maximum rated export with PV and BESS . . . . 40
3.5 BESS step sizes for maximum rated export with BESS . . . . . . . . 40
3.6 BESS step sizes for maximum rated import with BESS . . . . . . . . 41

4.1 Existing distribution system forward protection zones with generation

facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.2 Existing distribution system reverse protection zones with generation
facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.3 Existing microgrid system protection . . . . . . . . . . . . . . . . . . 59

5.1 Simulated system conditions for fault study in PSCAD . . . . . . . . 64

5.2 CTI margin between main and backup protection . . . . . . . . . . . 67
5.3 Proposed forward protection zones of distribution system with gener-
ation facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.4 Proposed reverse protection zones of distribution system with genera-
tion facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.5 Proposed forward protection zones of microgrid system . . . . . . . . 75
5.6 Proposed reverse protection zones of microgrid system . . . . . . . . . 75
5.7 Fault data recorded at 80K fuse for simulated faults within 80K fuse
zone during grid connection . . . . . . . . . . . . . . . . . . . . . . . 76
5.8 Fault data recorded at 80K fuse for simulated faults within 80K fuse
zone during microgrid operation . . . . . . . . . . . . . . . . . . . . . 77

vii
5.9 Fault data recorded at recloser 3 for simulated faults within 80K fuse
zone during grid connection . . . . . . . . . . . . . . . . . . . . . . . 77
5.10 Fault data recorded at recloser 3 for simulated faults within 80K fuse
zone during microgrid operation . . . . . . . . . . . . . . . . . . . . . 77
5.11 Recloser 3 protection settings . . . . . . . . . . . . . . . . . . . . . . 79
5.12 OTs of 80K fuse backed by recloser 3 for maximum and minimum faults
within the main protection zonea . . . . . . . . . . . . . . . . . . . . 80
5.13 Fault data recorded at recloser 3 for simulated faults within recloser 3
zone during grid connection . . . . . . . . . . . . . . . . . . . . . . . 86
5.14 Fault data recorded at recloser 3 for simulated faults within recloser 3
zone during microgrid operation . . . . . . . . . . . . . . . . . . . . . 87
5.15 Fault data recorded at PCC recloser for simulated faults within recloser
3 zone during grid connection . . . . . . . . . . . . . . . . . . . . . . 87
5.16 Fault data recorded at PCC recloser for simulated faults within recloser
3 zone during microgrid operation . . . . . . . . . . . . . . . . . . . . 87
5.17 Coordination of generation facility protection in reverse direction for
maximum and minimum faults within main protection zone . . . . . 88
5.18 Generation facility reverse protection settings . . . . . . . . . . . . . 88
5.19 OTs of generation facility in reverse direction for maximum and mini-
mum faults within the main protection zonea . . . . . . . . . . . . . . 89
5.20 Coordination of generation facility protection in forward direction for
maximum and minimum faults within main protection zone . . . . . 98
5.21 Generation facility forward protection settings . . . . . . . . . . . . . 99
5.22 OTs of generation facility in forward direction for maximum and min-
imum faults within the main protection zone . . . . . . . . . . . . . . 100
5.23 Fault data recorded at islanding recloser for simulated faults within
recloser 3 zone during grid connection . . . . . . . . . . . . . . . . . . 106
5.24 Coordination of islanding recloser in forward direction for maximum
and minimum faults . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.25 Islanding recloser forward protection settings . . . . . . . . . . . . . . 107
5.26 OTs of islanding recloser and coordinated devices in forward direction
for maximum and minimum faultsa . . . . . . . . . . . . . . . . . . . 107
5.27 Coordination of recloser 2 for maximum and minimum faults . . . . . 112
5.28 Recloser 2 protection settings . . . . . . . . . . . . . . . . . . . . . . 113
5.29 OTs of recloser 2 and coordinated devices for maximum and minimum
faultsa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.30 Coordination of recloser 1 for maximum and minimum faults . . . . . 117
5.31 Recloser 1 protection settings . . . . . . . . . . . . . . . . . . . . . . 117
5.32 OTs of recloser 1 and coordinated devices for maximum and minimum
faultsa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

6.1 Fault current limiter setpoint for BESS inverter based on fault type . 124

viii
6.2 Comparison of PSCAD and RSCAD steady-state currents during mi-
crogrid operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
6.3 Comparison of PSCAD and RSCAD steady-state voltages during mi-
crogrid operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
6.4 Comparison of PSCAD and RSCAD fault currents during microgrid
operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
6.5 Undervoltage ride-through settings for islanding recloser . . . . . . . 127
6.6 Overvoltage ride-through settings for islanding recloser . . . . . . . . 127
6.7 Underfrequency ride-through settings for islanding recloser . . . . . . 128
6.8 Overfrequency ride-through settings for islanding recloser . . . . . . . 128
6.9 Main and backup protection for grid-connected faults in forward direction129
6.10 Main and backup protection for grid-connected faults in reverse direction131
6.11 Designed and experimental OTs during grid connectiona . . . . . . . 135
6.12 Designed and experimental OTs of recloser 3 during grid connectiona 136
6.13 Designed and experimental OTs without the generation facilitya . . . 138
6.14 Designed and experimental OTs of recloser 3 without the generation
facilitya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
6.15 Main and backup protection for microgrid faults . . . . . . . . . . . . 139
6.16 Designed and experimental OTs during microgrid operationa . . . . . 141
6.17 Designed and experimental OTs of recloser 3 during microgrid operationa 142

ix
List of Figures

1.1 Projected generation capacity installations in 2023 [2] . . . . . . . . . 2

1.2 Duke Energy Hot Springs Microgrid [7] . . . . . . . . . . . . . . . . . 4
1.3 Comparison between traditional fault contribution of synchronous gen-
eration and inverters [7] . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Category I DER voltage ride-through [9] . . . . . . . . . . . . . . . . 8
1.5 Category II DER voltage ride-through [9] . . . . . . . . . . . . . . . . 8
1.6 Category III DER voltage ride-through [9] . . . . . . . . . . . . . . . 9
1.7 DER frequency ride-through [9] . . . . . . . . . . . . . . . . . . . . . 9
1.8 Delays of communication-based protection with settings changes . . . 12

2.1 Simplified oneline diagram of distribution system . . . . . . . . . . . 16

2.2 Detailed diagram of distribution system . . . . . . . . . . . . . . . . . 17
2.3 Microgrid generation facility . . . . . . . . . . . . . . . . . . . . . . . 18
2.4 PSCAD tap changer component [27] . . . . . . . . . . . . . . . . . . 23
2.5 PSCAD PV source component [24] . . . . . . . . . . . . . . . . . . . 24
2.6 Inverter controls from Clemson University inverter model [29] . . . . 28
2.7 Current limiter controls from Clemson University inverter model [29] 29

3.1 Inverter reference inputs based on grid status . . . . . . . . . . . . . 34

3.2 Inverter real power export and import in GFM while grid-connected . 35
3.3 Maximum rated export of 5.4 MW during peak load with no BESS
ramp rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.4 Maximum rated import of 6.9 MW during valley load with no BESS
ramp rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.5 Example of step ramp with maximum power import during peak load 42
3.6 Microgrid voltage during planned islanding . . . . . . . . . . . . . . . 44
3.7 Microgrid voltage during unplanned islanding . . . . . . . . . . . . . 46
3.8 Microgrid frequency during unplanned islanding . . . . . . . . . . . . 46
3.9 Current across islanding recloser during unplanned islanding . . . . . 47
3.10 Synchronization check . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.11 Utilizing inverter frequency to reduce phase difference between grid
and island . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.12 Inverter synchronization controls using inverter voltage and frequency
reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

x
3.13 Microgrid voltage during grid synchronization . . . . . . . . . . . . . 53
3.14 Microgrid frequency during grid synchronization . . . . . . . . . . . . 54

4.1 Existing distribution system protection zones with the generation facility 57
4.2 Existing microgrid system protection zones . . . . . . . . . . . . . . . 60

5.1 PV inverter fault response for 3PH fault at PCC . . . . . . . . . . . . 65

5.2 Voltage-restrained time overcurrent characteristic . . . . . . . . . . . 70
5.3 Dynamic response of 51V protection curve as a function of measured
voltage drop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.4 Proposed distribution system protection zones with the generation facility 72
5.5 Proposed microgrid system protection zones . . . . . . . . . . . . . . 73
5.6 Recloser 3 and 80K fuse coordination during grid connection . . . . . 78
5.7 Recloser 3 extended protection zone during microgrid operation . . . 79
5.8 Recloser 3 phase coordination with 80K fuse for grid faults . . . . . . 82
5.9 Recloser 3 ground coordination with 80K fuse for grid faults . . . . . 83
5.10 Recloser 3 phase operation for island faults . . . . . . . . . . . . . . . 83
5.11 Recloser 3 ground operation for island faults . . . . . . . . . . . . . . 84
5.12 Proposed reverse protection zones with the generation facility . . . . 85
5.13 PCC recloser phase coordination with recloser 3 for grid faults . . . . 92
5.14 PCC recloser phase coordination with recloser 3 for island faults . . . 92
5.15 PCC recloser ground coordination with recloser 3 for grid faults . . . 93
5.16 PCC recloser ground coordination with recloser 3 for island faults . . 93
5.17 PCC relay phase coordination with PCC recloser . . . . . . . . . . . 94
5.18 PCC relay ground coordination with PCC recloser . . . . . . . . . . . 94
5.19 BESS relay phase coordination with 65 BAY-O-NET fuse . . . . . . . 95
5.20 BESS relay ground coordination with 65 BAY-O-NET fuse . . . . . . 95
5.21 PV relay phase operation . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.22 PV relay ground operation . . . . . . . . . . . . . . . . . . . . . . . . 96
5.23 Proposed forward protection zones with the generation facility . . . . 98
5.24 AUX relay phase coordination with 12T fuse . . . . . . . . . . . . . . 101
5.25 AUX relay ground coordination with 12T fuse . . . . . . . . . . . . . 102
5.26 Proposed protection zones with islanding recloser . . . . . . . . . . . 103
5.27 Islanding recloser reverse settings . . . . . . . . . . . . . . . . . . . . 104
5.28 Islanding recloser phase coordination with 65KS fuse . . . . . . . . . 108
5.29 Islanding recloser ground coordination with 65KS fuse . . . . . . . . . 109
5.30 Islanding recloser phase coordination with recloser 3 . . . . . . . . . . 109
5.31 Islanding recloser ground coordination with recloser 3 . . . . . . . . . 110
5.32 Islanding recloser phase coordination with PCC recloser forward settings110
5.33 Islanding recloser ground coordination with PCC recloser forward settings111
5.34 Proposed protection zones with recloser 2 . . . . . . . . . . . . . . . . 112
5.35 Recloser 2 phase coordination with 80KS fuse . . . . . . . . . . . . . 114
5.36 Recloser 2 ground coordination with 80KS fuse . . . . . . . . . . . . . 114

xi
5.37 Recloser 2 phase coordination with islanding recloser . . . . . . . . . 115
5.38 Recloser 2 ground coordination with islanding recloser . . . . . . . . . 115
5.39 Proposed protection zones with recloser 1 . . . . . . . . . . . . . . . . 116
5.40 Recloser 1 phase coordination with 100KS fuse . . . . . . . . . . . . . 119
5.41 Recloser 1 ground coordination with 100KS fuse . . . . . . . . . . . . 119
5.42 Recloser 1 phase coordination with recloser 2 . . . . . . . . . . . . . . 120
5.43 Recloser 1 ground coordination with recloser 2 . . . . . . . . . . . . . 120
5.44 Recloser 1 phase coordination with recloser 4 . . . . . . . . . . . . . . 121
5.45 Recloser 1 ground coordination with recloser 4 . . . . . . . . . . . . . 121

6.1 RSCAD simulated faults during grid connection . . . . . . . . . . . . 130

6.2 Operation of Recloser 3 slow curve for AG fault within 80K fuse zone 132
6.3 Operation of Recloser 3 slow curve for CAG fault within 80K fuse zone 133
6.4 Operation of Recloser 3 slow curve for BC fault within 80K fuse zone 133
6.5 Operation of Recloser 3 slow curve for ABC fault within 80K fuse zone 134
6.6 RSCAD simulated faults without generation facility . . . . . . . . . . 137
6.7 RSCAD simulated faults during microgrid operation . . . . . . . . . . 140
6.8 Measurements at recloser 3 during induction motor startup . . . . . . 144
6.9 Measurements at recloser 3 during capacitor energization . . . . . . . 145

xii
Nomenclature

DC - Direct Current
AC - Alternating Current
MG - Microgrid
BESS - Battery Energy-Storage System
PV - Photovoltaic
AU X - Auxiliary
GF L - Grid-Following Mode
GF M - Grid-Forming Mode
DER - Distributed Energy Resources
IBR - Inverter Based Resources
P CC - Point of Common Coupling
RV C - Rapid Voltage Change
T CC - Time Current Curves
TD - Time Dial
M OC - Minimum Operating Current
OT - Operating Time
IN ST - Instantaneous
SHB - Second-Harmonic Blocking
P SCAD - Power Systems Computer Aided Design
RT DS - Real Time Digital Simulator
RSCAD - Real-Time Simulator Computer Aided Design
HIL - Hardware-in-the-loop

xiii
Chapter 1

Introduction and Literature

Review

Dating back to the Industrial Revolution, fossil fuels have been the domi-
nant source of energy worldwide. Oil, coal, and gas have long powered homes and
technological advancements, but they also contribute significantly to greenhouse gas
emissions, air pollution, and the risk of oil spills. Each year, this has led to millions
of premature deaths linked to the use of air-polluting fossil fuels [1].
In response, there has been a growing push for the adoption of clean, renewable
energy sources. Governments and organizations around the world have been investing
in the development of renewable energy technologies, such as hydropower, wind, and
solar. These renewable energy resources offer a sustainable alternative to fossil fuels as
they do not emit greenhouse gases. With the recognition of environmental concerns,
Fig. 1.1 demonstrates how more than 80% of the installed generation capacity within
the next year will be dominated by renewable energy resources [2].
The U.S. Energy Information Administration projects that more than half of
the new generation capacity added in 2023 will come from solar power and a quarter

1
 Figure 1.1: Projected generation capacity installations in 2023 [2]

from energy storage and wind [2]. In fact, projections forecast that renewable energy
capacity in the United States will increase by more than 280 GW between 2022 and
2027, with significant contributions from solar and wind energy [3]. Following this
trend, Duke Energy has set carbon goals and a plan to use this renewable generation
to increase reliability through microgrids.

1.1 Duke Energy Microgrids and Decarbonization

Established over a century ago, Duke Energy is a major utility provider in the
United States, serving seven states with reliable energy [4]. As a proponent of cleaner
energy, Duke Energy has committed to reducing 50% of its carbon emissions from
electrical generation by 2030 and to reach complete carbon neutrality by 2050. With
plans to generate 30 GW of renewable energy by 2035, the energy mix of the power
grid will increasingly rely on inverter-based resources (IBRs) [5]. Inverters are used
to connect direct current (DC) renewable resources, such as solar, wind, and battery

2
energy storage systems (BESS), with the alternating current (AC) power grid.
The increasing utilization of IBRs has motivated Duke Energy to strive for a
more intelligent and robust power grid. As opposed to the conventional centralized
generation system, IBRs can be decentralized and local, making them suitable for
microgrid implementation. According to the Department of Energy, microgrids are a
group of interconnected loads and distributed energy resources within clearly defined
electrical boundaries that acts as a single controllable entity with respect to the grid.
A microgrid can connect and disconnect from the grid to enable it to operate in both
grid-connected or island-mode [6]. These microgrids can both enhance the perfor-
mance of the overall grid and operate as an island during grid outages. Duke Energy
is advancing the implementation of microgrids by integrating IBRs into its existing
distribution network, prioritizing systems with regular grid service interruptions such
as the Hot Springs microgrid.

1.1.1 Hot Springs Microgrid

Duke Energy has applied microgrid technology to the town of Hot Springs
in order to provide reliable, clean energy to its local community. The Hot Springs
microgrid is a unique and innovative project, consisting of two IBRs, a 2 MW solar
farm, and a 4.4 MW lithium-ion BESS. This combination of resources allows the
microgrid to provide resilient energy to over 500 customers in the community. As
seen in Fig. 1.2, the microgrid sustains an isolated community by making use of
DERs [7].
The Hot Springs microgrid has the ability to operate both in connection with
the larger grid and as a standalone, self-sustained island. In grid-following mode
(GFL), it provides benefits such as voltage and frequency support and additional

3
 Figure 1.2: Duke Energy Hot Springs Microgrid [7]

capacity during peak times. In grid-forming mode (GFM), the microgrid can black
start the entire town’s load using only its renewable generation resources [7].
The Hot Springs microgrid serves as a practical example of how microgrid tech-
nology can improve grid reliability and make effective use of IBRs. It demonstrates
how microgrids can provide localized, resilient energy solutions to communities, es-
pecially those that are remote and have limited backup power options. Advanced
microgrids are equipped with a sophisticated seamless transition feature that ensures
uninterrupted power supply to customers, preventing any power blinks or interrup-
tions.

4
1.2 Seamless Transition and Limitations

As microgrid technology is still in its early stages, utility microgrids fed by

IBRs often undergo a blackstart during mode transitions. This results in customers
experiencing a temporary loss of power during a grid disturbance, which is then re-
stored by the microgrid generation facility. Similarly, customers also experience a loss
of power when the microgrid reconnects to the main grid. This creates significant dif-
ficulties for microgrids serving vital facilities such as hospitals, schools, government
buildings, prisons, and data centers, especially with the growing reliance on electron-
ics.
However, advanced microgrids have a feature known as seamless transition,
which enables a smooth, uninterrupted switch between grid-connected and island
operation. With this feature, customers within the microgrid system won’t experience
any power outages, especially during grid faults when the main grid source is lost.
By eliminating the need for blackstart, seamless transition reduces energy waste and
enhances reliability and resiliency.
To achieve a seamless transition, advanced inverter technology is necessary.
This technology must be capable of switching effortlessly between GFL and GFM
controls. Regrettably, many conventional inverters employed in microgrid applica-
tions often lack the capacity to transition seamlessly between the two control modes.
This is due to their reliance on a fixed reference voltage and frequency during GFM,
which can conflict with the grid voltage and frequency during a seamless transition
and cause disturbances. Hence, the inverter requires a momentary standby when
switching between modes, and failing to do so can potentially damage the inverter
or the grid. Therefore, unless the inverter is no longer required to switch between
modes, it often limits many microgrids to blackstart transitions only.

5
1.3 Inverter Fault Limitations and Standards

A power system fault or disturbance causes a disruption in the normal power

flow in the system. Faults can be caused by various factors, including natural disas-
ters, severe storms, and equipment failure. The regions that are more prone to grid
faults are those that experience harsh weather conditions, rugged terrain, and prox-
imity to water. There are two primary categories of faults: phase faults and ground
faults. Phase faults happen when phase conductors short-circuit, and they can be
two-phase or three-phase. Ground faults occur when phase conductors come into
contact with the ground, and they can be single-phase, two-phase, or three-phase.
Synchronous generators can produce up to six times their rated current during
a fault, while inverters are restricted to marginally more than their rated current,
as seen in Fig. 1.3 [8]. Inverters have distinctive fault characteristics, including
limited positive sequence current, limited to no negative sequence current, and no
zero-sequence current, making it a challenge to design protection for microgrids fed
by IBRs. Unfortunately, the response of inverters varies based on the manufacturer
and control scheme used. However, as a general requirement for manufacturers, they
all must conventionally conform with standards. For distribution-tied microgrids,
IEEE 1547-2018 is recognized as the accepted standard for inverters.
IEEE 1547-2018 is a standard that regulates the connection of DERs to the
power system. It establishes specific voltage and frequency standards that invert-
ers must comply with during faults, based on their categorization: Category I for
essential bulk system requirements, Category II for full coordination with all bulk
system stability and reliability needs, and Category III for complete reliability and
power quality needs of the bulk and distribution systems. As the penetration of re-
newable energy sources increases, it is expected that there will be a trend towards

6
Figure 1.3: Comparison between traditional fault contribution of synchronous gener-
ation and inverters [7]

higher category identifications. Each category outlines specific DER voltage and fre-
quency ride-through limits during fault and the range of allowable trip times. Figs.
1.4, 1.5, and 1.6 define the voltage ride-through requirements for a grid disturbance,
meanwhile, Fig. 1.7 defines the frequency ride-through requirements for a grid dis-
turbance [9].
As seen in Figs. 1.4, 1.5, and 1.6, an increase in the category identification
of IEEE 1547-2018 results in the mandatory operation region growing in terms of
voltage range and voltage ride-through time. Meanwhile, Fig. 1.7 depicts mandatory
overfrequency operation between 61.2 Hz to 61.8 Hz and mandatory underfrequency

7
Figure 1.4: Category I DER voltage ride-through [9]

Figure 1.5: Category II DER voltage ride-through [9]

8
Figure 1.6: Category III DER voltage ride-through [9]

Figure 1.7: DER frequency ride-through [9]

9
operation between 58.8 Hz to 57.0 Hz for all categories, with a minimum required
ride-through time of 300 seconds [9]. As a result, with increased DER penetration, the
standard requires IBRs to ride-through grid disturbances instead of instantaneously
tripping offline for any grid disturbance, avoiding voltage instability.
Distribution power systems use protective devices like relays and fuses to detect
and respond to faults. While mechanical devices like fuses have only one response type
and rely only on phase current, digital relays can have multiple settings with various
response types and even differentiate their operation for phase and ground faults,
resulting in fast response times. Digital relays often have separate configurations to
handle the limited fault currents caused by IBRs, while mechanical protective devices
sized for grid faults are unable to respond to inverter faults altogether. However, the
seamless transition of microgrids presents new challenges such as communication and
relay setting delays, making it necessary to develop more sophisticated protection
designs.

1.4 Challenges and Literature Review

System protection is essential for the reliable operation of any power distribu-
tion system. It must be able to operate dependably under any fault condition within
its zone, coordinate with adjacent protection, and not operate during non-fault dis-
turbances, thus offering security. With the addition of a microgrid, customers have
power availability regardless of grid conditions, but complexity is added to the distri-
bution system. An effective microgrid protection system must focus on fast, reliable
operation in the event of a disturbance, with consideration for cost, selectivity, and
safety.
To achieve a reliable microgrid protection scheme, the system must be able to

10
respond to both distribution system faults and microgrid faults, and the microgrid
should be able to isolate from the faulted grid as soon as possible. Therefore, during
a fault condition, the protection system must be capable of adapting to bidirectional
fault current and changes in fault current magnitude. Various surveys on microgrid
protection [10, 11, 12] and an IEEE PES technical report [13] shed light on the current
state of microgrid protection systems. These schemes employ a range of techniques,
including overcurrent, voltage, harmonic content-based, distance, differential, and
current traveling wave protection. It is worth noting that these protection schemes
rely on the existence of a communication infrastructure. Moreover, researchers have
put forth adaptive protection schemes that leverage these diverse techniques to en-
hance microgrid protection.
Patel et al. [14] propose an adaptive protection scheme that utilizes bidirec-
tional protective devices but requires different settings for different operating modes.
Gupta et al. [15] propose an algorithm that utilizes a settings lookup table for sys-
tem relays that is referenced in the event of a fault. Peiris et al. [16] replace fault
lookup tables by using a microgrid central processing unit to instantly calculate and
determine tripping current for any mode of operation. Wheeler et al. [17] utilize a
combination of adaptive settings based on communication and differential relays to
overcome the change in short-circuit contributions.
All of these proposed adaptive protection schemes rely on a fast and reliable
communication infrastructure. Traditionally, in utility distribution systems, a basic
communication infrastructure is utilized outside of the substation, such as cellular.
According to Verizon, latency delays can reach up to 45 milliseconds in the North
America, with successful packet delivery dependent on signal quality and network con-
gestion [18]. A distribution system with an upgraded communication infrastructure
may use faster, wired communication such as IEC 61850 GOOSE communication,

11
 Figure 1.8: Delays of communication-based protection with settings changes

with latency delays of up to 4 milliseconds. However, this is a much more costly

alternative [19]. Additionally, some of these proposed solutions also require a change
of relay group settings, which requires some commercial relays to be disabled for more
than 3 seconds during an active settings group change [20]. These communication
and relay delays can be seen depicted in the Fig. 1.8. For communication-based
protection schemes with settings changes, microgrids with seamless transition will
inevitably be left unprotected for some period of time after switching modes of oper-
ation, compromising protection performance.
Therefore, research is ongoing to find alternative techniques that do not re-
quire communication infrastructure or relay settings change. Nikkhajoei et al. [21]
suggest a protection scheme for microgrids that employs a static switch to isolate
the microgrid if the grid supplies any fault current, which requires a single setting

12
based on IBR faults. Nevertheless, the protection scheme assumes that the micro-
grid generation is always connected, which is not necessarily true. Ndjaba et al. [22]
discuss and compare various adaptive protection techniques that are communication
and non-communication based. As a non-communication based technique, [22] men-
tions voltage-restrained overcurrent protection and demonstrates the scheme working
in PSCAD on a simple two-bus network. However, in [22], the voltage-restrained
overcurrent technique does not conform to any standard relay operating practice and
the testing was not done on a realistic system.
Overall, many adaptive protection methods have been proposed in the field of
research. However, few have practically implemented and tested an adaptive protec-
tion scheme that is not reliant on communication and cost-effective.

1.5 Objective and Contribution of Thesis

The objective of this thesis is to design an adaptive protection scheme that

enables seamless transition on a Duke Energy microgrid system without relying on
communication or change in settings.
Currently, the microgrid system is designed for blackstart and is not standard-
capable of seamless transition due to the limitations of the inverter. To overcome this
limitation, this research explores the implementation of elemental controls on the
inverter in PSCAD simulation to ensure stable inverter operation in GFM while con-
nected to the grid and grid synchronization. This enables the microgrid to be islanded
at any time as the inverter is already in GFM, and also allows it to synchronize with
the grid after a grid disturbance is cleared. The successful demonstration of these
controls proves that seamless transition is possible with the current conventional in-
verter.

13
 Next, the existing protection system will be analyzed in the presence of faults in
different modes of operation. The insights gathered from this analysis will inform the
development and testing of an adaptive protection scheme on legacy relays using HIL.
The scheme will leverage bi-directional, voltage supervision, and voltage-restrained
overcurrent protective elements to enhance the microgrid’s ability to adapt to dynamic
changes and protect against all system faults.

14
Chapter 2

Distribution System Description

and Modeling

2.1 Feeder Description

The Duke Energy distribution system used for this work, consists of a 22.86
kV feeder that incorporates a microgrid generation facility. The feeder has a peak
load of around 11.1 MW and a valley load of approximately 2 MW. Fig. 2.1 shows a
simplified oneline diagram of the distribution system, while Fig. 2.2 shows a bird’s-eye
topology of the entire feeder.
The distribution feeder is supplied by a substation that employs a ∆/YG
transformer to step down the subtransmission voltage from 115 kV to 22.86 kV. The
transformer’s delta winding on the primary side provides balancing capability, while
the wye-grounded winding on the secondary side provides a source for ground fault
current. The substation overcurrent relay, which controls the feeder circuit breaker,
is the last protective device before the distribution line exits the substation.
The distribution system includes three 1.2 MVAR capacitor banks and three

15
16
Figure 2.1: Simplified oneline diagram of distribution system
 Figure 2.2: Detailed diagram of distribution system

voltage regulators, located towards the end of the feeder. These capacitor banks
provide power factor correction and voltage support, while the voltage regulators
step up the medium voltage along the system by utilizing an autotransformer. The
voltage-supporting devices keep the voltage range within the range of 0.95 pu to 1.05
pu, as per the ANSI standard [23].
While Chapter 4 of this thesis delves deeper into the system protection, Fig.
2.1 illustrates that electronic and hydraulic reclosers are employed to safeguard the
feeder backbone.

17
2.2 Microgrid Description

The microgrid is powered by a renewable, IBR generation facility that con-

sists of both BESS and PV generation. The microgrid load is 5.1 MW during peak
conditions and 0.96 MW during valley conditions. Prior to the study conducted in
this thesis, Duke Energy had sized the generation facility based on load restoration,
and developed a protection scheme for the system. They also conducted a study to
blackstart the microgrid in GFM, and export and import power in GFL [24, 25]. The
design of the generation facility can be seen in the Fig. 2.3.

Figure 2.3: Microgrid generation facility

18
 The BESS capacity is divided into two units, each with a capacity of 3.45 MW.
Both inverters are IEEE 1547-2018 compliant and follow category II specifications.
Each BESS inverter is connected to a ∆/YG medium voltage transformer that steps
up the voltage to 22.86 kV. When the generation facility solely powers the microgrid,
the transformer is solidly grounded, providing a source for ground fault current. How-
ever, when the facility is connected to the grid, a 40 Ω neutral grounding resistance is
used to limit ground current contributions. This resistance was sized by the previous
Duke Energy study to reduce contribution to grid ground faults, while maintaining
effective grounding.
During grid-connected ground faults, the absence of a neutral grounding re-
sistor can lead to desensitization of relays upstream to the microgrid, while causing
oversensitivity in relays downstream. The term ”upstream” refers to the direction
towards the substation, whereas ”downstream” indicates the direction towards the
end of the feeder. Upstream relays become desensitized because the generation fa-
cility contributes a significant portion to the ground faults, thereby reducing the
contribution from the substation. On the other hand, downstream relays become
excessively sensitive due to the combined substation and generation facility ground
current for ground faults. This situation results in a larger disparity in ground fault
current between grid-connected and disconnected scenarios. Therefore, a grounding
resistance was used to limit ground fault contributions by the generation facility.
However, introducing a higher resistance in the neutral grounding carries the risk of
ground-fault overvoltage. To mitigate this risk, the previous study determined that
the resistance should not exceed an 80% grounding coefficient or temporary overvolt-
age levels above 1.386 pu, as specified in IEEE Std. C62.92.1-2019 [26]. Taking these
regulations into account, Duke Energy opted for a 40 Ω neutral grounding resistor
during grid-connected operation, as it was the largest resistor that ensured the system

19
remained effectively grounded, as stated in the SIS report.
The PV solar farm has a maximum rated capacity of 3.25 MW, but account-
ing for diverse solar irradiance and solar panel degradation, the solar inverters have
a maximum rated capacity of 2.1 MW. The solar inverter is also IEEE 1547-2018
compliant and follows category I specifications. The PV inverter is connected to a
YG/YG medium voltage transformer that steps up the voltage to 22.86 kV.
The generation facility is completed with the addition of an auxiliary load,
which draws 50 kW per phase to support the cooling of battery containers, power
system controllers, and power system meters. The voltage is stepped down from
22.86 kV to 480 V through a YG/YG transformer. The entire system, including the
two BESS units, the PV unit, and the auxiliary load, is connected to the grid via a
four-way switchgear.
To ensure the proper functioning of the microgrid, Duke Energy has established
export and import limitations during grid-connected operation based on load demand,
voltage limitations, and thermal analysis. The limitations are outlined in Table 2.1 for
peak and valley loading. The export limitation is designed to prevent the generation
facility from exceeding regulator 3 seen in Fig. 2.1, to avoid conflicting with regulator
operation. Similarly, import limitations are set to prevent the distribution lines from
being overloaded and exceeding its thermal limit.

Table 2.1: Generation facility export and import restrictions

Mode Max Generation (BESS+PV) [MW] Max Charge (BESS) [MW]

Peak Load 5.4 3
Valley Load 3 6.9

Due to the BESS inverter’s incapability to switch between GFL and GFM with-
out entering a standby state, the microgrid must strictly disconnect and blackstart
during grid disturbances. For grid reconnection, microgrid customers will experience

20
a momentary outage before reconnecting. To ensure healthy voltage conditions dur-
ing a the microgrid blackstart, Duke Energy has added a load break switch within the
microgrid, which can be seen in Fig. 2.1. This switch enables a two-step blackstart
instead of a single-step blackstart, to avoid undervoltage concerns during peak load.
All reclosers and relays within the microgrid are also equipped with two separate
settings for grid-connected and islanded modes of operation. The former setting ac-
counts for high system fault currents, while the latter setting accounts for low inverter
fault currents [25].
This thesis seeks to enhance the design set by Duke Energy to improve micro-
grid resiliency and power reliability. In Chapter 3, the thesis aims to showcase the
operation of the inverter in GFM while connected to the grid, enabling a smooth tran-
sition for microgrid operation. However, this seamless transition can pose a challenge
for the current protection design. To overcome this obstacle, Chapter 5 proposes an
adaptive protection scheme that does not require communication. To achieve these
objectives, it is crucial to utilize an accurate system model to study the system in its
entirety.

2.3 System Description and Modeling

Duke Energy chose to use CYME simulation, a phasor-domain platform, to

model the original distribution system. However, for the microgrid study, the model
which was converted to PSCAD/EMTDC, a time-domain environment. PSCAD was
chosen due to its ability to incorporate vendor-specific inverter models into the gen-
eration facility, enabling simulations with greater accuracy and the identification of
potential issues during operation [24].
The CYME model was successfully transformed to PSCAD by extracting all

21
essential system data from CYME, including conductor type, sequence impedance
data between nodes, equipment specifications, and load data. The resulting PSCAD
model depicted the original distribution system, encompassing nearly 1000 nodes.
The construction of the PSCAD model involved several stages, starting with
the modeling of the 115 kV source and the ∆/YG substation transformer. In order to
create an accurate representation of the source, the impedance data previously used in
CYME was employed. The substation transformer, on the other hand, was modeled
using three single-phase transformers, and the relevant parameters are summarized
in Table 2.2.

Table 2.2: Substation transformer parameters

Type Transformer Specification

Nominal Capacity [MVA] 50
Primary Winding Configuration ∆
Secondary Winding Configuration YG
Primary Winding Voltage [kV] 115
Secondary Winding Voltage [kV] 22.86
Leakage Reactance 8%
Copper Losses 0.1%

To replicate the CYME distribution lines’ modeling, the ”Coupled PI Section”

component found in the PSCAD library was utilized and programmed with the ex-
ported CYME data for each line span. Accordingly, the PSCAD model also accounted
for mutual coupling. The ”fixed load” component was used to model the distribution
loads as constant power loads. The peak and valley loads, measured both from the
substation and from the generation facility, are shown in Table 2.3.

Table 2.3: System load data

Measurement Location Peak Load [MW] Valley Load [MW]

Substation 11.1 2
Generation Facility 5.1 0.96

22
 Available PSCAD libraries were utilized to incorporate capacitor banks and
voltage regulators into the model. The capacitor banks were connected in parallel at
their rated capacity of 1.2 MVAR, while each voltage regulator was modeled as three
single-phase transformers with tap changers and sharing the same parameters. The
tap changers were controlled by a ”Tap Changer” component from a PSCAD library,
initialized with the nominal system voltage, initial tap position, tap delay, desired
voltage setpoint, and voltage step per tap, as shown in Fig. 2.4 [27]. Detailed voltage
regulator specifications can be found in Table 2.4.

Figure 2.4: PSCAD tap changer component [27]

Table 2.4: Voltage regulator parameters

Type Voltage Regulator Specification

Nominal Capacity [MVA] 1
Nominal Voltage [kV] 22.86
Voltage Regulator Setting [pu] 1.04
Total Taps 32
Initial Tap Position 14
Tap Step [pu] 0.00625

Finally, blackbox inverter models and medium voltage transformers were used

23
to connect the IBRs with the grid, accurately modeling the generation facility. To
model the two BESS sources, single-phase DC voltage sources at 1.3 kV were used.
The PV generation, on the other hand, was simulated using a PSCAD ”Photovoltaic
Source” component, as illustrated in Fig. 2.5, which was programmed to comply with
the vendor’s rated specifications and output up to 3.25 MW [28].

Figure 2.5: PSCAD PV source component [24]

Table 2.5: Generation facility transformer parameters

Type BESS PV AUX

Transformer Transformer Transformer
Nominal Capacity [MVA] 3.45 2.1 0.5
Primary Winding Configuration ∆ YG YG
Secondary Winding Configuration YG YG YG
Primary Winding Voltage [kV] 0.6 0.6 0.48
Secondary Winding Voltage [kV] 22.86 22.86 22.86
Leakage Reactance 6% 6% 5%
Copper Losses 1% 1% 1%

To model the BESS transformers, Duke Energy employed a three-phase ∆/YG

transformer with a switch on the transformer neutral. This allowed for an alternate
between a 40 Ω neutral grounding resistance or a solid ground connection based on the
mode of operation. The PV transformer was modeled using a three-phase YG/YG
transformer. Both transformers were connected in series with the AC side of the

24
inverter [24]. Furthermore, an auxiliary three-phase YG/YG transformer was con-
nected to support all loads within the generation facility. The technical specifications
for these transformers are summarized in Table 2.5.
Table 2.6: Simulated steady-state voltage during peak load

Percent Distance Phase A Phase B Phase C

from Substation LG Voltage [kV] LG Voltage [kV] LG Voltage [kV]
0% 13.64 13.64 13.64
25% 13.38 13.49 13.21
50% 13.63 13.7 13.62
75% 13.37 13.6 13.39
100% 13.32 13.44 13.3

Table 2.7: Simulated steady-state current during peak load

Percent Distance Phase A Phase B Phase C

from Substation Current [A] Current [A] Current [A]
0% 265 272 282
25% 225 227 252
50% 157 138 152
75% 144 131 139

Table 2.8: Simulated steady-state voltage during valley load

Percent Distance Phase A Phase B Phase C

from Substation LG Voltage [kV] LG Voltage [kV] LG Voltage [kV]
0% 13.64 13.64 13.64
25% 13.59 13.65 13.57
50% 13.68 13.72 13.71
75% 13.72 13.82 13.78
100% 13.63 13.67 13.64

Model validation was conducted by comparing the PSCAD model results with
the CYME power flow results, without the generation facility. The validation showed
that the PSCAD model results had a maximum deviation of 2% from the CYME
power flow results. Tables 2.6, 2.7, 2.8, and 2.9 show the final steady-state voltages
and currents for peak and valley load without the generation facility.

25
 Table 2.9: Simulated steady-state current during valley load

Percent Distance Phase A Phase B Phase C

from Substation Current [A] Current [A] Current [A]
0% 48 53 47
25% 43 47 42
50% 40 44 38
75% 38 39 33

2.3.1 Inverter Model

The microgrid generation facility relies on the inverter as a crucial component

to connect DC renewable generation with the AC grid. To model the generation
facility accurately, vendor-provided blackbox inverter models were employed by Duke
Energy for the PV and BESS generation, which replicate the actual response of the
devices despite the unknown internal hardware and controls of the inverter.
The PV inverter adheres to IEEE 1547-2018, category I criteria and only
necessitates the reactive setpoint to operate effectively. Duke Energy has opted to
operate this system at unity power factor, or a reactive setpoint of zero, since the
primary aim of the PV generation is to support real power. This is due to the fact
that the substation acts as the reactive power source during grid connection, while
the BESS generation serves as the reactive power source during islanding [24].
On the other hand, the BESS inverter complies with IEEE 1547-2018, category
II criteria, and requires five input parameters, including a reference setpoint for real
power, reactive power, voltage, frequency, and inverter mode. Although the microgrid
master controller controls these setpoints in the real generation facility, they are
manually regulated in the simulation. The inverter can operate in two modes: GFL
and GFM. In GFL, the inverter employs the grid voltage and frequency as a reference,
but during GFM, it develops its own reference. Duke Energy has set the inverter
reference to 1.04 pu and 60 Hz during GFM, while the real and reactive setpoints are

26
based on load conditions [24]. To comprehend how the inverter operates in these two
modes, the following subsection provides an explanation.

2.3.1.1 Inverter Controls

An inverter model developed by Clemson University in [29] is referenced in

this subsection to understand inverter functionality in GFM and GFL. Since the
inverter controls used in this microgrid system are unknown, the Clemson inverter
model provides a high-level explanation of how an inverter operates.
In addition, this Clemson inverter model is also utilized for HIL testing later in
this thesis. This is due to the fact that the vendor inverter model mentioned previously
is only compatible with PSCAD and not available for the RSCAD platform, which
is required for real-time simulations within the RTDS. As such, the Clemson inverter
model serves as a suitable alternative for conducting HIL testing.
When operating in GFM mode, the inverter regulates voltage and frequency
using fixed references to ensure stable grid conditions. In contrast, when operating in
GFL mode, the inverter mirrors system conditions and provides real power support
to the grid, while reactive power is supplied by the grid source. By doing so, the
inverter helps to alleviate the load demand on the grid source, making it a crucial
component in modern power systems with continuously growing load demands.
Fig. 2.6 shows the controller used in [29], which is capable of switching between
these two modes of operation. This controller is a crucial component that enables
the inverter to operate effectively in both modes, meeting the varying demands of the
grid while ensuring stable and reliable operation.
The referenced inverter controls utilize a unique proportional-resonant (P R)
design that operates in the natural (abc) reference frame, eliminating the need for
transformation into the synchronous (dq) reference frame with a proportional-integral

27
 Figure 2.6: Inverter controls from Clemson University inverter model [29]

(P I) controller. This feature renders it ideal for real-time simulation as it elimi-

nates the need for additional processing associated with computation transformations.
Hence, the inverter in [29] was selected for the HIL testing.
To enable parallel operation of multiple IBRs without conflicts, a droop control
is implemented in this control scheme. The droop control employs reference inputs
∗
of real power (Pout ), reactive power (Q∗out ), frequency (f ∗ ), and voltage (Vrms
∗
) and
equivalent measured inputs (Pout , Qout , f ,Vrms ) to develop a 3-phase AC voltage
∗
reference (Vabc ) for the inverter, as illustrated in Fig. 2.6 [29].
In GFL, a positive-sequence current reference (i∗+
abc ) is generated by a PR con-

∗ ∗
troller that compares the measured voltage (Vabc ) with the voltage reference (Vabc ).
In contrast, in GFM, a PR controller with negative sequence components is used
in conjunction with the previous control loop to develop a current reference (i∗abc ).
−
These negative sequence components are extracted from the measured voltage (Vabc )

28
 ∗−
and voltage reference (Vabc ). The gain of the PR controller is depicted in (2.1), where
ωc is cut-off frequency and ω is the fundamental frequency [29].

2ωc s
GP R (s) = KP + Ki ∗ (2.1)
s2 + 2ωc s + ω 2

As previously mentioned, inverters are equipped with fault current limitations

to safeguard the internal power electronics components. In case of a fault condition,
this inverter is equipped with a current limiter that compares the output current
(iabc ), current reference (i∗abc ), and current limit (ILIM to detect any anomaly. If a
fault condition is detected, the controls clamp down the current reference to ensure
that it does not exceed the inverter current limit, which is illustrated in Fig. 2.7 [29].

Figure 2.7: Current limiter controls from Clemson University inverter model [29]

Subsequently, a current controller is employed to produce a full form PWM

29
waveform that executes firing pulses. The resulting output is then passed through
a low-pass filter. To filter out high-frequency outputs, this inverter incorporates an
LCL filter. As a result, the inverter is now successfully interfaced with the grid.

30
Chapter 3

Implementing Inverter Controls

and Seamless Transition

In the previous chapter, the distribution system and generation facility were
discussed, highlighting the battery generation as the primary resource to support the
microgrid. To connect the BESS to the AC grid, a conventional inverter is used as an
interface. However, this inverter is not capable of instantly switching between GFL
and GFM modes, as it requires a standby period. Consequently, achieving seamless
transition is not a standard capability of the inverter.
Nevertheless, it is possible to achieve seamless transition if the inverter is set
to operate in a single mode. To this end, this thesis proposes keeping the inverter in
GFM mode continuously and implementing additional external controls to ensure its
stability while connected to the grid.

31
3.1 Implementing Inverter operation in Grid-Forming

Mode while Grid-Connected

By continuously operating in GFM mode, the inverter can eliminate the re-
quired outage that arises from the transition. Furthermore, external controls were
designed in PSCAD to maintain the inverter’s stability during grid connection and
prevent any undesired voltage or frequency variations in the grid.
In the previous chapter, it was mentioned that the microgrid generation facility
used blackbox vendor models for the BESS and PV inverters. However, the PV
inverter does not allow for operation in GFM and always remains in GFL. Therefore,
the primary focus of the development of external controls was on the BESS inverter.
To enable parallel operation of the two BESS inverters, a droop control scheme is
implemented by the vendor. The frequency and voltage droop equations that the
inverter adheres to in GFM are shown in equations 3.1 and 3.2 [30].

finverter = f ∗ − KP ∗ (P − P ∗ ) (3.1)

Vinverter = V ∗ − KQ ∗ (Q − Q∗ ) (3.2)

Equations 3.1 and 3.2 reveal that the inverter relies on four input references:
voltage, frequency, real power, and reactive power which are denoted as V ∗ , f ∗ , P ∗ ,
and Q∗ , respectively. The droop gains are represented by KP and KQ , while P and
Q correspond to the measured real and reactive power values.
In equation 3.1, the frequency setpoint of the inverter is denoted as finverter.
It is calculated as the difference between the frequency reference and the product of
the real power difference and the real power gain. During steady state operation, the

32
real power difference is minimal and the frequency reference plays a dominant role in
determining the value of finverter. Similarly, in equation 3.2, the voltage setpoint of
the inverter is represented by Vinverter. It is calculated as the difference between the
voltage reference and the product of the reactive power difference and the reactive
power gain. During steady state operation, the reactive power difference is also small
and the voltage reference is the primary driving variable that determines the value of
Vinverter.
In a previous microgrid study, the voltage reference in GFM was set at a fixed
1.04 pu, the frequency reference was set at a fixed 60 Hz, and power references were
set by the microgrid master controller to meet load conditions. Moreover, the real
power droop coefficient was set to 0.02, and the reactive power droop was set to
0.2 [24]. However, this control scheme would inherently conflict with the grid due to
the fixed voltage and frequency references.
To ensure smooth operation of the inverter in GFM while grid-connected,
changes were made to the voltage and frequency references. Specifically, they were
set to be the measured voltage and frequency at the point of common coupling (PCC),
which mimics the inverter operation during GFL by taking the grid measurements
as reference. Meanwhile, the real power setpoint was set to the desired output of
the BESS plant to provide ancillary grid support, within the limits discussed in the
previous chapter. As the BESS generation output is restricted to unity power factor
during grid-connection, the reactive power setpoint was always kept at zero. As a
result, the microgrid master controller would adjust the reference inputs of the inverter
based on the grid connection status to continuously run the inverter in GFM while
grid-connected. Fig. 3.1 illustrates the different inputs into the inverter depending
on the grid connection status.
As depicted in Fig. 3.1, with the inverter always in GFM, the first switch

33
 Figure 3.1: Inverter reference inputs based on grid status

position represents the controlled inputs during grid connection, while the second
position represents the controlled inputs during grid isolation. In the first position,
the inverter reference voltage and frequency are set to match the measured values at
the PCC, ensuring smooth operation of the inverter in GFM while grid-connected. In
the second position, the inverter reference inputs are set as fixed values to generate
an independent reference to support the island. This control scheme was successfully
implemented on the PSCAD blackbox inverter in GFM and operated without con-
flicting with the grid during grid connection, enabling both charging and discharging.
Fig. 3.2 shows the inverter real power output given a real power setpoint of 1 MW
for discharge and -1 MW for charge.

34
 Figure 3.2: Inverter real power export and import in GFM while grid-connected

Fig. 3.2 clearly shows that the inverter performed discharge and charge cycles
successfully, with no deviation from the rated real power setpoint. This is because,
when the inverter is in GFM while grid-connected, it utilizes the measured grid voltage
and frequency as reference, enabling it to operate similarly to GFL. While the inverter
operated smoothly during charging and discharging cycles with this control loop, the
inverter’s GFL has a feature absent in GFM, which is ramp rate. As such, if setting
a higher desired real power output, it must be done in steps to avoid surpassing the
rapid voltage change (RVC) limits and triggering overvoltages. The next subsection
explores extreme discharge and charge scenarios and demonstrates how implementing
power steps can ensure stable operation.

35
3.1.1 System Response with Grid-Forming Inverter

To prevent damage to customer electronics and restrict voltage protection

misoperation, the microgrid generation facility must be equipped to provide ancillary
support to the grid within defined limits. Failure to do so may result in overvoltages
and RVCs which could lead to premature equipment degradation and damage.
The ANSI C84.1 sets the standard for steady-state voltage limits, specifying
that they should be within ±5% of the nominal voltage or within the range of 0.95
pu to 1.05 pu [23]. Meanwhile, the IEEE 1453 standard defines the most conservative
RVC for medium voltage systems as 3% [31]. Therefore, it is important for the
microgrid generation facility to be capable of supporting the grid within these limits
to ensure safe and reliable operation of the distribution system.
Duke Energy had previously examined the system response using the BESS
inverter in GFL. However, with the new GFM setup, it was essential to re-evaluate the
system response, especially given that GFM lacks the ramp rate option. It is worth
noting that the PV inverter’s ramp rate of 30 kW/s implemented in the previous study
remained unchanged since no alterations were made to the PV inverter controls [24].
In the preceding chapter of this thesis, the export and import limits of the
generation facility were established based on the system load conditions, as seen in
Table 2.1. Since the generation facility is composed of both BESS and PV generation,
the maximum power export of the facility can involve both BESS and PV export or
solely BESS export. Meanwhile, for maximum power import of the facility, only the
BESS can undergo a charge cycle. Consequently, to study the system response with
the new inverter configuration while grid-connected, three scenarios were explored
on the PSCAD system for peak and valley load conditions: maximum generation
facility export with PV and BESS, maximum generation facility export with BESS,

36
and maximum generation facility import with BESS, accounting for both generation
plant ramp-up and ramp-down. Details of these scenarios are presented in Tables 3.1,
3.2, and 3.3.

Table 3.1: Generation facility maximum export with PV and BESS

Mode PV Generation [MW] BESS Discharge [MW]

Peak Load 2.1 3.3
Valley Load 1.25 0

Table 3.2: Generation facility maximum export with BESS

Mode BESS Discharge [MW]

Peak Load 5.4
Valley Load 1.25

Table 3.3: Generation facility maximum import with BESS

Mode BESS Charge [MW]

Peak Load 3
Valley Load 6.9

With the BESS inverter in GFM and the absence of ramping rate, the study
in PSCAD found that majority of scenarios considered failed to meet voltage or
RVC limits. When the generation facility was set to maximum rated export for the
scenarios in Tables 3.1 and 3.2, the voltage inevitably exceeded the 1.05 pu overvoltage
limit due to the sudden high power injection, particularly as the facility is located
closely downstream of a voltage regulator. As outlined in the previous chapter, the
voltage regulator setpoint is 1.04 pu, resulting in the nominal voltage at the PCC
remaining consistently close to 1.04 pu. This allows for easy exception of the 1.05 pu
limit during power export. On the other hand, returning the generation facility from
maximum rated export to no export regularly resulted in a violation of the RVC limit

37
Figure 3.3: Maximum rated export of 5.4 MW during peak load with no BESS ramp
rate

of 3%, with the voltage regularly dropping near or below 0.95 pu with the sudden
loss of power injection.
Similarly, when setting the facility to maximum rated import for the scenario in
Table 3.3, it was also routine for RVC violations to occur, while the voltage sometimes
dropped below 0.95 pu due to the sudden addition of load. Conversely, returning the
facility from maximum rated import to no import often caused violations of the 1.05
pu overvoltage limit, as the distribution system saw this as a sudden loss of load.
Extreme examples of BESS exporting and importing with no ramp rate can be
seen in Figs. 3.3 and 3.4. In Fig. 3.3, sudden power export and loss of power export

38
Figure 3.4: Maximum rated import of 6.9 MW during valley load with no BESS ramp
rate

are shown for BESS discharge of 5.4 MW during peak load, taken from Table 3.2.
Similarly, Fig. 3.4 depicts sudden power import and loss of power import for BESS
charge of 6.9 MW during valley load, taken from Table 3.3. In the figures, power
export is denoted by positive values, while power import is indicated by negative
values. Both figures display the voltage at the PCC, illustrating the voltage and RVC
violations.
To resolve the observed violations of voltage and RVC limits, incremental
ramping of the BESS inverter’s GFM real power reference was implemented. With
the selected steps, the BESS inverter’s real power reference was incremented after

39
each voltage regulator operation, allowing for the voltage regulator to correct the
system voltage in between steps. This replicates the ramp rate functionality available
in the GFL inverter and prevents abrupt changes in system voltage. The step size
for ramping was carefully calibrated to ensure that it did not exceed any voltage or
RVC limits.
For the scenarios in Tables 3.1, 3.2, and 3.3, step sizes were selected to demon-
strate how voltage and RVC limits can be maintained, with an integer number of
steps chosen for each scenario. Tables 3.4 and 3.5 outline the step number and sizes
chosen for ramping up and down of the BESS and PV generation collectively and
solely BESS generation scenarios. Note that in the valley load scenario in Table 3.4,
the BESS does not export any power as the maximum rated export of the generation
facility is 1.25 MW during valley load, which is met by the PV system. However,
Table 3.5 shows the step number and sizes selected for the BESS exporting 1.25 MW
during valley load, as this scenario assumes solely BESS generation.
Table 3.4: BESS step sizes for maximum rated export with PV and BESS

Mode Number of Steps Step Size [MW/step]

Peak Load- Ramp Up 5 0.66
Valley Load- Ramp Up 0 0
Peak Load- Ramp Down 2 1.65
Valley Load- Ramp Down 0 0

Table 3.5: BESS step sizes for maximum rated export with BESS

Mode Number of Steps Step Size [MW/step]

Peak Load- Ramp Up 9 0.6
Valley Load- Ramp Up 2 0.625
Peak Load- Ramp Down 3 1.8
Valley Load- Ramp Down 1 1.25

Based on the simulated results, step sizes of approximately 0.63 MW are sug-
gested for ramping up the BESS export, while larger step sizes of about 1.6 MW can

40
be taken for ramping down. Due to the nominal voltage at the PCC remaining at
1.04 pu from the voltage regulator, the incremental steps for ramping up the BESS
export is smaller than those for ramping it down. Similarly, Table 3.6 displays the
step number and sizes for ramping of the BESS for maximum rated import scenarios,
taken from Table 3.3.
Table 3.6: BESS step sizes for maximum rated import with BESS

Mode Number of Steps Step Size [MW/step]

Peak Load- Ramp Up 2 1.5
Valley Load- Ramp Up 4 1.725
Peak Load- Ramp Down 3 1
Valley Load- Ramp Down 5 1.38

The simulated results suggest using step sizes of approximately 1.6 MW for
ramping up the BESS import, and smaller step sizes of 1.2 MW for ramping down.
In other words, ramping up the BESS to import power required smaller steps com-
pared to ramping down, which is opposite of the trend seen during power export.
This observation is reasonable since ramping up power export and ramping down
power import both cause a voltage increase, whereas ramping down power export
and ramping up power import result in a voltage drop.
To illustrate the interaction of a step ramp with the distribution system, con-
sider the ramp design in Table 3.6 for maximum rated import during peak load, which
is depicted in Fig. 3.5. For ramping up the BESS import, two steps of 1.5 MW are
taken to maintain the voltage within standard limits. On the other hand, for ramping
down the BESS import, three steps of 1 MW are taken to ensure the voltage remains
within the limits. The figure shows that tStart denotes the time the ramping begins,
tStep on the power plot indicates the time of each step, and tV R on the voltage plot
indicates the time when the voltage regulator operates. It is worth noting that since
ramping up the BESS import only required two steps, the third or final tStep and tV R

41
shown in Fig. 3.5 are specific to ramping down the BESS import, which required
three steps. Nevertheless, Fig. 3.5 shows that the voltage stays within the range of
0.95 pu to 1.05 pu, and no RVC exceeding 3% occurs, demonstrating the effectiveness
of a step ramp.

Figure 3.5: Example of step ramp with maximum power import during peak load

The careful selection of the ramping step sizes ensures that system limitations
are not exceeded under various generation conditions. Therefore, these results con-
clusively demonstrate the capability of the BESS inverter to operate in GFM mode
while grid-connected, while still meeting voltage and RVC limits.

42
3.2 Seamless Transition

The BESS inverter has been demonstrated to run in GFM mode while being
grid-connected for both discharge and charge cycles. This advancement eliminates
the need for the inverter to switch between operating modes, resulting in a single op-
erating mode and enabling seamless transition functionality. As a result, the inverter
is now capable of seamless islanding and grid reconnection.
In the next two subsection, the latest inverter configuration is successfully
demonstrated for seamless planned and unplanned islanding scenarios. The third
subsection showcases the synchronization of the microgrid with the grid to provide
seamless grid reconnection. With these advancements, the inverter provides a more
reliable solution for microgrid customers, making it a valuable asset for critical facil-
ities that cannot risk any outage.

3.2.1 Planned Islanding

An intermittent scenario to be required of the microgrid generation facility is

a planned islanding scenario. Planned islanding can be required if utility crews are
working upstream of the microgrid or if the substation is back-feeding another circuit
and the microgrid load needs to be shed from the total system load. For the sake of
the microgrid customers, it is crucial that the process of planned islanding be done
in a reliable and undetected fashion.
To ensure successful planned islanding, the generation facility must generate
sufficient power to exclusively power the microgrid load, resulting in minimal power
across the islanding recloser. A simple control loop was implemented in PSCAD to
incrementally increase the real and reactive power reference until the power across
the islanding recloser was less than 10 kW/kVAR. Therefore, with the BESS inverter

43
already in GFM and no inverter switching required, the island was successfully formed
once minimal power flow across the islanding recloser was detected. Fig. 3.6 depicts
the voltage measured at the PCC of the microgrid during the planned islanding
scenario.

Figure 3.6: Microgrid voltage during planned islanding

Fig. 3.6 demonstrates that the planned islanding scenario results in a neg-
ligible voltage change, with only a slight balance shift, as anticipated. Since the
planned islanding occurs under regular operating conditions rather than a grid dis-
turbance, the microgrid frequency remains steady at 60 Hz and is not shown. This
approach illustrates that the microgrid can smoothly transition from the main grid
to autonomous islanding without any blackstart, ensuring continuity of power supply
for customers.

44
3.2.2 Unplanned Islanding

A frequent scenario that a microgrid generation facility can encounter is an

unplanned islanding event. In such a scenario, if there is an external fault, the
islanding recloser disconnects in compliance with IEEE 1547-2018 standards. As the
BESS inverter is already in GFM, it can provide seamless support to the microgrid
load upon grid disconnection. Although the detailed design of the proposed protection
settings for the islanding recloser is discussed in Chapter 5, this section provides an
overview to demonstrate the unplanned islanding functionality.
According to IEEE 1547-2018, the islanding recloser must meet the minimum
ride-through times outlined in Chapter 1, Section 3 of this thesis. As specified by
Duke Energy, the microgrid islanding recloser is Category II compliant and must
comply with the ride-through times depicted in Fig. 1.5 for voltage ride-through and
Fig. 1.7 for frequency ride-through.
To illustrate the worst-case unplanned islanding scenario, or the one requiring
the longest ride-through time, a three-phase fault is applied at the farthest point
upstream of the microgrid. This resulted in a 0.16 second ride-through time for
the islanding recloser as the voltage dropped to 0.38 pu at the islanding recloser, as
seen in Fig. 1.5. Meanwhile, the frequency drop did not require any ride-through
time. During the unplanned islanding, Fig. 3.7 shows the voltage at the PCC of the
microgrid, Fig. 3.8 shows the frequency of the microgrid, and Fig. 3.9 shows the
current across the islanding recloser.
Fig. 3.7 shows a decrease in microgrid voltage to 0.38 pu at the onset of the
upstream fault, which gradually rises back to normal operating conditions after 0.16
seconds, or when the islanding recloser operates. Similarly, the frequency drops mo-
mentarily during the fault, but quickly recovers once the islanding recloser operates,

45
 Figure 3.7: Microgrid voltage during unplanned islanding

Figure 3.8: Microgrid frequency during unplanned islanding

46
 Figure 3.9: Current across islanding recloser during unplanned islanding

as depicted in Fig. 3.8. The fault current across the islanding recloser provided by
the microgrid generation facility can be seen in Fig. 3.9. Initially, it rises suddenly to
support the fault. After 0.03 seconds, the fault current is clamped and stabilizes to
0.15 kA, regulated by the inverter control loops. Finally, the fault current diminishes
to zero when the islanding recloser opens. The proposed islanding recloser protection
settings discussed in Chapter 5 will leverage this reverse fault current alongside the
IEEE 1547-2018 ride-through requirements.
The demonstration of unplanned islanding functionality was a crucial test, as
it showcased the microgrid’s ability to remain operational during unexpected power
outages or grid faults. By keeping the BESS inverter in GFM, the system was able to
seamlessly transition from the main grid to autonomous islanding, without the need
for a blackstart to restore power to microgrid customers.

47
3.2.3 Grid Synchronization

The previous two subsections focused on the process of seamlessly islanding

from the main grid. In contrast, this section will address grid synchronization and
the procedure used in PSCAD for reconnecting the island to the grid. Demonstrating
this capability is essential to exhibit the inverter’s fully seamless transition function-
ality. Additionally, connecting the generation facility to the main grid also requires
synchronization. Since the BESS inverter is now continuously operating in GFM, it
considers itself an independent source. Therefore, before the generation facility can
connect to the grid, it must synchronize with the grid.
In order to achieve successful synchronization, it is crucial to align three key
parameters: voltage, phase angle, and frequency, to ensure smooth transition. The
limits for these synchronization parameters are outlined in IEEE 1547-2018, which
specifies that the voltage difference must be within 3%, phase angle difference within
10◦ , and frequency difference within 0.1 Hz.
The objective of the synchronization process is to ensure the phasor difference
between corresponding phase voltages on either side of the synchronizing recloser is
close to zero when the recloser closes. It should be noted that the voltage phasors on
either side of the recloser are rotating at speeds corresponding to their frequencies.
The closer the frequencies are to each-other the smaller the relative speed of rotation
of the phasors. If such small relative speeds are achieved and the corresponding phase
voltage magnitudes on either side are nearly equal, there will be enough time for the
recloser to close during the time interval when the phasor difference between the
corresponding phase voltages is nearly zero - which is the ideal time to synchronize
the power systems on either side of the recloser. This is a two stage process: First
the phase voltage magnitudes of the microgrid are brought to the same values as

48
those of the grid. Second, the frequency of the microgid is brought close to the
grid frequency (not exactly the same). Then, the recloser is closed when the phasor
difference between the corresponding voltages is small enough as characterized by the
values given in Fig. 3.10.
In the PSCAD simulation, the BESS inverter is a blackbox model that limits
access to its internal controls and only allows control using external variables that are
acceptable for GFM. As a result, the BESS inverter’s external voltage and frequency
references must be adjusted to synchronize the island with the grid. It should be
noted that the inverter does not accept any phase reference in GFM mode, and
therefore, the frequency must be adjusted to bring the phase angle into alignment.
To illustrate the process, refer to Fig. 3.11, where θdif f represents the phase angle
difference between the grid phase angle, θgrid , and the island phase angle, θisland .
Assuming a counterclockwise rotation of the phase angle, adjusting the fre-
quency reference of the inverter can align the phase angle difference between the grid
and the island. If the grid phase angle leads the island phase angle, the inverter
frequency reference should increase to speed up phase rotation. Conversely, if the
grid phase angle lags the island phase angle, the inverter frequency reference should
decrease to slow down the rotation. Thus, Fig. 3.11 demonstrates that if the phase
angle difference falls within the first or second quadrant (θdif f 1 or θdif f 2 ), the inverter
frequency should increase, whereas if it falls within the third or fourth quadrant (θdif f 3
or θdif f 4 ), the frequency should decrease.
To achieve synchronization, two control loops were designed in PSCAD: one for
voltage and one for frequency, as depicted in Fig. 3.12. The voltage synchronization
loop ensures alignment of the voltage magnitudes between the grid and microgrid.
Simultaneously, the frequency loop serves a dual purpose: frequency and phase angle
synchronization. It aims to bring the microgrid and grid frequencies close, ultimately

49
Figure 3.10: Synchronization check

50
Figure 3.11: Utilizing inverter frequency to reduce phase difference between grid and
island

aligning the relative phase angles after a small time interval. As the frequency dif-
ference decreases, the relative phase angle becomes more precise and assumes a more
significant role in defining the frequency.
As shown in Fig. 3.12, the voltage control loop computed the RMS voltage dif-
ference (Vdif f ) between the grid RMS voltage (Vgrid ) and island RMS voltage (Visland )
and scaled it by a gain (Gv ). This difference was then added to a reference voltage
of 1.04 pu to output the inverter voltage reference (Vref ). Meanwhile, the frequency
control loop determined the frequency reference (fref ) using both the phase angle

51
Figure 3.12: Inverter synchronization controls using inverter voltage and frequency
reference

difference and frequency difference. The phase angle difference (θdif f ) was calculated
as the difference between the grid phase angle (θgrid ) and island phase angle (θisland ).
This difference was multiplied by a gain (Gθ ) and added to a reference frequency
of 60 Hz. Similarly, the frequency difference (fdif f ) was calculated as the difference
between the grid frequency (fgrid ) and island frequency (fisland ), and was multiplied
by a gain (Gf req ) before being added to the same loop as the phase angle.
An example was conducted in PSCAD to demonstrate the functionality of the
synchronization control loop, where the microgrid was synchronized with the main
grid during a worst-case scenario of the phase difference being in the third quadrant.
While achieving synchronization, the voltage measured at the PCC of the microgrid

52
can be observed in Fig. 3.13, while Fig. 3.14 displays the microgrid frequency.

Figure 3.13: Microgrid voltage during grid synchronization

Fig. 3.13 displays the voltage magnitude rising to meet the grid voltage mag-
nitude. Meanwhile, Fig. 3.14 shows that the frequency drops at the beginning of the
synchronization process to reduce the relative rotation of voltage phasors. This drop
in frequency occurs because the phase angle of the island was leading the phase angle
of the grid, as the phase angle difference was in the third quadrant. After 1.2 seconds,
the synchronization check passes, and the islanding recloser closes, connecting the mi-
crogrid with the main grid. Subsequently, Fig. 3.13 indicates that the voltage adjusts
to grid conditions, while Fig. 3.14 depicts the frequency quickly returning to 60 Hz.
This example effectively illustrates the synchronization control loop’s functionality.
While the BESS inverter cannot switch seamlessly between GFL and GFM,
this chapter showcased its ability to operate continuously in GFM. By maintaining the

53
 Figure 3.14: Microgrid frequency during grid synchronization

inverter in GFM mode, the study demonstrated the possibility of seamless transitions
through planned and unplanned islanding, as well as grid synchronization in PSCAD.
Although this technique enhances microgrid resilience and customer power reliability,
it presents additional protection challenges. The remainder of the thesis proposes and
tests a protection scheme that does not depend on external communication or relay
settings adjustments to maintain system protection during seamless transitions.

54
Chapter 4

Existing Protection Devices

Before proceeding with the proposed adaptive protection scheme for the dis-
tribution system, the current protection devices and settings established by Duke
Energy are first examined. As outlined in Chapter 2 of this thesis and illustrated in
Figs. 2.1 and 2.3, the distribution system and generation facility are protected by
various devices such as fuses, electronic and hydraulic reclosers, and digital relays,
which utilize overcurrent as the method of protection.
The reclosers are equipped with inverse time-overcurrent settings which means
that they operate based on measured current and determine their associated operating
time based on a time-current curve (TCC), given the current exceeds a specified
threshold. Fuses are installed across multiple laterals to safeguard downstream loads,
while reclosers positioned along the distribution system backbone can implement
fuse-saving. Reclosers are specialized protective devices that can perform multiple
operations of opening and closing to attempt to clear temporary faults before locking
out for a permanent fault. Fuse-saving is a protective philosophy that aims to save a
fuse against temporary faults by allowing the recloser to operate before the minimum
melting curve for initial reclosing operations, while allowing it to blow for permanent

55
faults for later reclosing operations. Reclosers also act as a backup for the fuses and
safeguard sections of the system backbone [32].
Moreover, relays in the generation facility are equipped with inverse time-
overcurrent settings to safeguard against on-site faults. However, the introduction
of the generation facility poses two major protection challenges: fault current di-
rection and limited fault current magnitude. Firstly, in a traditional distribution
system, overcurrent protection was only required to operate in a single direction as
the system was single-sourced and radial. Yet, with the inclusion of added generation,
bi-directional settings become crucial. Secondly, traditional synchronous generation
allowed for a significant disparity between load current and fault current, which aided
in system protection. However, in the case of IBRs, load current and fault current
are closely matched in magnitude, as explained in Chapter 1 and depicted in Figure
1.3. Figure 4.1 provides a visual representation of the protection zones within the
distribution system incorporating the generation facility.
As the generation facility introduces bi-directional fault current, a forward and
reverse protection zone is defined, with the forward direction assumed to be away from
the substation, while the reverse direction is assumed to be towards the substation.
The forward main and backup protection zones can be seen in Table 4.1, while the
reverse main and backup protection zones can be seen in Table 4.2.
Fig. 4.1 displays a total of 14 protection zones across the distribution network.
As mentioned in Chapter 1, the previous microgrid study assumed the microgrid to
be blackstart-only due to the BESS inverter’s limitations. Therefore, Duke Energy
engineered the PCC protection settings to isolate the generation facility during an
external fault, enabling the main grid protection to isolate the external fault. As a
result, only the protective devices within the generation facility were designed with bi-
directional settings, with reverse settings for faults external to the generation facility,

56
57
Figure 4.1: Existing distribution system protection zones with the generation facility
Table 4.1: Existing distribution system forward protection zones with generation
facility

Zone Number Main Protection Backup Protection

1 Recloser 1 Substation Breaker
2 Recloser 2 Recloser 1
3 Islanding Recloser Recloser 2
4 Recloser 3 Islanding Recloser
5 Recloser 4 Recloser 1
6 100KS Fuse Recloser 1
7 80KS Fuse Recloser 2
8 65KS Fuse Islanding Recloser
9 80K Fuse Recloser 3
10 PCC Recloser Islanding Recloser
11 BESS Relay PCC/PV Relay
12 PV Relay PCC/BESS Relay
13 AUX Relay PCC/BESS/PV Relay
14 12T Fuse AUX Relay

Table 4.2: Existing distribution system reverse protection zones with generation fa-
cility

Zone Number Main Protection Backup Protection

3 PCC Recloser PCC Relay
10 PCC Relay BESS/PV Relay
11 BAY-O-NET 65A Fuse Cooper ELSP 125A Fuse
12 BAY-O-NET 50A Fuse Cooper ELSP 80A Fuse

and forward settings for any internal faults from the main grid. All generation facility
protection settings were undervoltage supervised to distinguish between load current
and fault current. Within the generation facility, two fuses on the MV side of the
BESS and PV transformer provide main and backup protection for the generation
tie lines, with additional vendor-provided protection for all faults further within the
generation plants.
Additionally, Fig. 4.1 shows four fuses - 100KS, 80KS, 65KS, and 80K -
representing the various fused laterals downstream recloser 1, recloser 2, islanding
recloser, and recloser 3, respectively. The existing fuse design incorporates single fuse

58
type within each recloser zone. These reclosers coordinated with the respective fuses
for fuse-saving.
With this existing protection scheme, when the microgrid disconnects from
the main grid, the protection layout changes, as depicted in Fig. 4.2. The resultant
main and backup protection for each zone can be seen in Table 4.3. Furthermore,
the existing protection scheme requires all protective devices within the microgrid
to have a second set of settings for the different protection layout, highlighting the
disadvantage of the existing scheme for microgrids with seamless transition.

Table 4.3: Existing microgrid system protection

Zone Number Main Protection Backup Protection

11 BAY-O-NET 65A Fuse Cooper ELSP 125A Fuse
12 BAY-O-NET 50A Fuse Cooper ELSP 80A Fuse
13 AUX Relay BESS/PV Relay
14 12T Fuse AUX Relay
10 PCC Relay BESS/PV Relay
3 PCC Recloser PCC Relay
4 Recloser 3 PCC Relay

Fig. 4.2 illustrates 7 protection zones for the microgrid. During island oper-
ation, the microgrid relies solely on the generation facility for power supply, so the
protection scheme can follow a simpler, radial method, with each device strictly pro-
tecting downstream loads from faults. As a result, bi-directional settings were not
required as there was no flow of current from the main grid. It is important to note
that all of the microgrid protection settings were specifically designed to operate at
a faster pace compared to the primary distribution system protection. This design
choice was made to meet the fault ride-through constraints of the BESS inverter.
Additionally, these settings cannot be coordinated with the fused laterals since the
fault current observed by the fuse is insufficient to trip within an appropriate time
frame. Consequently, zone 8 and 9 in Fig. 4.2 are included within zone 3 and 4,

59
60
Figure 4.2: Existing microgrid system protection zones
respectively. Hence, Duke Energy has devised two sets of protection settings that
vary based on the mode of operation of the system. These changes in settings are
allowed to occur during the necessary open transition period between grid-connected
and islanded modes. However, as this transitional period is no longer present, a more
robust scheme must be implemented.

61
Chapter 5

Proposed Adaptive Protection

Scheme

In light of the seamless transition capability discussed in Chapter 3, the exist-

ing protection scheme is no longer adequate. The inherent delays in communication
and relay settings adjustment when switching between grid-connected and islanded
modes of operation make the use of a two-setting protection scheme problematic,
resulting in miscoordination or even loss of protection during the transition period.
This is because the two setting groups differ greatly in terms of their fault current
magnitude, coordination, and directional settings. The protection scheme for grid-
connected operation requires higher fault current magnitudes, fuse coordination, and
bi-directional settings, while the protection scheme for microgrid operation operates
on a much lower fault current magnitude, assumes that fuse operation is neglected,
and is radial. In response, this chapter proposes a new adaptive protection scheme
that remains immune to seamless transition, avoiding the need for communication
and change of relay settings.

62
5.1 Protection Design Criteria and Assumptions

The protective devices discussed in Chapter 4 remain unchanged. However,

now all devices are designed with a single set of settings that locally adapt to the fault
conditions. Additionally, instead of isolating only the generation facility in the event
of an external fault, the entire microgrid must now isolate by utilizing the islanding
recloser, as the BESS inverter is always running in GFM and capable of seamless
transition.
Before designing any protection setpoints, a comprehensive fault study was
conducted on the PSCAD system under various modes of operation. This study
aimed to gather all the necessary fault data to design a robust protection scheme
capable of addressing all extreme scenarios. Four fault types were applied to locations
that resulted in maximum and minimum fault currents: three-phase (3PH) fault,
line-to-line (LL) fault, line-to-line-to-ground (LLG) fault, and single line-to-ground
(SLG) fault. To determine the most severe configuration for each fault type, different
combinations of phases were utilized within each category.
In a study by Burke [33], an extensive four-year fault analysis was carried out
in the United States. The study encompassed fifty feeders from thirteen utilities,
operating at various voltage levels. One notable finding was the negligible fault
resistance observed in the recorded faults. This observation was further supported by
the close agreement between the actual and calculated fault current values. In line
with the findings of Burke’s study, all simulated faults within the scope of this study
are bolted.
The fault scenarios were applied to both the distribution system, with and
without the generation facility, as well as the microgrid system, which relies solely on
the generation facility for support. These specific fault scenarios can be seen listed

63
in Table 5.1.
Table 5.1: Simulated system conditions for fault study in PSCAD

Fault Scenarios Grid Fault Study Microgrid Fault Study

BESS and PV Discharge Considered Considered
BESS Discharge Considered Considered
BESS Charge Considered Not Considered
No IBR Considered Not Considered

The grid fault study included different scenarios involving maximum rated
discharge and charge of the generation facility, with variations in BESS and PV
connections, as well as the absence of an IBR connection. On the other hand, the
microgrid fault study encompassed scenarios involving the supply of microgrid load
through both BESS and PV generation or solely through BESS. In the design of the
protection scheme, only the extreme fault values were taken into consideration.
During the grid fault study, it was observed that maximum fault currents were
generated with BESS and PV discharge or during strictly BESS discharge. Con-
versely, minimum fault currents in the grid fault study were recorded in cases where
there was no IBR connection. Meanwhile, in the island fault study, both BESS and
PV generation and strictly BESS generation resulted in the same fault currents. This
can be attributed to the fact that the BESS generation plants serves as the primary
source of fault current. While the PV generation plant supplies power to the micro-
grid load during normal operations, it is limited to GFL and only meets the baseline
requirements outlined in IEEE 1547-2018 Category I. As a result, the PV inverter re-
peatedly entered a state of momentary cessation of energization due to undervoltage
conditions. This behavior is exemplified in the example in Fig. 5.1, which depicts
a 3PH fault at the PCC during grid connection. Nonetheless, the maximum and
minimum measured fault currents at each protective device will be further discussed
in subsequent sections as the protection scheme is being devised.

64
 Figure 5.1: PV inverter fault response for 3PH fault at PCC

The development of the overcurrent protection settings for each device adheres
to the constraints specified in IEEE Std. 242-2001 and references the procedure
outlined by Blackburn and Domin [32]. The ANSI protection element for inverse
time-overcurrent is ANSI 51, where 51P denotes phase overcurrent and 51G represents
ground overcurrent. For phase settings, the minimum operating current (MOC) for
each relay or recloser is set to 200% or more of the maximum rated load current while
being 50% or less of the minimum far-end phase fault current. The phase settings
are designed to operate for 3PH, LL, and LLG faults. Regarding ground settings, the
MOC is set to 200% or more of the maximum unbalance, 33% or less than the phase
MOC, and 50% or less than the minimum far-end ground fault current [32]. Similarly,
the ground settings are designed to operate for primarily SLG faults, but can also
operate for LLG faults in the presence of high enough ground current. It is worth

65
noting that the phase element also detects SLG faults, however, the ground element
is the first to operate. Nonetheless, if the MOC cannot meet these specifications,
which was common for IBR faults, voltage supervision is utilized. The design of each
recloser and relay incorporates a curve derived from the US TCC curves, as defined
by equations 5.1-5.5 [20].

 
0.0104
tp = T D ∗ 0.02260 + 0.02 (5.1)
M −1
 
5.95
tp = T D ∗ 0.18000 + 2 (5.2)
M −1
 
3.88
tp = T D ∗ 0.09630 + 2 (5.3)
M −1
 
5.67
tp = T D ∗ 0.03520 + 2 (5.4)
M −1
 
0.00342
tp = T D ∗ 0.00262 + 0.02 (5.5)
M −1

Equations 5.1, 5.2, 5.3, 5.4, and 5.5 represent different overcurrent charac-
teristic curves used for the design of reclosers and relays. The moderately inverse
curve, known as U1, is defined by equation 5.1. The inverse curve, denoted as U2,
is described by equation 5.2. Equation 5.3 represents the very inverse curve, referred
to as U3. Likewise, equation 5.4 showcases the extremely inverse curve, identified
as U4. Finally, the short-time inverse curve, labeled as U5, is outlined in equation
5.5. In these equations, the variable T D represents the time-dial setting selected for
each curve, which can range from 0.5 to 15 [20]. The variable tp corresponds to the
operating time associated with the specific curve, and M represents the ratio of the
fault current and the MOC.
In order to ensure reliable and coordinated system protection, the design in-

66
corporates a coordination margin between the main protection and backup protection
devices, which is crucial in the event of a breaker or relay failure. The coordination
time interval (CTI) between each overcurrent protective device is defined and specified
in Table 5.2, where CS represents clear space between device curves [34].

Table 5.2: CTI margin between main and backup protection

Backup Protection
Main LV Circuit Electromechanical Digital
Protection Fuse Breaker Relay/Recloser Relay/Recloser
Fuse CS CS 0.22 s 0.12 s
LV Circuit Breaker CS CS 0.22 s 0.12 s
Electromechanical
Relay/Recloser 0.2 s 0.2 s 0.3 s 0.2 s
Digital Relay/Recloser 0.2 s 0.2 s 0.3 s 0.2 s

The CTI margins encompass various factors, including breaker operation delay,
overshoot, relay timing errors, CT errors, and an additional safety margin. These
margins are incorporated to ensure reliable device coordination between the main and
backup protection systems. To establish this coordination, a coordination range is
initially determined, representing the range of fault currents within which the backup
protection should operate. The coordination range for phase curves encompasses the
maximum and minimum fault currents for 3PH, LL, and LLG faults. On the other
hand, the coordination range for ground curves includes SLG faults and occasionally
LLG faults, depending on the ground current magnitude. Within this coordination
range, the CTI requirements must be satisfied.
When designing a recloser fast curve for fuse-saving protection, the curve is
designed to operate faster than the fuse minimum-melting curve. Conversely, when
designing the slow curve, the fuse is employed as the primary protection, while the
slow curve is utilized as the backup protection. This configuration ensures rapid
response to faults for initial recloser operations and allows for fuse operation in later

67
recloser operations, prioritizing fuse-saving for temporary faults.
To achieve coordination between reclosers, a coordination range is defined be-
tween main and backup protection, but sequence coordination is also assumed. With
sequence coordination, the backup recloser keeps track of the main recloser opera-
tions and increments its shot counter accordingly [20]. This ensures that repetitive
recloser operations are avoided and prevents any potential miscoordination between
the fast and slow curves. With sequence coordination functionality, the reclosers work
together harmoniously, allowing for an efficient and coordinated response to faults.
During the design of protection settings for devices within the microgrid, it
is crucial to consider the FRT capability of the BESS inverter. Analysis of the sys-
tem fault study reveals that the BESS inverter repeatedly tripped due to voltage
ride-through issues, while frequency ride-through was not a cause for tripping. As
previously mentioned, the BESS inverter complies with IEEE 1547-2018 Category II
standards, which mandates a minimum FRT duration of 2 seconds for severe faults,
as depicted in Fig. 1.5. Consequently, to prevent the BESS inverter from tripping, all
protective devices within the microgrid must operate within a 2 second time-frame,
ensuring their response aligns with the FRT capability of the BESS inverter. Mean-
while, to achieve coordination with the substation circuit breaker, an upper limit of
3 seconds is set for all protective devices within the grid.
To streamline the process of developing system settings, MATLAB functions
were created to iterate through different device settings. The functions helped facili-
tate the identification of the fastest protection settings for each device while consid-
ering the specified constraints. To do this, the script utilized the CTI margin, the
necessary fault data, and corresponding operating times (OTs) to find coordinated
settings for all fault types. Additionally, the script accounts for the individually mea-
sured fault currents at both main and backup protection devices to ensure precise

68
coordination. The various MATLAB functions associated with this process can be
found in Appendix A. Based on the fault study parameters and the assumed protec-
tion setting assumptions, the subsequent sections will delve into the development of
the proposed protection scheme.

5.2 Voltage-Restrained Time Overcurrent Protec-

tion

During a fault event, the behavior of a synchronous generator differs from

an IBR in terms of current and voltage characteristics, as mentioned in Chapter 1.
Synchronous generators can deliver higher fault currents and maintain higher system
voltages compared to IBRs. This discrepancy in voltage response between the two
types of resources becomes advantageous when it comes to distinguishing between
different fault conditions within the system.
In order to develop protection settings that can adapt locally, a voltage-
restrained time overcurrent protection method is employed, also known as ANSI 51V.
The implementation of 51V in this thesis adheres to industry standards [35]. By uti-
lizing this approach, the MOC of the TCC is dynamically adjusted based on local
voltage conditions, as shown in Fig. 5.2, enhancing its adaptability and reliability.
The setting of 51V involves the selection of a reference voltage, denoted as
Vref . When the measured line voltage at the relay falls below this reference voltage,
a multiplier is generated to decrease the MOC of the relay. It is important to note
that the minimum MOC multiplier is set to 0.125 and the maximum is set to 1.0,
following the standard industry practice.
The reference voltage is fine-tuned to ensure optimal operation based on the

69
 Figure 5.2: Voltage-restrained time overcurrent characteristic

Figure 5.3: Dynamic response of 51V protection curve as a function of measured

voltage drop

70
findings of the system fault study. This approach allows for a lower MOC to be
applied in response to more severe voltage drops, effectively adapting the OT of the
protective device to island faults.
A graphical representation of the TCC variation as the voltage drops below
the reference can be observed in Fig. 5.3. This depiction illustrates how the MOC
reduces in response to voltage drops, demonstrating an improved OT for more severe
voltage drops. By incorporating this adaptive capability, the protection scheme can
efficiently and accurately respond to different fault scenarios.

5.3 Protection Design Overview

While the existing protective devices discussed in Chapter 4 remain in use,

there are modifications to the main and backup protection devices for each zone. This
ensures reliable and selective operation during both grid-connected and microgrid
operation. To provide a clearer visual representation, Figs. 5.4 and 5.5 depict the
labeled zones during grid connection and microgrid operation respectively.
In Chapter 4, the power flow direction in the system was defined as follows:
power flowing away from the substation is referred to as the forward direction, while
power flowing towards the substation is referred to as the reverse direction. To im-
prove the protective scheme, a revision has been made to the main and backup pro-
tection configuration for both forward and reverse directions during grid-connected
operation. These revisions for the protection zones in Fig. 5.4 can be found in Tables
5.3 and 5.4. Similarly, during microgrid operation, the main and backup protection
for both forward and reverse directions are redefined. Tables 5.5 and 5.6 provide the
main and backup protective devices for the zones in Fig. 5.5.
The proposed protection scheme incorporates an additional role for the island-

71
72
Figure 5.4: Proposed distribution system protection zones with the generation facility
73
Figure 5.5: Proposed microgrid system protection zones
Table 5.3: Proposed forward protection zones of distribution system with generation
facility

Zone Number Main Protection Backup Protection

1 Recloser 1 Substation Breaker
2 Recloser 2 Recloser 1
3 Islanding Recloser Recloser 2
4 Recloser 3 Islanding/PCC Recloser
5 Recloser 4 Recloser 1
6 100KS Fuse Recloser 1
7 80KS Fuse Recloser 2
8 65KS Fuse Islanding/PCC Recloser
9 80K Fuse Recloser 3
10 PCC Recloser Islanding Recloser
11 BESS Relay PCC Relay
12 PV Relay PCC/BESS Relay
13 AUX Relay PCC/BESS Relay
14 AUX Relay 12T Fuse

Table 5.4: Proposed reverse protection zones of distribution system with generation
facility

Zone Number Main Protection Backup Protection

3 PCC Recloser PCC Relay
10 PCC Relay BESS Relay
11 BAY-O-NET 65A Fuse Cooper ELSP 125A Fuse
12 BAY-O-NET 50A Fuse Cooper ELSP 80A Fuse

ing recloser, which is to isolate the entire microgrid in the event of external faults.
This differs from the previous approach, where the PCC would isolate itself as the
BESS inverter operated in GFL. However, with the BESS inverter continuously op-
erating in GFM, the generation facility is now capable of seamlessly supporting the
microgrid, allowing the islanding recloser to disconnect from the main grid.
To ensure system protection amidst seamless transition, all devices are pro-
grammed with single settings while still meeting the criteria outlined in Tables 5.3,
5.4, 5.5, and 5.6.
Within the generation facility, with the exception of the AUX relay, all devices

74
 Table 5.5: Proposed forward protection zones of microgrid system

Zone Number Main Protection Backup Protection

12 PV Relay AUX Relay
13 AUX Relay BESS Relay
14 AUX Relay 12T Fuse
10 PCC Relay BESS Relay
3 PCC Recloser PCC Relay
4 Recloser 3 PCC Relay

Table 5.6: Proposed reverse protection zones of microgrid system

Zone Number Main Protection Backup Protection

11 BAY-O-NET 65A Fuse Cooper ELSP 125A Fuse
12 BAY-O-NET 50A Fuse Cooper ELSP 80A Fuse

are programmed with bi-directional overcurrent settings, allowing for operation in

both forward and reverse directions. Similarly, the islanding recloser is also designed
with bi-directional settings. This enables it to detect reverse faults and provide fuse-
saving capabilities in the forward direction.
In contrast, protective devices located outside the microgrid are programmed
with only forward settings. This is due to the presence of the islanding recloser, which
takes on the responsibility of isolating the microgrid from external faults, eliminating
the reverse fault contribution. As a result, the external protective devices can focus
solely on addressing fault conditions originating from the main grid.
As depicted in Fig. 5.5, during islanded operation, recloser 3 must operate for
faults within the 80K fused zone. In contrast, as shown in Fig. 5.4, recloser 3 must
coordinate with the 80K fused zone to provide selective operation. Therefore, the
settings of recloser 3 are chosen to be programmed using 51V to ensure coordination
with downstream fuses during grid connection, as well as rapid operation during
island mode. The subsequent section will provide a comprehensive description of
the configuration and operation of recloser 3, presenting a detailed explanation of its

75
designed settings.

5.4 Recloser 3

Recloser 3 serves as the sole recloser located downstream of both the genera-
tion facility and the islanding recloser, as depicted in Figs. 5.4 and 5.5. Consequently,
recloser 3 encounters a wide range of fault currents originating from either the substa-
tion or the limited generation facility. To account for this variability, recloser 3 was
programmed with the 51V protection element, which adjusts the MOC based on the
measured line voltage. As 51V operation is dependent on both voltage and current,
all fault types were individually considered rather than just a general coordination
range.
Recloser 3 must coordinate with downstream 80K fused laterals. From the
system fault study, Tables 5.7 and 5.8 present the maximum and minimum simulated
faults within the protection zone of all 80K fused laterals (referred to as zone 9 in
Fig. 5.4) during grid connection and microgrid operation. Similarly, Tables 5.9 and
5.10 display the same faults within the protection zone of the 80K fuse but provide
the corresponding voltage and current measurements obtained at recloser 3.

Table 5.7: Fault data recorded at 80K fuse for simulated faults within 80K fuse zone
during grid connection

Fault Type Maximum Fault [A] Minimum Fault [A]

SLG 815 504
LLG 1024 701
LL 927 679
3PH 1032 784

The fault study results, presented in Tables 5.9 and 5.10, highlights a signifi-
cant difference in line voltages between grid-connected and microgrid faults. Notably,

76
Table 5.8: Fault data recorded at 80K fuse for simulated faults within 80K fuse zone
during microgrid operation

Fault Type Maximum Fault [A] Minimum Fault [A]

SLG 282 267
LLG 308 286
LL 230 216
3PH 211 206

Table 5.9: Fault data recorded at recloser 3 for simulated faults within 80K fuse zone
during grid connection

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 818 513
SLG Ground Current [A] 720 470
Line Voltage [kV] 14.79 17.52
Phase Current [A] 1029 712
LLG Ground Current [A] 513 353
Line Voltage [kV] 2.58 3.89
Phase Current [A] 980 735
LL
Line Voltage [kV] 2.6 3.88
Phase Current [A] 1035 795
3PH
Line Voltage [kV] 2.54 3.88

Table 5.10: Fault data recorded at recloser 3 for simulated faults within 80K fuse
zone during microgrid operation

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 283 272
SLG Ground Current [A] 270 248
Line Voltage [kV] 7.71 7.99
Phase Current [A] 309 288
LLG Ground Current [A] 447 384
Line Voltage [kV] 0.52 1.03
Phase Current [A] 232 214
LL
Line Voltage [kV] 0.5 0.97
Phase Current [A] 211 209
3PH
Line Voltage [kV] 0.52 1.03

there is a more pronounced voltage drop observed during islanded conditions, as cor-
roborated by [36, 37, 38]. Therefore, this can be leveraged by utilizing the 51V

77
protection element to dynamically adjust MOC based on voltage drop.
In the context of microgrid operation, the 80K fuse experiences a minimum
fault current of 206 A. Consequently, the fuse operates after 300 seconds, which is not
a practical OT. Given that the fault current magnitude during microgrid operation
is not sufficiently high, it is necessary to extend the protection zone of recloser 3
during island mode to include the zone protected by the 80K fuse, labeled as zone 9
in Fig. 5.5. This arrangement ensures proper coverage. Fig. 5.6 visually illustrates
how recloser 3 provides backup protection for the 80K fuse during grid connection
and Fig. 5.7 demonstrate how it expands its protection zone to encompass the region
protected by the 80K fuse during island operation.

Figure 5.6: Recloser 3 and 80K fuse coordination during grid connection

Additionally, the fault study reveals that during island operation, the ground
current of an LLG fault surpasses the phase current, in contrast to grid-connected
operation. Therefore, when determining the ground setpoint, the elevated magnitude
of ground current from island LLG faults were taken into consideration.

78
 Figure 5.7: Recloser 3 extended protection zone during microgrid operation

The protection settings for recloser 3 were determined by considering all the
provided fault data in conjunction with the specified CTI values in Table 5.2. For
coordinating recloser 3 with the fuse total clearing curve, a CTI of 0.12 seconds was
taken, meanwhile, 0.2 seconds was taken for coordinating with the fuse minimum
melting curve. The designed settings for recloser 3 are summarized in Table 5.11.
Table 5.12 displays the OTs of the 80K fuse backed by recloser 3, presenting the
specific OT for the maximum and minimum fault scenarios.

Table 5.11: Recloser 3 protection settings

Function MOC [A] TD Curve Supervisor

51V-P- Slow Curve 274 4.03 U5 Vref = 5 kV
51V-G- Slow Curve 55 11.66 U5 Vref = 17.5 kV
51P- Fast Curve 104 0.5 U5 ANSI 27
51G- Fast Curve 55 0.5 U5 None

Table 5.11 demonstrates the implementation of 51V for the slow curves, en-
suring coordination with the 80K fuse during grid-connection and faster operation

79
Table 5.12: OTs of 80K fuse backed by recloser 3 for maximum and minimum faults
within the main protection zonea

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.19 0.44 0.04, 0.35 0.04, 0.56
80K Fuse Rec 3
Ground 0.31 0.82 0.03, 0.63 0.04, 0.94
a
#, # - Represents recloser OTs for fast, slow curves.

than the fuse during island operation. Selecting the 51V base MOC for the phase
and ground slow curves resulted in a phase MOC of 274 A and a ground MOC of
55 A, decreasing based on voltage. These values were determined considering that
the maximum rated phase current at recloser 3 is 137 A, while the maximum rated
ground current is 27.1 A. The chosen MOCs were set at 200% of the rated currents,
as specified in the first section of this chapter.
As the 51V protection element adjusts the MOC dynamically, a reference
voltage was carefully selected considering the distinct voltage drop. By conducting
MATLAB optimization to ensure coordination for all faults, the phase reference volt-
age was determined to be 5 kV accompanied by a short-time inverse curve, while the
ground reference voltage was set to 17.5 kV alongside a short-time inverse curve.
Table 5.12 reveals that the coordination time of 0.12 seconds is primarily
influenced by the minimum fault. This can be attributed to the dynamic nature of
the 51V protection element, which considers both voltage and current parameters.
During fault conditions, as the voltage drops, the lowest fault current emerges as the
dominant factor for coordination.
On the other hand, the fast curves for recloser 3 were based on a standard in-
verse time-overcurrent curve, or ANSI 51, as the main requirement was for the curve
to operate faster than the fuse curve. Therefore, the fast curve did not necessitate a
dynamic response. Similar to the selection of MOC for the slow curves, determining

80
the MOC for both the phase and ground fast curves involved considering the maxi-
mum rated load and ground current. However, as the MOC remains fixed and does
not decrease, it was important to also consider that the minimum far-end phase fault
current of 209 A and the minimum far-end ground fault current of 248 A. Meeting
both criteria of exceeding 200% of the load current and remaining below 50% of the
minimum far-end fault current proved to be unattainable for the phase MOC.
In order to address this challenge, a priority was placed on selecting a phase
MOC that corresponds to 50% of the far-end phase fault current, accompanied by
the implementation of an undervoltage supervisor. Consequently, a phase MOC of
104 A was chosen, incorporating an undervoltage supervisor set at 0.8 pu. The
corresponding protection element for undervoltage is ANSI 27, and a threshold of
0.8 pu was selected to ensure the protection operates effectively during abnormal
grid conditions. Meanwhile, the ground fast curve MOC did not encounter the same
constraint. Thus, it was possible to adopt 200% of the maximum rated ground current
as the MOC.
To facilitate efficient coordination with all upstream recloser fast curves, the
time dial (TD) setpoint for the recloser fast curves was set to the fastest setting of
0.5. This decision was made to enable fast coordination among upstream recloser fast
curves, as they also need to coordinate their operation.
In Fig. 5.8, the coordination between the recloser 3 phase curves and the 80K
fuse is illustrated for both maximum and minimum phase faults occurring within the
80K fuse’s protection zone. It is important to note that the phase slow curve operates
based on the 51V reference and is dynamic in nature. To depict the coordination, an
average line was used, incorporating the maximum and minimum OTs for the phase
faults to outline the curve.
Similarly, Fig. 5.9 demonstrates the coordination between the recloser 3

81
ground curves and the 80K fuse. However, since the fuse does not operate based
on ground current but rather phase current, the recloser ground curves are referred
to phase to align with the fuse curve.
Additionally, Figs. 5.10 and 5.11 employ the same methodology, illustrating
the operation of recloser 3 for all island faults within its expanded microgrid zone,
which incorporates the 80K fuse protection zone. This demonstrates that the recloser
no longer coordinates with the fuse and operates at a faster rate for island faults.

Figure 5.8: Recloser 3 phase coordination with 80K fuse for grid faults

The TCC plots serve as evidence of the successful coordination between re-
closer 3 and the 80K fused laterals for all grid faults, as well as the effective operation
of recloser 3 with its expanded protection zone during islanded mode. The OT of
recloser 3 is successfully coordinated for all types of faults, ensuring a minimum of
0.12 seconds of CTI following the operation of the 80K fuse during grid faults. On the

82
Figure 5.9: Recloser 3 ground coordination with 80K fuse for grid faults

Figure 5.10: Recloser 3 phase operation for island faults

83
 Figure 5.11: Recloser 3 ground operation for island faults

other hand, recloser 3 also achieves fast OTs during island operation, taking into ac-
count the more pronounced voltage drop. By adapting its MOC based on the voltage
conditions, recloser 3 ensures effective coordination and provides reliable protection
with a single setting.

5.5 Generation Facility Protection in Reverse Di-

rection

The PCC recloser differs from recloser 3 in its placement, as it is positioned at

the connection point between the generation facility and the grid. Consequently, the
PCC recloser and all protective devices within the generation facility are only exposed
to a single range of fault current. As a result, dynamic settings like 51V are no longer

84
necessary. However, due to the presence of two sources in the system, namely the
substation and the generation facility, the protective devices face the potential of fault
current in both directions, requiring bi-directional settings. This section will delve
into the protection settings for the devices operating in the reverse direction, or the
direction of exporting fault current out of the generation facility.

Figure 5.12: Proposed reverse protection zones with the generation facility

The protection diagram, depicted in Fig. 5.12, reveals that downstream of the
generation facility lies recloser 3, which requires coordination. Additionally, 65KS
fused laterals are present downstream of the PCC recloser. However, since the gener-
ation facility cannot generate sufficient fault current to operate the fuse, fuse-saving
measures are not implemented with the PCC recloser. Instead, the islanding recloser,
which serves as the pathway for fault current from the substation, incorporates the
fuse-saving functionality and is presented later.
Additionally, the generation facility contains the PCC relay, BESS relay, and
PV relay, all of which require reverse protection settings. As shownin Fig. 5.12, the
PCC relay serves as the backup for the PCC recloser, while the BESS and PV relays

85
provide backup for the PCC relay.
In the preceding section, recloser 3 was equipped with the 51V element, which
relies on both voltage and current parameters. Consequently, to ensure proper coor-
dination between the PCC recloser and recloser 3, all fault types were individually
considered rather than a coordination range. Tables 5.13 and 5.14 showcase the mea-
surements for maximum and minimum faults within the protection zone of recloser
3 during grid connection and microgrid operation. Meanwhile, to coordinate with
recloser 3, Tables 5.15 and 5.16 exhibit the measured currents at the PCC recloser
during faults within the primary protection zone of recloser 3.

Table 5.13: Fault data recorded at recloser 3 for simulated faults within recloser 3
zone during grid connection

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 927 550
SLG Ground Current [A] 821 509
Line Voltage [kV] 13.64 17.16
Phase Current [A] 1138 777
LLG Ground Current [A] 586 380
Line Voltage [kV] 0 1.86
Phase Current [A] 1068 797
LL
Line Voltage [kV] 0 1.85
Phase Current [A] 1137 867
3PH
Line Voltage [kV] 0 1.85

Given that the generation facility consistently exports fault currents within
a specific range, reaching up to 1.45 pu for both grid and island faults, the protec-
tion settings for the facility are primarily based on traditional overcurrent protection
principles. However, in cases where the MOC limitations cannot be met, voltage su-
pervision is incorporated for reliable operation. In this case, the programming of these
settings only required the consideration of a coordination range. Table 5.17 outlines
the coordination range between the main and backup protection systems within the

86
Table 5.14: Fault data recorded at recloser 3 for simulated faults within recloser 3
zone during microgrid operation

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 290 272
SLG Ground Current [A] 284 248
Line Voltage [kV] 7.38 7.99
Phase Current [A] 334 288
LLG Ground Current [A] 516 384
Line Voltage [kV] 0 1.03
Phase Current [A] 222 214
LL
Line Voltage [kV] 0 0.97
Phase Current [A] 211 209
3PH
Line Voltage [kV] 0 1.03

Table 5.15: Fault data recorded at PCC recloser for simulated faults within recloser
3 zone during grid connection

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 304 257
SLG
Ground Current [A] 380 356
Phase Current [A] 275 251
LLG
Ground Current [A] 271 258
LL Phase Current [A] 233 182
3PH Phase Current [A] 208 201

Table 5.16: Fault data recorded at PCC recloser for simulated faults within recloser
3 zone during microgrid operation

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 290 272
SLG
Ground Current [A] 287 246
Phase Current [A] 334 289
LLG
Ground Current [A] 516 382
LL Phase Current [A] 222 214
3PH Phase Current [A] 211 210

generation facility for faults occurring in the reverse direction.

A CTI margin of 0.12 seconds is considered between the PCC recloser and
the PCC relay, equivalent to the CTI of fuse coordination. This smaller margin was

87
Table 5.17: Coordination of generation facility protection in reverse direction for
maximum and minimum faults within main protection zone

Main Backup Fault Coordination Range [A]

Protection Protection Type Max Fault Min Fault
Phase 344 188
PCC Rec PCC Relay
Ground 546 270
Phase 344 197
PCC Relay BESS/PV Relay
Ground 546 292

taken due to the short distance and absence of load between these devices. For the
remaining relays, a margin of 0.2 seconds is implemented, as shown in Table 5.2, to
ensure adequate coordination between the relays.
Following the methodology employed earlier for determining MOC and protec-
tion settings, Table 5.18 presents the reverse settings for all protective devices within
the generation facility. Meanwhile, Table 5.19 shows the OTs of the main and backup
protection devices within the generation facility in the reverse direction, demonstrat-
ing the coordinated OTs for both the maximum and minimum fault scenarios.

Table 5.18: Generation facility reverse protection settings

Device Function MOC [A] TD Curve Supervisor

51P 91 6.88 U5 ANSI 27
PCC Recloser
51G 55 11.9 U5 None
50/51P 91 1.75 Definite Time ANSI 27
PCC Relay
50/51G 55 1.21 Definite Time None
50/51P 91 1.95 Definite Time ANSI 27
BESS Relay
50/51G 55 1.41 Definite Time None
50/51P 15 1.95 Definite Time ANSI 27
PV Relay
50/51G 5 1.41 Definite Time None

The phase MOCs were chosen to be 50% of the far-end fault current and were
supplemented by undervoltage supervision with a threshold of 0.8 pu. This revised
approach was implemented to ensure reliable operation for all fault scenarios, as the
primary MOC criteria could not be met. On the other hand, the ground MOCs were

88
Table 5.19: OTs of generation facility in reverse direction for maximum and minimum
faults within the main protection zonea

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.03, 0.21 0.04, 0.89 1.45 1.09
Rec 3 PCC Rec
Ground 0.03, 0.58 0.04, 0.89 0.99 1.09
Phase 0.92 1.63 1.75 1.75
PCC Rec PCC Relay
Ground 0.89 1.09 1.21 1.21
Phase 1.75 1.75 1.95 1.95
PCC Relay BESS/PV Relay
Ground 1.21 1.21 1.41 1.41
a
#, # - Represents recloser OTs for fast, slow curves.

established at 200% of the rated ground current.

In Table 5.18, it can be observed that the PCC recloser, PCC relay, and BESS
relay share the same phase MOC of 91 A and ground MOC of 55 A, as the BESS
is the main fault contributor. As mentioned in the beginning of the chapter, it was
found that the PV inverter entered momentary cessation for all faults simulated in the
study. As a precautionary measure, a high-impedance fault scenario was simulated
to ensure that the undervoltage threshold of 0.7 pu was not exceeded. In these
simulations, the PV inverter contributed fault currents ranging from 30 to 70 A.
Although the primary response of the PV inverter to system faults remains to be
momentary cessation, assigning redundant protection measures becomes necessary in
case it fails to enter this state. Consequently, a phase MOC of 15 A was assigned
to the PV relay. Meanwhile, the maximum current unbalance observed at the PV
system was 2.3 A, leading to a ground MOC of 5 A for the PV relay.
The selection of curves and calculation of TD were performed using MATLAB
optimization, to achieve optimal coordination. Based on the results, the PCC, BESS,
and PV relays were configured with definite time curves. In the case of the PCC
recloser, a short-time inverse curve was selected for both phase and ground protection
to best coordinate with recloser 3. Unlike the other reclosers in the system, since the

89
implementation of fuse-saving functionality was not considered in its operation, the
PCC recloser only required a single curve for both phase and ground protection.
Table 5.19 demonstrates the minimum fault at recloser 3 as the leading factor
in developing the PCC phase and ground settings. This is attributed to the dynamic
characteristics of coordinating with the 51V element, but also the results from the
grid-connected fault study. In Table 5.13, it can be seen that recloser 3 experiences
the highest fault current during 3PH faults and the lowest during SLG faults, as
the substation serves as the primary source of fault current. Conversely, Table 5.15
demonstrates that for the same faults, the generation facility contributes the highest
fault current during SLG faults and the least during 3PH faults. This indicates that
in the case of grid-connected faults, a minimum fault at recloser 3 corresponds to
maximum fault current export at the PCC, complicating the method of coordination
between the devices.
In the previous section, recloser 3 was designed to coordinate with an 80K fuse
during grid connection. This coordination resulted in slower OTs for ground faults,
as they were of less magnitude. As a result, recloser 3 exhibited its slowest OT of
0.89 seconds for a SLG fault occurring at the end of its primary protection zone.
Considering the fault observations made at the PCC and coordination requirements,
it was observed that the PCC phase element also needed to coordinate with this OT
to prevent premature operation, as highlighted in Table 5.19.
For the remaining PCC, BESS, and PV relays, the definite time curves were
utilized. As mentioned previously, the BESS inverter is capable of a FRT duration
of 2 seconds for severe faults. Initially, it was observed that the utilizing a standard
TCC for all the relays resulted in optimal coordination for the maximum fault and
a delayed coordination for the minimum fault, exceeding the inverter’s FRT limit.
To address this issue, the PCC relay definite time curve was coordinated with the

90
slowest OT of the PCC recloser, allowing the relay to achieve optimal operation for
the minimum fault within the coordination range. Then, the definite time curves of
the BESS and PV relays were set using the CTI margin. This adjustment successfully
addressed the FRT limitation and ensured coordination requirements were met.
Fig. 5.13 demonstrates the coordination between the PCC recloser and recloser
3 phase curves for faults occurring during grid connection. In this scenario, the PCC
recloser does not measure the same fault current as recloser 3, so the PCC recloser
phase curve is referred to the magnitude of fault current observed by recloser 3, with
the primary aim of demonstrating coordination with the minimum fault. On the
other hand, Fig. 5.14 illustrates the same coordination scenario but during island
operation, where both devices are subjected to the same fault current.
Similarly, Fig. 5.15 adopts the same approach, depicting the coordination
between the PCC recloser and recloser 3 ground curves for faults occurring during
grid connection, with a focus on demonstrating coordination with the minimum fault.
Fig. 5.16 showcases the coordination between these devices during island operation.
Since all devices within the generation facility experience a single range of
fault current, the remaining coordination plots are presented with the entire range of
fault currents from both grid-connected and islanded scenarios. Figs. 5.17 and 5.18
depict the coordination between the PCC recloser and PCC relay. Similarly, Figs.
5.19 and 5.20 showcase the coordination of the BESS relay with the 65 BAY-O-NET
fuse located at the BESS transformer MV terminal. Likewise, Figs. 5.21 and 5.22
illustrates the PV relay’s quicker response compared to the 50 BAY-O-NET fuse at
the PV transformer MV terminal, which does not operate for the specified range.
It is worth noting that in Figs. 5.20 and 5.22, the ground curves for both
the BESS and PV relays are referred to phase current as the fuse operates based
on phase current. Additionally, since the coordination between the PCC relay and

91
Figure 5.13: PCC recloser phase coordination with recloser 3 for grid faults

Figure 5.14: PCC recloser phase coordination with recloser 3 for island faults

92
Figure 5.15: PCC recloser ground coordination with recloser 3 for grid faults

Figure 5.16: PCC recloser ground coordination with recloser 3 for island faults

93
Figure 5.17: PCC relay phase coordination with PCC recloser

Figure 5.18: PCC relay ground coordination with PCC recloser

94
Figure 5.19: BESS relay phase coordination with 65 BAY-O-NET fuse

Figure 5.20: BESS relay ground coordination with 65 BAY-O-NET fuse

95
Figure 5.21: PV relay phase operation

Figure 5.22: PV relay ground operation

96
BESS/PV relay is evident due to the nature of their definite time curves, a TCC for
this coordination is not presented.
These figures illustrate the successful coordination of all devices within the
generation facility for faults in the reverse direction. With these reverse settings in
place, particularly during island operation, the system will effectively isolate the fault
before it reaches the FRT limit of the BESS inverter. The next section will design the
settings for the generation facility in the forward direction. This will ensure reliable
operation in both reverse and forward fault scenarios.

5.6 Generation Facility Protection in Forward Di-

rection

For the generation facility protection, the forward direction refers to fault
currents entering the generation facility or faults occurring within the facility itself.
During grid connection, as depicted in Fig. 5.23, the PCC recloser, PCC relay, BESS
relay, PV relay, and AUX relay are designed to operate on the high-magnitude fault
currents originating from the grid. On the other hand, during island operation, the
fault current contribution is supplied by the BESS. Therefore, only the PV relay and
AUX relay need to detect and respond to forward fault currents.
The protection settings for the devices within the generation facility are de-
signed based on a range of fault currents, following the traditional overcurrent ap-
proach. Table 5.20 shows the coordination range for each set of main and backup
protection in the generation facility for faults occurring in the forward direction. The
coordination range is defined by the extreme fault scenarios encountered during the
fault study, including both grid-connected and islanded scenarios.

97
 Figure 5.23: Proposed forward protection zones with the generation facility

Table 5.20: Coordination of generation facility protection in forward direction for

maximum and minimum faults within main protection zone

Main Backup Fault Coordination Range [A]

Protection Protection Type Max Fault Min Fault
Phase 982 831
PCC Relay PCC Rec
Ground 641 422
Phase 982 831
BESS Relay PCC Relay
Ground 641 422
Phase 1055 211
PV Relay PCC/BESS Relay
Ground 810 572
Phase 1055 115
AUX Relay PCC/BESS Relay
Ground 810 572

In Table 5.20, it can be observed that the PV relay and AUX relay experience
the highest fault current magnitude, which includes the combined contribution from
both the grid and the BESS. It is important to note that these relays are backed up
by the PCC relay in the forward direction and the BESS relay in the reverse direction,
ensuring the disconnection of both sources in the event of a failed operation of the

98
main protection. Meanwhile, the minimum fault within the PV and AUX protection
zones are measured during island operation. Conversely, the coordination range of
the PCC recloser, PCC relay, and BESS relay are equivalent and defined from the
grid-connected fault study.
Similar to the last section, a CTI of 0.12 seconds is taken between the PCC
recloser and PCC relay. Meanwhile, a CTI margin of 0.2 seconds is considered for the
coordination between the remaining protective devices within the generation facility.
Following the same procedure used earlier to determine the MOC and coordinate
the protective devices, Table 5.21 presents the forward settings for all the protective
devices within the generation facility. Table 5.22 displays the associated OTs for the
main and backup protection based on the designed settings.

Table 5.21: Generation facility forward protection settings

Device Function MOC [A] TD Curve Supervisor

50/51P 346 0.32 Definite Time None
PCC Recloser
50/51G 22 0.32 Definite Time None
50/51P 346 0.2 Definite Time None
PCC Relay
50/51G 22 0.2 Definite Time None
50P 346 N/A N/A SHB
BESS Relay
50G 22 N/A N/A SHB
50P 15 N/A N/A SHB
PV Relay
50G 5 N/A N/A SHB
50P 22 N/A N/A SHB
AUX Relay
50G 9 N/A N/A SHB

In Table 5.21, all the selected MOCs are based on 200% the rated current in
the forward direction. The BESS, PV, and AUX relays are designed using the ANSI
50 protection element, which is the instantaneous overcurrent element. Since these
relays protect the first forward protection zone, an instantaneous trip was selected
for each relay, supplemented by a second-harmonic blocking (SHB) supervisor.
The SHB supervisor enables the relays to differentiate between transformer

99
Table 5.22: OTs of generation facility in forward direction for maximum and minimum
faults within the main protection zone

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.2 0.2 0.32 0.32
PCC Relay PCC Rec
Ground 0.2 0.2 0.32 0.32
Phase INST INST 0.2 0.2
BESS Relay PCC Relay
Ground INST INST 0.2 0.2
Phase INST INST 0.2 0.2
PV Relay PCC Relay
Ground INST INST 0.2 0.2
Phase INST INST 0.2 0.2
AUX Relay PCC Relay
Ground INST INST 0.2 0.2

inrush current, which contains second harmonics, and fault current, which does not
contain the second harmonics. This allows the relays to operate only in the presence of
actual fault conditions while avoiding unnecessary tripping during transformer inrush
currents.
The design of the AUX relay also accounted for faults on the AUX trans-
former’s LV side or zone 14 in Fig. 5.12. Although the fault in zone 14 is within the
primary protection zone of the 12T fuse, it is considered part of the AUX relay’s zone
to save the fuse from blowing. Conversely, it is important to emphasize that the LV
side of the BESS and PV transformers are not included in the operation range of the
corresponding relays. The protection of the BESS and PV LV zones are handled by
the vendor of the generation units, while the auxiliary load is managed by the utility.
Table 5.22 shows the operating times of the BESS, PV, and AUX relays, all
of which are instantaneous (INST) due to the use of the ANSI 50 protection element.
Therefore, considering a CTI margin of 0.2 seconds, the PCC relay is configured to
operate on a definite time curve with a 0.2 second delay. Similarly, the PCC recloser
is set to operate on a definite time curve with a 0.32 second delay, with a CTI margin
of 0.12 seconds.

100
 In Fig. 5.24, the coordination between the AUX phase setting and the 12T
fuse is depicted for a range of LV phase faults, spanning from 233 to 115 A. Similarly,
Fig. 5.25 demonstrates the coordination between the AUX ground setting and the
12T fuse for a range of LV ground faults, ranging from 227 to 154 A. These figures
highlight that the operation of the AUX relay is faster than that of the 12T fuse for
transformer LV faults, effectively saving the 12T fuse.
For the remaining protection devices within the generation facility, including
the PCC recloser, PCC relay, BESS relay, and PV relay, the coordination is evident
due to the characteristics of their instantaneous and definite time curves. As a result,
the TCC is not shown for this coordination.

Figure 5.24: AUX relay phase coordination with 12T fuse

Having established the settings outlined in the preceding sections, the coor-
dination and effectiveness of all protective devices within the microgrid have been

101
 Figure 5.25: AUX relay ground coordination with 12T fuse

successfully achieved, rendering the use of only a single setting. This advancement
not only overcomes the drawbacks of the existing protection scheme discussed in
Chapter 4 but also enables continuous protection amidst seamless transition. Sub-
sequently, the upcoming section will delve into the design aspects of the islanding
recloser, the protective device responsible for disconnecting the microgrid from the
main grid.

5.7 Islanding Recloser

Similar to the generation facility, the islanding recloser is equipped with both
forward and reverse protection settings. In the forward direction, it coordinates with
recloser 3, 65KS fused laterals, and the PCC recloser forward settings. This coordi-
nation arrangement can be seen in Fig. 5.26. These forward settings are designed to

102
operate for faults within the microgrid.

Figure 5.26: Proposed protection zones with islanding recloser

On the other hand, in the reverse direction, the islanding recloser must detect
any faults occurring upstream of the microgrid and promptly open its breaker. This
isolates the microgrid from the main grid to continue to provide power to microgrid
customers. The reverse settings will be designed first, followed by the forward settings.
The islanding recloser must conform to the minimum ride-through require-
ments specified in IEEE 1547-2018 Category II, as alluded to in Chapter 3. These re-
quirements necessitate the inclusion of ride-through curves for voltage and frequency,
as depicted in Figs. 1.5 and 1.7, within the settings of the islanding recloser.
To ensure that the islanding recloser remains connected to the grid during
instances such as the generation facility being disconnected or undergoing mainte-
nance, the ride-through limitations are combined with a reverse overcurrent element.
This arrangement prevents the recloser from unnecessary disconnection from the grid.
Fig. 5.27 provides a visual representation of how the various protection elements are
utilized in the design of the reverse settings for the islanding recloser.
Fig. 5.27 demonstrates the implementation of frequency ride-through using

103
 Figure 5.27: Islanding recloser reverse settings

the ANSI 81O (overfrequency) and 81U (underfrequency) protection elements. Sim-
ilarly, the voltage ride-through is achieved through the utilization of the ANSI 59
(overvoltage) and 27 (undervoltage) protection elements. The practical implemen-
tation of utilizing these protection elements and how they are used to model the
corresponding ride-through curves is further discussed in Chapter 6. Nonetheless, it
can be seen by the final AN D gate that the islanding recloser will not operate unless
the instantaneous overcurrent element, ANSI 50, in the reverse direction also trips.
Considering the generation export limits, it was determined that the most
severe reverse load and ground current observed across the islanding recloser were
17 A and 9 A. Therefore, an MOC of 34 A and 18 A were set for the reverse phase

104
and ground overcurrent element, ensuring the detection of excessive current flow in
the reverse direction. Furthermore, a simulated example depicting an unplanned
islanding scenario for the farthest upstream fault was presented in Chapter 3, Section
2 of the thesis.
As shown in Fig. 5.26, the design of the forward protection settings for the is-
landing recloser involves incorporating fuse-saving functionality in coordination with
the 65KS fused laterals, as well as providing backup protection for recloser 3 and
the PCC recloser. Therefore, the forward settings of the islanding recloser was pro-
grammed with both a fast and a slow curve.
The fast curve of the islanding recloser was coordinated with the forward
settings of the PCC recloser. This coordination ensures that in the event of a fault
occurring within the generation facility, the forward settings of the generation facility
will operate before the fast curve of the islanding recloser. This arrangement is crucial
as it allows the generation facility’s protective settings to respond first, preventing
unnecessary reclosing of the entire microgrid for a fault within the generation facility.
As recloser 3 dynamically responds to different faults based on voltage and
current, a comprehensive analysis of each fault point was necessary for effective co-
ordination. Unlike the coordination between the PCC recloser and recloser 3, where
various scenarios were taken into account, the coordination between the islanding
recloser and recloser 3 only needed to consider the grid-connected scenario. In Table
5.13, which was previously shown, the measurements at recloser 3 for both maximum
and minimum faults within its primary zone are listed. In parallel, Table 5.23 pro-
vides the corresponding measurements at the islanding recloser for those same fault
points.
For the remaining device coordination, a coordination margin can be defined
to ensure sufficient margin between the device curves. The coordination range with

105
the PCC recloser forward settings and the 65KS fused laterals can be found detailed
in Table 5.24. As done previously, for coordination with the 65KS fused laterals, the
extreme fault scenarios across all laterals were considered within the range.

Table 5.23: Fault data recorded at islanding recloser for simulated faults within re-
closer 3 zone during grid connection

Fault Type Measured Value Maximum Fault Minimum Fault

Phase Current [A] 620 541
SLG
Ground Current [A] 506 487
Phase Current [A] 862 777
LLG
Ground Current [A] 396 377
LL Phase Current [A] 855 761
3PH Phase Current [A] 943 852

Table 5.24: Coordination of islanding recloser in forward direction for maximum and
minimum faults

Main Backup Fault Coordination Range [A]

Protection Protection Type Max Fault Min Fault
Phase 982 831
PCC Rec Island Rec
Ground 641 604
Phase 963 853
65KS Fuse Island Rec
Ground 556 508

A CTI of 0.12 seconds is taken between the 65KS fuse and islanding recloser.
Meanwhile, a CTI margin of 0.2 seconds is employed to coordinate the islanding
recloser with recloser 3 and the PCC recloser. Following the established procedure
for developing the settings, the resulting forward settings for the islanding recloser
are presented in Table 5.25. Furthermore, Table 5.26 provides the associated OTs for
the main and backup protection based on the designed settings.
After employing MATLAB optimization, an inverse curve was chosen for the
islanding recloser slow curves. Specifically, the inverse curve provided the fastest
operation for faults within the lower range of coordination. Although steeper TCC

106
 Table 5.25: Islanding recloser forward protection settings

Function MOC [A] TD Curve

51P- Slow Curve 430 0.69 U2
51G- Slow Curve 87 6.79 U2
51P- Fast Curve 430 0.51 U3
51G- Fast Curve 87 3.05 U3

Table 5.26: OTs of islanding recloser and coordinated devices in forward direction for
maximum and minimum faultsa

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.32 0.32 0.52, 1.09 0.77, 1.62
PCC Rec Island Rec
Ground 0.32 0.32 0.52, 1.98 0.54, 2.07
Phase 1.02 1.26 0.54, 1.14 0.71, 1.51
65KS Fuse Island Rec
Ground 2.03 2.32 0.59, 2.23 0.65, 2.44
Phase 0.03, 0.21 0.04, 0.34 0.57, 1.19 0.96, 1.98
Rec 3 Island Rec
Ground 0.03, 0.58 0.04, 0.89 0.65, 2.44 0.68, 2.55
a
#, # - Represents recloser OTs for fast, slow curves.

curves can lead to faster OTs for higher magnitude faults, it was observed that lower
magnitude faults were approaching the 3 second limitation as these curves were too
steep in comparison to the 65KS fuse. Unlike ”T” or ”K” fuses, ”KS” fuses have
a distinct characteristic of being slower in terms of their slope. Therefore, when
considering the coordination with the 65KS fuse, the inverse curve demonstrated the
closest slope to that of the fuse within the margin of coordination. Although it is
possible to replace the 65KS fused laterals with a faster fuse, the present study does
not consider resizing the fuses due to practical cost constraints.
For the fast curves of the islanding recloser, a very inverse curve was chosen
for both the phase and ground settings, as it provided the fastest operating time
before the 65KS fuse. Consequently, the fast curve of the islanding recloser maintains
a sufficient CTI margin of 0.2 seconds with respect to the 65KS minimum melting
curve.

107
 Figs. 5.28 and 5.29 demonstrate the coordination between the islanding re-
closer and the 65KS fuse. It is important to note that the islanding recloser ground
curves are referred to the phase in order to showcase their coordination with the 65KS
fuse.

Figure 5.28: Islanding recloser phase coordination with 65KS fuse

Figs. 5.30 and 5.31 highlight the coordination between the islanding recloser
and recloser 3, displaying the 51V base curves. While the previous method of pre-
senting the 51V curves could have been employed, it is important to note that the
coordination with the 65KS fuse primarily influenced the definition of the islanding
recloser slow curve. Consequently, both figures effectively illustrate the coordination
between the islanding recloser and the 51V base curves of recloser 3.
Finally, the coordination between the islanding recloser fast curves and the
PCC recloser forward settings are illustrated in Figs. 5.32 and 5.33.

108
Figure 5.29: Islanding recloser ground coordination with 65KS fuse

Figure 5.30: Islanding recloser phase coordination with recloser 3

109
 Figure 5.31: Islanding recloser ground coordination with recloser 3

Figure 5.32: Islanding recloser phase coordination with PCC recloser forward settings

110
Figure 5.33: Islanding recloser ground coordination with PCC recloser forward set-
tings

The TCC plots serve as evidence of the successful implementation of fuse-

saving with the 65KS fuse. They also demonstrate that the islanding recloser fast
curves do not conflict with the operation of the PCC recloser forward settings. Addi-
tionally, these plots demonstrate that the islanding recloser’s fast curves do not clash
with the operation of the PCC recloser’s forward settings. In the subsequent section,
the design of settings for recloser 2 will be addressed.

5.8 Recloser 2

Based on the previous section, the islanding recloser was specifically designed
to isolate the microgrid from the main grid in the event of an external fault. Conse-
quently, recloser 2 solely requires forward settings. Fig. 5.34 illustrates that recloser

111
2 acts as backup protection for both the islanding recloser and 80KS fused laterals in
the forward direction.

Figure 5.34: Proposed protection zones with recloser 2

To establish appropriate settings for recloser 2, coordination ranges are defined

and outlined in Table 5.27. These ranges are crucial for ensuring effective coordination
among the devices involved.

Table 5.27: Coordination of recloser 2 for maximum and minimum faults

Main Backup Fault Coordination Range [A]

Protection Protection Type Max Fault Min Fault
Phase 983 835
Island Rec Rec 2
Ground 665 532
Phase 1299 1169
80KS Fuse Rec 2
Ground 798 714

A CTI of 0.12 seconds is considered between the 80KS fuse and recloser 2.
Meanwhile, a CTI of 0.2 seconds is utilized between recloser 2 and the islanding
recloser forward settings. Following the previous procedure for selecting MOCs and
developing settings, the settings for recloser 2 are presented in Table 5.28. Table
5.29 provides the OTs for both the main and backup protection based on the defined
coordination range.
In line with the design approach taken for the islanding recloser, inverse curves

112
 Table 5.28: Recloser 2 protection settings

Function MOC [A] TD Curve

51P- Slow Curve 484 0.92 U2
51G- Slow Curve 99 7.38 U2
51P- Fast Curve 484 0.56 U3
51G- Fast Curve 99 3.94 U3

Table 5.29: OTs of recloser 2 and coordinated devices for maximum and minimum
faultsa

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.52, 1.09 0.75, 1.61 0.72, 1.84 1.04, 2.82
Island Rec Rec 2
Ground 0.5, 1.92 0.61, 2.33 0.7, 2.25 0.82, 2.59
Phase 0.89 1.04 0.4, 1.01 0.5, 1.29
80KS Fuse Rec 2
Ground 1.83 2.07 0.62, 2.01 0.67, 2.19
a
#, # - Represents recloser OTs for fast, slow curves.

were carefully selected for the recloser’s slow curves to ensure optimal coordination
with the 80KS fuse while maintaining effective coordination with the islanding re-
closer. On the other hand, very inverse curves were chosen for the recloser’s fast
curves to achieve the best possible coordination with the fast curves of the islanding
recloser. Furthermore, when coordinating with the slow ”KS” fuses, the OTs speci-
fied in Table 5.29 remain below the 3-second limit. This ensures that the protection
requirements are met while providing sufficient protection against faults.
Figs. 5.35 and 5.36 visually showcase the coordination between the 80KS fuse
and recloser 2, demonstrating an ample margin to enable fuse-saving functionality for
the 80KS fuse. As done with the previous TCC plots, the ground curve of recloser 2
in Fig. 5.36 is referred to phase current to demonstrate fuse coordination. Similarly,
Figures 5.37 and 5.38 illustrate the coordination between the islanding recloser and
recloser 2 slow and fast curves, with a CTI of 0.2 seconds maintained between each
respective curve.

113
Figure 5.35: Recloser 2 phase coordination with 80KS fuse

Figure 5.36: Recloser 2 ground coordination with 80KS fuse

114
Figure 5.37: Recloser 2 phase coordination with islanding recloser

Figure 5.38: Recloser 2 ground coordination with islanding recloser

115
 Based on the plots, the settings designed for recloser 2 have resulted in suc-
cessful coordination with the corresponding downstream devices, while also ensuring
compliance with the 3-second OT limit. Consequently, in the upcoming section, the
settings for the final protective device will be formulated.

5.9 Recloser 1

Recloser 1, positioned furthest upstream in the system as depicted in Fig.

5.39, plays a vital role in coordinating with various devices including the 100KS fused
laterals, recloser 2, and recloser 4.

Figure 5.39: Proposed protection zones with recloser 1

Similar to recloser 2, recloser 1 only requires forward protection settings. How-

ever, it is worth noting that recloser 4 is a hydraulic recloser with a fixed coil rating
of 140 A, utilizing standard Cooper A and B curves. Consequently, the settings for
recloser 4 cannot be programmed or modified, and are fixed similar to fuse coordina-
tion.
To establish effective coordination, the range of coordination between recloser
1 and the respective protective devices is outlined in Table 5.30. This table includes
the extreme fault values measured during the fault study discussed in the first section
of this chapter.

116
 Table 5.30: Coordination of recloser 1 for maximum and minimum faults

Main Backup Fault Coordination Range [A]

Protection Protection Type Max Fault Min Fault
Phase 1598 895
Rec 2 Rec 1
Ground 829 612
Phase 2236 1652
Rec 4 Rec 1
Ground 1122 937
Phase 1695 1541
100KS Fuse Rec 1
Ground 983 836

To achieve coordination with the 100KS fuse, a CTI margin of 0.12 seconds.
Similarly, a CTI margin of 0.2 seconds is employed for coordination with recloser 2.
However, a CTI of 0.3 seconds is established between the hydraulic recloser (recloser
4) and recloser 1, as indicated in Table 5.2.
Utilizing these parameters and following the same established procedure, the
designed settings for recloser 1 are presented in Table 5.31. Furthermore, the OTs
resulting from the defined coordination range can be found in Table 5.32.

Table 5.31: Recloser 1 protection settings

Function MOC [A] TD Curve

51P- Slow Curve 662 2.04 U1
51G- Slow Curve 137 7.82 U1
51P- Fast Curve 662 2.76 U5
51G- Fast Curve 137 8.85 U5

As shown in Table 5.31, recloser 1 is equipped with moderately inverse curves

for both the phase and ground slow curves. Meanwhile, short-time inverse curves were
selected for the phase and ground fast curves. These curve selections were determined
through multiple test scenarios conducted in MATLAB, ensuring that the recloser
achieves fast OTs while also maintaining the slowest OT below the 3-second limit.
The iterative testing process also maintained coordination with all relevant devices
within the system.

117
Table 5.32: OTs of recloser 1 and coordinated devices for maximum and minimum
faultsa

Main Backup Fault Main OTs [s] Backup OTs [s]

Protection Protection Type Max Fault Min Fault Max Fault Min Fault
Phase 0.27, 0.72 0.95, 2.44 0.5, 1.16 1.15, 2.62
Rec 2 Rec 1
Ground 0.6, 1.93 0.78, 2.5 0.82, 2.2 0.98, 2.85
Phase 0.03, 0.15 0.05, 0.22 0.39, 0.9 0.51, 1.19
Rec 4 Rec 1
Ground 0.06, 0.34 0.08, 0.43 0.72, 2.07 0.79, 2.25
Phase 0.91 1.05 0.5, 1.15 0.56, 1.29
100KS Fuse Rec 1
Ground 2.05 2.24 0.77, 2.19 0.84, 2.36
a
#, # - Represents recloser OTs for fast, slow curves.

Figs. 5.40 and 5.41 provide a visual representation of the successful coor-
dination achieved between the 100KS fuse and recloser 1, with the recloser ground
curve referred to phase current. Figs. 5.42 and 5.43 exhibit the coordination be-
tween recloser 2 and recloser 1, maintaining a minimum CTI of 0.2 seconds. Lastly,
Figures 5.42 and 5.43 showcase the coordination between recloser 2 and recloser 4,
demonstrating a minimum CTI margin of 0.3 seconds.
The TCC plots provide clear evidence of the maintained effective margin be-
tween each protective device amidst the different curve slopes and OTs. Overall, all
the TCC plots in this chapter demonstrate that all the coordinated protection de-
vices operate within the specified time limit of 3 seconds. It is important to note that
while faster operating times could have been achieved by altering the ”KS” fuses, this
protection design prioritizes the utilization of the existing protection elements within
the system.
In conclusion, the system has been effectively designed to operate with single-
settings, eliminating the need for communication between devices. Additionally, the
system is immune to seamless transitions, ensuring smooth and reliable protection,
while utilizing the available protection features.

118
Figure 5.40: Recloser 1 phase coordination with 100KS fuse

Figure 5.41: Recloser 1 ground coordination with 100KS fuse

119
Figure 5.42: Recloser 1 phase coordination with recloser 2

Figure 5.43: Recloser 1 ground coordination with recloser 2

120
Figure 5.44: Recloser 1 phase coordination with recloser 4

Figure 5.45: Recloser 1 ground coordination with recloser 4

121
Chapter 6

Testing Adaptive Protection

Scheme using HIL Simulations

In order to validate the functionality of the protection scheme outlined in

Chapter 5, hardware-in-the-loop (HIL) simulations were conducted. To carry out
these simulations, the designed settings were programmed into legacy relays, which
were then connected to a multi-processor computer system called the real-time digital
simulator (RTDS). This setup allowed for effective demonstration of the protection
scheme’s dependability, selectivity, and security.
To facilitate the HIL simulations, the PSCAD system was converted to the
RSCAD platform [39]. This conversion enabled the execution of real-time simulations
on the RTDS, which is intended for power system simulations. The subsequent section
will delve into the details of how the system was converted and modeled on the
RSCAD platform.

122
6.1 Testing Background and Parameters

The distribution system was initially modeled using PSCAD simulation, as

explained in Chapter 2, to develop the proposed protection scheme. However, for
the purpose of HIL real-time simulation, the conversion of the PSCAD model to the
RSCAD platform was necessary. Although RSCAD provides a conversion tool from
PSCAD, it had limitations in successfully converting certain components. Specifi-
cally, the loads and the generation facility in the system were not converted properly,
while the PI sections, regulators, capacitors, and substation source were successfully
converted.
To simplify the load modeling process, each load lateral in PSCAD was consol-
idated into a single load in RSCAD. This not only streamlined the modeling process
but also ensured that the RTDS, which had a maximum capacity of 6 cores, could
handle the number of system load components. The loads were modeled as fixed loads,
similar to the PSCAD model. However, the largest commercial customer within the
microgrid, with a maximum rated capacity of 1000 kVA, was modeled using a 1000
kVA induction motor. This was achieved by utilizing the ”Induction Machine” model
available in RSCAD, where the motor was connected around the voltage zero-crossing
and ramp rate was reduced to simulate severe startup and inrush. The objective of
this choice was to later validate the reliability and security of the protection scheme by
confirming that the protection does not trigger false operations from motor starting.
When it came to modeling the generation facility, only the BESS generation
was considered, as the PV inverter did not contribute to any faults studied in Chapter
5. Since the PSCAD inverter was a blackbox vendor model, the Clemson inverter
described in Chapter 2, Section 3 was utilized and adjusted to exhibit a similar
system and fault response. To emulate the fault contribution of the PSCAD inverter,

123
the fault current limiter of the RSCAD inverter was dynamically modified based on
the type of fault, as indicated in Table 6.1. Furthermore, the RSCAD inverter was
set with a real power droop of 0.5 and a reactive power droop of 0.1, denoted as n
and m in Fig. 2.6 of Chapter 2.

Table 6.1: Fault current limiter setpoint for BESS inverter based on fault type

Fault Type Fault Current Limit Rating [pu]

SLG 1.08
LLG 1.18
LL 1.45
3PH 1.25

The RSCAD inverter model resulted in a system and fault response within a
maximum margin of error of 4.2% in comparison with PSCAD model. This assessment
was made by comparing the currents and voltages obtained from both simulations. In
Tables 6.2 and 6.3, a comprehensive comparison of the steady-state currents and volt-
ages during the normal operation of the microgrid, as simulated in both PSCAD and
RSCAD, is presented. Furthermore, Table 6.4 provides a comparison of the microgrid
fault currents contributed by the BESS inverter. While the RSCAD inverter did not
perfectly match the PSCAD model, it should be noted that the primary objective of
the HIL simulations was to assess the performance of the proposed protection scheme.

Table 6.2: Comparison of PSCAD and RSCAD steady-state currents during microgrid
operation

Percent Distance Phase A Current [A] Phase B Current [A] Phase C Current [A]
from PCC PSCAD RSCAD PSCAD RSCAD PSCAD RSCAD
0% 137 138 133 135 136 136
25% 122 119 129 125 126 121
50% 104 96 98 89 94 90
75% 54 49 68 62 42 40

To conduct the HIL testing of the protection scheme, the SEL platform was

124
 Table 6.3: Comparison of PSCAD and RSCAD steady-state voltages during microgrid
operation

Percent Distance Phase A LG Voltage [kV] Phase B LG Voltage [kV] Phase C LG Voltage [kV]
from PCC PSCAD RSCAD PSCAD RSCAD PSCAD RSCAD
0% 13.69 13.71 13.73 13.74 13.71 13.72
25% 13.67 13.69 13.7 13.72 13.7 13.71
50% 13.66 13.66 13.69 13.67 13.67 13.66
75% 13.62 13.63 13.66 13.65 13.64 13.63
100% 13.6 13.6 13.63 13.61 13.61 13.59

Table 6.4: Comparison of PSCAD and RSCAD fault currents during microgrid oper-
ation

Percent Distance SLG Fault [A] LLG Fault [A] LL Fault [A] 3PH Fault [A]
from PCC PSCAD RSCAD PSCAD RSCAD PSCAD RSCAD PSCAD RSCAD
0% 293 295 344 346 223 222 212 214
25% 290 289 334 331 222 221 211 210
50% 283 286 309 311 220 219 211 210
75% 274 278 289 281 216 214 210 209

employed. In these simulations, three SEL relays were utilized in conjunction for fault
testing, while the remaining system relays were modeled in RSCAD using software
relays. The selection of relays depended on the simulated fault location, with a focus
on using physical relays for faults in the main and backup protection zones, and
software relays for the remainder of the system.
Two SEL 421 relays were connected to the RTDS using Sampled Value (SV)
communication, allowing them to receive system data through communication [40].
The third relay, an SEL 411, was connected to the RTDS through low-voltage wiring
from the front of the relay [41]. This wiring arrangement was necessary as no am-
plifier was available to amplify the RTDS output and interface it with the relay.
Although these relays have the capability to serve various protection purposes, they
were specifically equipped with the required protection elements for testing the pro-

125
posed protection scheme. These elements included time-overcurrent, overvoltage,
undervoltage, overfrequency, underfrequency, and directional protection.
The subsequent section will explore the construction of the relay settings and
the customized programming of the voltage-restrained time-overcurrent element, pro-
viding additional insights into its integration within the relay.

6.2 Relay Programming

To establish protection settings on the SEL platform, the software application

”acSELerator QuickSet” was employed. In general, the protection settings relied
on the standard time-overcurrent logic templates found in the relays, namely the
protection templates for 51P, 51G, 50P, and 50G. However, the 51V element required
custom programmed logic.
For protection devices utilizing directional overcurrent, the relays were pro-
grammed with the provided line parameters and set to ”AUTO” mode. This allowed
the relays to automatically calculate the settings for the directional element [40]. With
the initialized directional element, it provided three states: one for the forward direc-
tion, one for the reverse direction, and one for the unknown state. This third state
occurred during changes of power flow direction, where the relay can require up to a
cycle to determine the correct direction. Nonetheless, only the forward and reverse
state elements were utilized as overcurrent supervisors for bi-directional settings.
Meanwhile, for protection devices incorporating voltage supervision (27 and
59) or frequency supervision (81), the limits were configured within the correspond-
ing template of relay logic. Then, these supervision elements were implemented in
conjunction with the ”Trip Logic” of the overcurrent element within the relay.
In programming the settings for the islanding recloser, as mentioned in Chapter

126
5, the IEEE 1547-2018 voltage and frequency ride-through curves were programmed
into the recloser. This was done by sampling the ride-through curves in Figs. 1.5
and 1.7 with their corresponding ride-through times, eliminating the need for custom
curve programming. The specific settings for the undervoltage protection element
can be found in Table 6.5, which includes the associated voltage limits and ride-
through times. Similarly, the overvoltage protection element settings can be seen in
Table 6.6. The underfrequency protection element settings are presented in Table 6.7
with the associated frequency limits and ride-through times, while the overfrequency
protection element settings are displayed in Table 6.6. Within the islanding recloser,
these supervisor elements were integrated within the ”Trip Logic” in conjunction with
reverse overcurrent, as previously illustrated in Fig. 5.27.

Table 6.5: Undervoltage ride-through settings for islanding recloser

Voltage [pu] Ride-Through Time [s]

0.88 5
0.8 4.31
0.7 3.44
0.65 0.32
0.45 0.16
0.3 INST

Table 6.6: Overvoltage ride-through settings for islanding recloser

Voltage [pu] Ride-Through Time [s]

1.10 1
1.15 0.5
1.175 0.2
1.2 INST

In order to implement the custom 51V logic for recloser 3, as discussed in

Chapter 5, reference was made to an SEL application guide documented in [42]. The
finalized protection logic code can be found in Appendix B.

127
 Table 6.7: Underfrequency ride-through settings for islanding recloser

Frequency [Hz] Ride-Through Time [s]

58.8 299
57 INST

Table 6.8: Overfrequency ride-through settings for islanding recloser

Frequency [Hz] Ride-Through Time [s]

61.2 299
62 INST

The protection logic was programmed to incorporate ”Protection SELogic

math variables” in order to set dynamic pickup values for the phase and ground
inverse time-overcurrent elements [42]. This means that the overcurrent pickup val-
ues were defined as variables that could be adjusted during operation.
The protection logic implemented in the system compares the three measured
line voltages at the recloser and determines the minimum voltage among them. This
minimum voltage value is then used to adjust the pickup of the recloser. To deter-
mine the appropriate pickup value based on the voltage drop, a pickup multiplier is
calculated by dividing the voltage drop by the voltage reference. This relationship is
illustrated in Fig. 5.2 from the previous chapter.
By incorporating the pickup values as variables within the overcurrent ele-
ment, the 51V logic is seamlessly integrated into the recloser’s overcurrent protection
settings. This integration enables the recloser to respond accurately and adaptively
to various fault conditions by utilizing the voltage drop as a basis for changing its
operation.
The following section will delve into the discussion of experimental results
pertaining to various fault scenarios. These results will serve as evidence of the
successful operation of the proposed protection scheme with single settings outlined

128
in Chapter 5.

6.3 Feeder Protection Test

The proposed protection scheme was validated through the use of the RSCAD
distribution system and legacy relays wired in HIL with the RTDS. The testing process
involved simulating a variety of faults, resembling those analyzed in the PSCAD fault
study, across different sections of the system. A visual representation of the tested
faults across the system can be observed in Fig. 6.1.
Fig. 6.1 displays all the tested faults during grid connection, highlighted in
orange or magenta color. However, the faults highlighted in magenta are significant
as they are chosen for in-depth analysis in this section. These seven fault scenar-
ios effectively summarize the successful operation of the proposed protection scheme
discussed in Chapter 5 during grid connection.
To enhance understanding of the protective operation during each analyzed
fault, Tables 6.9 and 6.10 present detailed information on the main and backup pro-
tective devices associated with each fault, considering both the forward and reverse
directions. Notably, the numerical fault numbers listed in these tables directly corre-
spond to the fault numbers indicated in Fig. 6.1.

Table 6.9: Main and backup protection for grid-connected faults in forward direction

Fault Number Main Protection Backup Protection

1 Recloser 2 Recloser 1
2 Islanding Recloser Recloser 2
3 PCC Recloser Islanding Recloser
4 Recloser 3 Islanding/PCC Recloser
5 80K Fuse Recloser 3
6 80K Fuse Recloser 3
7 AUX Relay PCC/BESS Relay

129
130
Figure 6.1: RSCAD simulated faults during grid connection
Table 6.10: Main and backup protection for grid-connected faults in reverse direction

Fault Number Main Protection Backup Protection

1 Islanding Recloser PCC Recloser
2 PCC Recloser PCC Relay
3 PCC Relay BESS Relay

These seven faults are simulated to test the proposed single settings for each
protective device, with each device serving as either main or backup protection. First,
these settings will be evaluated for faults that occur with the presence of the substa-
tion source and the generation facility.
The experiment demonstrates the operation of recloser 3 with its proposed
dynamic 51V element through the simulation of faults 4, 5, and 6. Both faults 5
and 6 are considered to accurately demonstrate the coordination of the dynamic 51V
element with the 80K fuse for all range of faults.
Additionally, the experiment validates the proposed forward settings of the
islanding recloser using faults 2 and 3, while the proposed reverse settings of the
islanding recloser are tested through fault 1. The operation of the proposed forward
settings of the PCC recloser are verified using fault 3, and the proposed reverse
settings are tested with faults 2 and 4. Similarly, the PCC relay’s proposed reverse
settings are evaluated with fault 2 and 3, and the relay’s proposed forward settings
are tested using fault 7. The BESS relay’s proposed reverse settings are validated
using fault 3. Furthermore, the testing includes the assessment of the AUX relay’s
forward settings using fault 7.
Although it would have been possible to demonstrate additional faults within
the generation facility to showcase the effective operation of all the proposed forward
settings, it was deemed sufficient to focus on faults 3 and 7, considering that all the
devices within the generation facility employ similar forward settings.

131
 To provide a comprehensive understanding of how each fault was tested in
RSCAD using HIL, Figs. 6.2, 6.3, 6.4, and 6.5 showcase the 51V operation of recloser
3 for maximum faults occurring within the 80K fuse zone. These fault scenarios
correspond to fault 5 in Fig. 6.1, where the 80K fuse serves as the primary protective
device and the 51V curve of recloser 3 functions as backup.

Figure 6.2: Operation of Recloser 3 slow curve for AG fault within 80K fuse zone

Within each figure, the first subplot displays the measured line voltage drop
at recloser 3, which enables the dynamic adjustment of the MOC by the 51V element.
In the second subplot, the figures display the measured fault current, giving insights
into the magnitude of the current during the fault event. The third subplot presents
the trip signal, indicating the precise moment when the recloser detects the fault and
initiates the breaker operation in the RSCAD simulation. It is important to note
that the interruption of current on each phase does not occur simultaneously; rather,

132
Figure 6.3: Operation of Recloser 3 slow curve for CAG fault within 80K fuse zone

Figure 6.4: Operation of Recloser 3 slow curve for BC fault within 80K fuse zone

133
Figure 6.5: Operation of Recloser 3 slow curve for ABC fault within 80K fuse zone

the breaker in simulation operates specifically at the zero crossing of the current
waveform.
The remaining faults tested in this chapter will be presented in a tabulated
format, showcasing their experimental OTs. These OTs will then be compared with
the OTs of the designed protection scheme discussed in the previous chapter.
As 51V operation is dependent on both voltage and current, the OTs for faults
related to recloser 3 are presented in detail, with the OT for each specific fault (SLG,
LLG, LL, or 3PH faults) individually shown. However, for the remaining devices, the
fastest and slowest OTs are shown, representing the range of OTs for each device.
In Table 6.11, the designed and experimental OTs for each protective device
are categorized based on fault number, showing the fastest and slowest OTs for faults
at that location with the generation facility connected. Meanwhile, Table 6.12 follows

134
 a similar approach but presents a more detailed comparison between the experimental
OTs for recloser 3 and the OTs derived from the designed settings.

Table 6.11: Designed and experimental OTs during grid connectiona

Fault Protective Protection Designed OT [s] Experimental OT [s]

Direction
Number Device Type Fastest Slowest Fastest Slowest
Rec 2 Main Forward 0.3, 0.79 0.63, 1.95 0.29, 0.78 0.61, 1.92
Island Rec Main Reverse INST INST 0.02 0.02
1
Rec 1 Backup Forward 0.53, 1.18 0.84, 2.23 0.52, 1.18 0.83, 2.21
PCC Rec Backup Reverse 0.98 1.52 1.0 1.56
Island Rec Main Forward 0.56, 1.17 0.53, 1.96 0.58, 1.19 0.55, 1.98
PCC Rec Main Reverse 0.97 1.51 0.98 1.51
2
Rec 2 Backup Forward 0.74, 1.88 0.73, 2.31 0.76, 1.87 0.74, 2.29
PCC Relay Backup Reverse 1.21 1.75 1.22 1.76
PCC Rec Main Forward 0.32 0.32 0.33 0.33
PCC Relay Main Reverse 1.21 1.75 1.22 1.76
3
Island Rec Backup Forward 0.56, 1.18 0.55, 1.98 0.55, 1.17 0.55, 1.97
BESS Relay Backup Reverse 1.41 1.95 1.41 1.95
Rec 3 Main Forward 0.03, 0.21 0.03, 0.58 0.03, 0.2 0.03, 0.59
4 Island Rec Backup Forward 0.57, 1.19 0.65, 2.44 0.55, 1.18 0.63, 2.41
PCC Rec Backup Reverse 0.99 1.45 0.98 1.46
AUX Relay Main Forward INST INST 0.02 0.02
7
PCC Relay Backup Forward 0.2 0.2 0.2 0.2
BESS Relay Backup Reverse 1.41 1.95 1.41 1.95
a
#, # - Represents recloser OTs for fast, slow curves.

Tables 6.11 and 6.12 demonstrate the close alignment between the experimen-
tal OTs and the OTs derived from the proposed protection scheme design. However,
during the HIL testing of the proposed scheme, it was observed that the directional
element occasionally experienced a slight delay of 0.01 to 0.02 seconds, specifically
when there was a change of direction. Nonetheless, the CTI employed in the pre-
vious chapter incorporated a safety margin to accommodate any additional delays
effectively.
Overall, based on the experimental results, the proposed single settings for
each protective device have demonstrated successful operation in line with the design,

135
 Table 6.12: Designed and experimental OTs of recloser 3 during grid connectiona

Fault Number Fault Type Designed OT [s] Experimental OT [s]

SLG 0.03, 0.58 0.03, 0.59
LLG 0.03, 0.21 0.03, 0.2
4
LL 0.03, 0.21 0.03, 0.2
3PH 0.03, 0.21 0.03, 0.2
SLG 0.03, 0.63 0.04, 0.65
LLG 0.04, 0.35 0.04, 0.37
5
LL 0.04, 0.36 0.04, 0.38
3PH 0.04, 0.35 0.04, 0.32
SLG 0.04, 0.96 0.04, 0.97
LLG 0.04, 0.61 0.04, 0.62
6
LL 0.04, 0.59 0.04, 0.59
3PH 0.04, 0.55 0.04, 0.56
a
#, # - Represents recloser OTs for fast, slow curves.

considering the combined presence of the substation source and generation facility.
Next, the same settings are tested with solely substation connection, i.e., without the
generation facility being connected.
The analysis with the substation connection focuses on the same faults consid-
ered in Fig. 6.1, with the exception of faults 3 and 7. These two faults are excluded
as they occur within the generation facility, which is now disconnected. Moreover,
in the absence of the generation facility, all protective devices operate in a forward
direction only. Fig. 6.6 displays the analyzed faults, following the previous naming
convention, with faults 3 and 7 removed to reflect the disconnection of the generation
facility.
Table 6.13 provides a comparison between the experimental and designed OTs
for the fastest and slowest OTs at each fault location, similar to the previous tabulated
data. On the other hand, Table 6.14 specifically focuses on the OTs of recloser 3 for
each fault type, highlighting the successful operation of the 51V element within the
recloser.

136
137
Figure 6.6: RSCAD simulated faults without generation facility
 Table 6.13: Designed and experimental OTs without the generation facilitya

Fault Protective Protection Designed OT [s] Experimental OT [s]

Direction
Number Device Type Fastest Slowest Fastest Slowest
Rec 2 Main Forward 0.27, 0.72 0.6, 1.93 0.25, 0.69 0.61, 1.95
1
Rec 1 Backup Forward 0.5, 1.16 0.82, 2.2 0.48, 1.13 0.81, 2.19
Island Rec Main Forward 0.52, 1.09 0.5, 1.92 0.5, 1.07 0.5, 1.91
2
Rec 2 Backup Forward 0.72, 1.84 0.7, 2.25 0.74, 1.86 0.71, 2.25
Rec 3 Main Forward 0.04, 0.26 0.05, 0.61 0.03, 0.23 0.05, 0.6
4
Island Rec Backup Forward 0.58, 1.2 0.66, 2.46 0.56, 1.19 0.67, 2.45
a
#, # - Represents recloser OTs for fast, slow curves.

Table 6.14: Designed and experimental OTs of recloser 3 without the generation
facilitya

Fault Number Fault Type Designed OT [s] Experimental OT [s]

SLG 0.05, 0.61 0.05, 0.6
LLG 0.05, 0.33 0.05, 0.36
4
LL 0.05, 0.31 0.05, 0.34
3PH 0.04, 0.26 0.03, 0.23
SLG 0.06, 0.69 0.06, 0.71
LLG 0.06, 0.41 0.06, 0.4
5
LL 0.06, 0.42 0.06, 0.41
3PH 0.05, 0.39 0.05, 0.38
SLG 0.06, 0.94 0.04, 0.97
LLG 0.06, 0.58 0.04, 0.61
6
LL 0.06, 0.56 0.04, 0.59
3PH 0.06, 0.53 0.04, 0.55
a
#, # - Represents recloser OTs for fast, slow curves.

Under the same protection settings, all protective devices performed in align-
ment with the design. Furthermore, it was also confirmed that the islanding recloser
did not operate for upstream faults due to the absence of the connected generation
facility. In summary, during the HIL experimentation, the proposed single protection
settings demonstrated successful operation in the event of grid faults. All devices ad-
hered closely to the intended design requirements, remaining resilient to the presence
of the generation facility. The subsequent section will concentrate on evaluating these

138
identical settings within the context of microgrid operation.

6.4 Microgrid Protection Test

During microgrid operation, the generation facility assumes sole responsibility

as the source of fault current. This section aims to establish that, without making
any adjustments to the device settings, all microgrid protection will effectively operate
within the limitations of the fault current generated by the facility.
For the analysis in this section, the same previous fault scenarios depicted in
Fig. 6.1 are considered, with the exclusion of faults 1 and 2 as they are external
to the microgrid. Fig. 6.7 illustrates the faults analyzed within this section during
microgrid operation.
In order to gain insight into the operation of the protective devices during
each analyzed fault, Table 6.15 provides an overview of the main and backup protec-
tive devices associated with each fault shown in Fig. 6.7. This table demonstrates
protective measures taken in response to each fault scenario.

Table 6.15: Main and backup protection for microgrid faults

Fault Number Main Protection Backup Protection

3 PCC Relay BESS Relay
4 Recloser 3 PCC Recloser
5 Recloser 3 PCC Recloser
6 Recloser 3 PCC Recloser
7 AUX Relay BESS Relay

The experimental analysis will showcase the improved response time of recloser
3 in operating for faults that were previously coordinated with the 80K fuse, specif-
ically faults 5 and 6. Moreover, the proposed reverse settings for the PCC recloser
is validated through the examination of faults 4, 5, and 6. Similarly, the proposed

139
140
Figure 6.7: RSCAD simulated faults during microgrid operation
 reverse settings for the PCC relay is tested using fault 3. The AUX relay’s proposed
forward settings is tested through fault 7. Additionally, the proposed reverse settings
for the BESS relay is validated through fault 7.
These chosen fault scenarios encompass and effectively summarize the success-
ful operation of all microgrid protection, as a form of main or backup protection. As
mentioned in the previous section, although additional faults within the generation
facility could have been examined, it was determined that faults 3 and 7 would suffice
due to the similarity of the protection settings implemented in the generation facility.
Table 6.16 presents a comparison between the experimental OTs for the faults
depicted in Fig. 6.7 and the corresponding OTs derived from the design. This
comparison encompasses the fastest and slowest OTs for each fault, as done in the
previous section. A more in-depth analysis of the OTs specific to recloser 3 is shown
in Table 6.17, as it utilizes the dynamic 51V element. Table 6.17 presents a detailed
breakdown of the OTs for each individual fault, allowing for a focused examination
of the OTs, which are a function of both voltage and current.
Table 6.16: Designed and experimental OTs during microgrid operationa

Fault Protective Protection Designed OT [s] Experimental OT [s]

Direction
Number Device Type Fastest Slowest Fastest Slowest
PCC Relay Main Reverse 1.21 1.75 1.21 1.75
3
BESS Relay Backup Reverse 1.41 1.95 1.41 1.95
Rec 3 Main Forward 0.04, 0.31 0.05, 0.81 0.04, 0.29 0.05, 0.83
4
PCC Rec Backup Reverse 0.91 1.41 0.88 1.39
Rec 3 Main Forward 0.04, 0.37 0.06, 0.84 0.04, 0.35 0.06, 0.85
5
PCC Rec Backup Reverse 0.96 1.42 0.94 1.41
Rec 3 Main Forward 0.04, 0.43 0.06, 0.88 0.04, 0.44 0.06, 0.86
6
PCC Rec Backup Reverse 1.02 1.43 0.99 1.41
AUX Relay Main Forward INST INST 0.02 0.02
7
BESS Relay Backup Reverse 1.41 1.95 1.41 1.95
a
#, # - Represents recloser OTs for fast, slow curves.

With no change in settings, all microgrid protection devices in HIL have proven

141
Table 6.17: Designed and experimental OTs of recloser 3 during microgrid operationa

Fault Number Fault Type Designed OT [s] Experimental OT [s]

SLG 0.05, 0.81 0.05, 0.83
LLG 0.04, 0.31 0.04, 0.29
4
LL 0.09, 0.37 0.1, 0.39
3PH 0.1, 0.38 0.11, 0.4
SLG 0.06, 0.84 0.06, 0.85
LLG 0.04, 0.37 0.04, 0.35
5
LL 0.07, 0.42 0.07, 0.41
3PH 0.07, 0.44 0.07, 0.43
SLG 0.06, 0.88 0.06, 0.86
LLG 0.04, 0.43 0.04, 0.44
6
LL 0.1, 0.49 0.11, 0.52
3PH 0.1, 0.52 0.11, 0.54
a
#, # - Represents recloser OTs for fast, slow curves.

to experimentally operate as predicted in the proposed design. For all main and
backup protective devices, the protection operation is less than 2 seconds in order to
meet the FRT limitation of the BESS inverter. From the test results, both the main
and backup protective devices have consistently shown protection operation within
2 seconds, meeting the FRT limitation of the BESS inverter. Additionally, although
demonstrating some short delays, the directional element has accurately detected
fault current direction, allowing for proper directional overcurrent operation.
In summary, both the previous section and this section validate the adequacy
of the proposed protection scheme presented in Chapter 5, which utilizes a single set
of settings for both grid and microgrid operation. Moving forward, the next section
will focus on demonstrating the security of the protection scheme, specifically testing
its resilience against extreme operating conditions to demonstrate that it does not
operate.

142
6.5 Protection Security

In Chapter 5, it was highlighted that voltage supervision plays a crucial role in

the majority of protective devices deployed in the microgrid. Recloser 3 utilizes the
51V protection element with a voltage drop threshold to adapt its MOC. Meanwhile,
several protective devices in the generation facility employ overcurrent protection
alongside undervoltage supervision at 0.8 pu. Furthermore, Chapter 5 demonstrated
that the voltage drop experienced during microgrid operation is more pronounced
compared to grid connection.
To assess the security of the proposed protection scheme, two tests are con-
ducted specifically during microgrid operation, as the generation facility faces greater
challenges in maintaining microgrid voltage stability. The first test involves the start-
ing of a 1000 kVA induction motor, representing the largest commercial microgrid
customer. As previously mentioned in the chapter’s first section, this commercial
customer has a transformer rating of 1000 kVA. Therefore, to account for the worst-
case scenario, a 1000 kVA induction motor with a minimal ramp rate is used. This
test aims to ensure that the microgrid protection, primarily recloser 3 due to its sensi-
tivity to voltage drop, does not operate from the voltage drop. The second test focuses
on capacitor energization. Within the microgrid, two 2.1 MVAR capacitor banks are
present. Initially disconnected and de-energized, this test verifies that the protection
system remains unaffected when both capacitor banks are connected simultaneously.
By conducting these tests, the aim is to validate the security of the protection scheme,
confirming that it does not operate under normal operating conditions.
Starting with the inrush current of induction motor startup, Fig. 6.8 illustrates
the abrupt connection of the induction motor during island operation, depicting the
resultant surge in current and the corresponding decline in measured voltage. These

143
measurements, shown in Fig. 6.8, were captured at recloser 3, which is situated
directly upstream the induction motor within the microgrid system.

Figure 6.8: Measurements at recloser 3 during induction motor startup

Fig. 6.8 displays the current and voltage behavior during the test scenario,
revealing that the highest current magnitude occurs in phase A and reaches approxi-
mately 147 A. Meanwhile, the line voltage experiences a significant drop to 22.72 kV
or 0.99 pu, remaining above the 51V reference voltages and the undervoltage super-
visor threshold of 0.8 pu. Therefore, the 51V protection element did not even pick up
on the current draw. Furthermore, although the phase MOC threshold of 91 A for
the protective devices in the generation facility were exceeded, as the voltage drop
did not surpass 0.8 pu, these protection devices did not operate.
From this security test, no protective devices operated during the induction
motor start, indicating the successful outcome of the first security test. Similarly,

144
Fig. 6.9 illustrates the sudden energization of both 2.1 MVAR capacitor banks at
peak system voltage. Since both capacitor banks are situated downstream of recloser
3, the measurements at recloser 3 are presented in Fig. 6.9. Moreover, to highlight
the capacitor banks’ energization at the voltage peak, the instantaneous voltage and
current waveforms are shown in the figure.

Figure 6.9: Measurements at recloser 3 during capacitor energization

Figure 6.9 depicts the energization of the capacitors at the voltage peak, specif-
ically on phase A, leading to a peak instantaneous current draw of 379 A. However,
since the protective devices operate based on current phasors, the maximum cur-
rent magnitude measured at recloser 3 is 136 A. Also, it is shown that there was
no sustained voltage drop observed during this event, with only an observed voltage
transient that lasted a few milliseconds. Given the defined voltage supervision, none
of the protective devices operated, effectively showcasing the security and reliability

145
of the protection scheme during capacitor energization.
Overall, this chapter has conclusively showcased the dependability, security,
and selectivity of the proposed protection scheme through experimental validation
utilizing HIL simulations. Notably, it has been established that all protective devices
consistently operate in accordance with the designated design, utilizing a single set
of settings for both grid and island fault scenarios. At the same time, the protection
scheme has been proven to not operate during induction motor starting and capacitor
energization events. This demonstrates the protection scheme’s ability to effectively
discriminate between normal system operation and potential fault conditions, ensur-
ing the reliable system operation.

146
Chapter 7

Conclusion

This project focused on analyzing a 22.86 kV Duke Energy system that fea-
tured a generation facility designed to facilitate the blackstarting of a microgrid sys-
tem. The main goal was to enhance microgrid resiliency and customer reliability by
transitioning from a blackstart microgrid to a seamless transition microgrid.
The analysis began by examining the existing BESS inverters and their ability
to provide seamless transition functionality in PSCAD. This involved operating the
inverters continuously in GFM while manipulating the voltage and frequency refer-
ences during grid connection. This revised configuration allowed for both planned and
unplanned islanding scenarios. Additionally, PSCAD controls were designed to syn-
chronize the microgrid with the main grid after islanding, manipulating the inverter’s
voltage and frequency reference to meet IEEE 1547-2018 synchronization check limits.
The implementation of seamless transition functionality rendered the existing
protection scheme reliant on communication obsolete due to inevitable delays in com-
munication and relay settings changes. As a result, an adaptive protection scheme
was proposed, utilizing the existing protection devices and offering single settings for
each device. This scheme incorporated elements such as inverse time-overcurrent,

147
bi-directional settings, voltage supervision, and voltage-restrained time overcurrent
protection.
To validate the proposed protection scheme, HIL simulations were conducted
on the RTDS platform. These simulations encompassed fault scenarios to demonstrate
the operation of the protection devices, as well as extreme loading conditions to
confirm the system security. The experimental results aligned with the designed
values, demonstrating the effective operation of all protective devices with single
settings and their immunity to seamless transition.
These achievements highlight the significant advancements made in both the
operational capabilities of the inverter and the system protection. Through the imple-
mentation of seamless transition and the removal of dependence on communication for
protection settings, this research effectively showcases a distribution system solution
that enhances the resilience and reliability of microgrid customers.

148
Bibliography

[1] H. Ritchie, M. Roser, and P. Rosado, “Renewable energy,” Oct 2022. [Online].
Available: https://ourworldindata.org/renewable-energy

[2] E. Fasching and S. Ray, “More than half of new u.s. electric-generating
capacity in 2023 will be solar,” Feb 2023. [Online]. Available: https:
//www.eia.gov/todayinenergy/detail.php?id=55419#

[3] IEA, “Renewables 2022 – analysis.” [Online]. Available: https://www.iea.org/

reports/renewables-2022

[4] Duke Energy, “Our history- honoring our past as we transform the future.”
[Online]. Available: https://www.duke-energy.com/our-company/about-us/
our-history

[5] N. Nissan and J. Sullivan, “Duke energy outlines progress on clean energy
transition,” Oct 2022. [Online]. Available: https://news.duke-energy.com/
releases/duke-energy-outlines-progress-on-clean-energy-transition

[6] D. T. Ton and M. A. Smith, “The u.s. department of energy’s microgrid initia-
tive,” The Electricity Journal, vol. 25, no. 8, pp. 84–94, 2012. [Online]. Available:
https://www.sciencedirect.com/science/article/pii/S1040619012002254

[7] R. Wheeless, “Duke energy places advanced microgrid into service in hot
springs, nc,” Feb 2023. [Online]. Available: https://news.duke-energy.com/
releases/duke-energy-places-advanced-microgrid-into-service-in-hot-springs-nc

[8] K. Prabakar, “Protection.” [Online]. Available: https://www.nrel.gov/grid/

protection.html

[9] “Ieee standard for interconnection and interoperability of distributed energy re-
sources with associated electric power systems interfaces,” IEEE Std 1547-2018
(Revision of IEEE Std 1547-2003), pp. 1–138, 2018.

[10] N. A. Mohamed and M. M. A. Salama, “A review on the proposed solutions to

microgrid protection problems,” in 2016 IEEE Canadian Conference on Electri-
cal and Computer Engineering (CCECE), 2016, pp. 1–5.

149
[11] A. R. Haron, A. Mohamed, and H. Shareef, “A review on protection schemes
and coordination techniques in microgrid system,” Journal of Applied Sciences,
vol. 12, pp. 101–112, 02 2012.

[12] I. Almutairy, “A review of coordination strategies and techniques for overcoming

challenges to microgrid protection,” in 2016 Saudi Arabia Smart Grid (SASG),
2016, pp. 1–4.

[13] M. e. a. Higginson, “Microgrid protection systems,” IEEE Power Energy Society-

Power System Relaying and Control Committee (PSRC), Aug 2019.

[14] T. Patel, S. Brahma, J. Hernandez-Alvidrez, and M. J. Reno, “Adaptive protec-

tion scheme for a real-world microgrid with 100% inverter-based resources,” in
2020 IEEE Kansas Power and Energy Conference (KPEC), 2020, pp. 1–6.

[15] A. Gupta, A. Varshney, O. G. Swathika, and S. Hemamalini, “Dual simplex algo-

rithm aided adaptive protection of microgrid,” in 2015 International Conference
on Computational Intelligence and Communication Networks (CICN), 2015, pp.
1505–1509.

[16] W. L. T. Peiris, W. H. Eranga, K. T. M. U. Hemapala, and W. D. Prasad,

“An adaptive protection scheme for small scale microgrids based on fault current
level,” in 2018 2nd International Conference On Electrical Engineering (EECon),
2018, pp. 64–70.

[17] K. A. Wheeler, S. O. Faried, and M. Elsamahy, “A microgrid protection scheme

using differential and adaptive overcurrent relays,” in 2017 IEEE Electrical
Power and Energy Conference (EPEC), 2017, pp. 1–6.

[18] “Ip latency statistics.” [Online]. Available: https://www.verizon.com/business/

terms/latency/

[19] O. Hegazi, E. Hammad, A. Farraj, and D. Kundur, “Iec-61850 goose traffic

modeling and generation,” in 2017 IEEE Global Conference on Signal and In-
formation Processing (GlobalSIP), 2017, pp. 1100–1104.

[20] SEL 351s, Nov 2022. [Online]. Available: https://selinc.com/products/351/

[21] H. Nikkhajoei and R. H. Lasseter, “Microgrid protection,” in 2007 IEEE Power

Engineering Society General Meeting, 2007, pp. 1–6.

[22] S. Ndjaba, G. T. Machnida, M. Nthontho, S. Chowdhury, S. P. Chowdhury,

and N. Mbuli, “Modeling and simulation of fault detection methods for power
electronic interfaced microgrids,” in 2012 47th International Universities Power
Engineering Conference (UPEC), 2012, pp. 1–6.

150
[23] ANSI C84.1 Voltage Rating for Electric Power Systems and Equipment. Amer
Natl Standards Inst. [Online]. Available: https://books.google.com/books?id=
YVnwOQAACAAJ

[24] D. Energy, Microgrid: System Impact Studies. Internal Document.

[25] ——, Microgrid: Protection Design Document. Internal Document.

[26] IEEE Std. C62.92.1-2019. IEEE, 2019.

[27] “On-load tap changer for transformer.” [Online]. Available: https://www.pscad.

com/knowledge-base/article/498

[28] “Photovoltaic source.” [Online]. Available: https://www.pscad.com/webhelp/

Master Library Models/Sources/Photovoltaic Source.htm

[29] T. Patel, P. Gadde, S. Brahma, J. Hernandez-Alvidrez, and M. J. Reno, “Real-

time microgrid test bed for protection and resiliency studies,” in 2020 52nd North
American Power Symposium (NAPS), 2021, pp. 1–6.

[30] D. C. Raj and D. N. Gaonkar, “Frequency and voltage droop control of par-
allel inverters in microgrid,” in 2016 2nd International Conference on Control,
Instrumentation, Energy Communication (CIEC), 2016, pp. 407–411.

[31] IEEE Standard for Measurement and Limits of Voltage Fluctuations and Asso-
ciated Light Flicker on AC Power Systems. IEEE Std 1453-2022 (Revision of
IEEE Std 1453-2015), 2022.

[32] J. L. Blackburn and T. J. Domin, Protective relaying: Principles and applica-

tions. CRC Press, an imprint of Taylor and Francis, 2014.

[33] J. J. Burke, Power Distribution Engineering. Marcel Dekker, 1994.

[34] IEEE Recommended Practice for Protection and Coordination of Industrial and
Commercial Power Systems (IEEE Buff Book). IEEE Std 242-2001 (Revision
of IEEE Std 242-1986) [IEEE Buff Book], 2001.

[35] SEL 700G, Nov 2022. [Online]. Available: https://selinc.com/products/700g/

[36] S. Brahma, “Protection of distribution system islands fed by inverter-interfaced

sources,” in 2019 IEEE Milan PowerTech, 2019, pp. 1–6.

[37] M. Roos, P. Nguyen, J. Morren, and J. Slootweg, “Modeling of distributed energy

resources for simulating fault-initiated islanding of microgrids,” in 2019 IEEE
Milan PowerTech, 2019, pp. 1–6.

151
[38] J. Valbuena G and A. Pavas, “Loss of coordination in a protection scheme due
to dg assessed by means of reliability analysis,” in 2019 IEEE Milan PowerTech,
2019, pp. 1–6.

[39] “Rscad articles and manuals.” [Online]. Available: https://knowledge.rtds.com/

hc/en-us/articles/360037537653-RSCAD-Modules

[40] SEL 421, Nov 2022. [Online]. Available: https://selinc.com/products/421/

[41] SEL 411, Nov 2022. [Online]. Available: https://selinc.com/products/411/

[42] M. Bryson, R. Turner, K. Garg, and P. Mahadanaarachchi, Implementing a 51V

Voltage-Restrained Inverse-Time Overcurrent Element in the SEL-487E Relay,
2011. [Online]. Available: https://selinc.com/api/download/97743/?lang=en

152
Appendices

153
Appendix A : MATLAB functions for producing
protection settings
function Ot = oT( I , Ip ,TD,TCC) % US TCC Curves
M = I / Ip ;
i f (M>30)
M = 30;
end

k = [0.0226 ,0.180 ,0.0963 ,0.0352 ,0.00262];

B = [0.0104 ,5.95 ,3.88 ,5.67 ,0.00342];
a = [0.02 ,2 ,2 ,2 ,0.02];
Ot = TD . ∗ ( k (TCC)+( B(TCC) . / (M. ˆ ( a (TCC) ) − 1 ) ) ) ;
end

% TripTimes . x l s x are t h e o p e r a t i n g t i m e s used f o r c o o r d i n a t i o n

% S p e c i f i c F a u l t P o i n t s . x l s x are t h e measured f a u l t s

function S e t p o i n t s= DT( I p m a t r i x , CTI , s h e e t F a u l t 1 ,

s h e e t F a u l t 2 , sheetCoord1 , sheetCoord2 )
% s e t p o i n t c a l c u l a t i o n f o r d e f i n i t e time
% I p m a t r i x i s o r g a n i z e d as Phase and Ground
% TCC matrix & TD matrix i s o r g a n i z e d as Phase Slow ,
% Ground Slow , Phase Fast , Ground Fast
c o o r d P o i n t 1= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord1 ) ;
f a u l t P o i n t 1= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 1 ) ;
c o o r d P o i n t 2= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord2 ) ;
f a u l t P o i n t 2= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 2 ) ;

f a u l t L o c= [ 1 3 5 7 9 ] ;
TD matrix =[0 0 ] ; % i n i t i a l v a l u e
TD check= 0 ;
for i= 1 : 1 : 2 % For Grid & I s l a n d c o o r d i n a t i o n
for k= 1 : 1 : 2 % For PS , GS
for j = 1 : 1 : 5
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= I p m a t r i x ( k ) ;
v= f a u l t L o c ( j ) ;
% Reference to navigate f a u l t current t a b l e
i f i ==1; coordMat= c o o r d P o i n t 1 ;
f a u l t M a t= f a u l t P o i n t 1 ;

154
 end % Grid c o o r d i n a t i o n p o i n t s
i f i ==2; coordMat= c o o r d P o i n t 2 ;
f a u l t M a t= f a u l t P o i n t 2 ;
end % I s l a n d c o o r d i n a t i o n p o i n t s

i f ( k==1 && j >2);

TD check= round ( coordMat { j , p}+CTI , 2 ) ; end
i f ( k==1 && j <3);
TD check= round (min( abs ( coordMat { j , p } ) ,
abs ( coordMat { j , p+2}))+CTI ) ; end
i f ( k==2 && j <3);
TD check= round (min( abs ( coordMat { j , p } ) ,
abs ( coordMat { j , p+2}))+CTI , 2 ) ; end

i f ( TD matrix ( k))< TD check ; TD matrix ( k)=TD check ; end

TD check =0; % Reset Value
end
end
end
end
S e t p o i n t s= [ I p m a t r i x ; TD matrix ] ;
end

function f i n a l T a b l e= oT DT( S e t p o i n t s , sheetName )

% o p e r a t i n g time f o r d e f i n i t e time
f a u l t P o i n t s= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” ,
’ S he e t ’ , sheetName ) ;
% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;
for k = 1 : 1 : 2 % Loop t h r o u g h Phase Slow , Ground Slow
for j = [ 1 3 5 7 9 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
TD= S e t p o i n t s ( 2 , k ) ;
i f o r ( k==1, k==3); timeTable P ( g , p−1)= TD; end
i f o r ( k==2, k==4); timeTable G ( t , p−1)= TD; end
end
i f o r ( k==1, k==3); g=g +1; end
i f o r ( k==2, k==4); t=t +1; end
end
end

155
f i n a l T a b l e= a r r a y 2 t a b l e ( [ timeTable P timeTable G ] ,
’ VariableNames ’ , { ’ PhaseTime MaxFault ’ , ’ PhaseTime MinFault ’ ,
’ GroundTime MaxFault ’ , ’ GroundTime MinFault ’ } ,
’RowNames ’ , { ’ LG S ’ , ’ LLG S ’ , ’ LL S ’ , ’ LLL S ’ , ’ LLLG S ’ } ) ;
end

function f i n a l T a b l e= oT woV woF ( S e t p o i n t s , sheetName )

% o p e r a t i n g time c a l c u l a t i o n f o r r e l a y s l o w c u r v e
f a u l t P o i n t s= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” ,
’ S he e t ’ , sheetName ) ;
% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;
for k = 1 : 1 : 2 % Loop t h r o u g h Phase Slow , Ground Slow
for j = [ 1 3 5 7 9 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= S e t p o i n t s ( 1 , k ) ;
TD= S e t p o i n t s ( 2 , k ) ;
TCC= S e t p o i n t s ( 3 , k ) ;

i f o r ( k==1, k==3); timeTable P ( g , p−1)=

oT( f a u l t P o i n t s { j , p }∗1000 , Ip , TD, TCC) ; end
i f o r ( k==2, k==4); timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +1,p }∗1000 , Ip , TD, TCC) ; end
end
i f o r ( k==1, k==3); g=g +1; end
i f o r ( k==2, k==4); t=t +1; end
end
end

f i n a l T a b l e= a r r a y 2 t a b l e ( [ timeTable P timeTable G ] ,
’ VariableNames ’ , { ’ PhaseTime MaxFault ’ , ’ PhaseTime MinFault ’ ,
’ GroundTime MaxFault ’ , ’ GroundTime MinFault ’ } , ’RowNames ’ ,
{ ’ LG S ’ , ’ LLG S ’ , ’ LL S ’ , ’ LLL S ’ , ’ LLLG S ’ } ) ;
end

function f i n a l T a b l e= oT woV wF ( S e t p o i n t s , sheetName )

% o p e r a t i n g time f o r r e c l o s e r s
f a u l t P o i n t s= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , sheetName ) ;
% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;

156
for k = 1 : 1 : 4
% Loop t h r o u g h Phase Slow , Ground Slow , Phase Fast , Ground Fast
for j = [ 1 3 5 7 9 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= S e t p o i n t s ( 1 , k ) ;
TD= S e t p o i n t s ( 2 , k ) ;
TCC= S e t p o i n t s ( 3 , k ) ;

i f o r ( k==1, k==3); timeTable P ( g , p−1)=

oT( f a u l t P o i n t s { j , p }∗1000 , Ip , TD, TCC) ; end
i f o r ( k==2, k==4); timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +1,p }∗1000 , Ip , TD, TCC) ; end
end
i f o r ( k==1, k==3); g=g +1; end
i f o r ( k==2, k==4); t=t +1; end
end
end

f i n a l T a b l e= a r r a y 2 t a b l e ( [ timeTable P timeTable G ] ,
’ VariableNames ’ , { ’ PhaseTime MaxFault ’ , ’ PhaseTime MinFault ’ ,
’ GroundTime MaxFault ’ , ’ GroundTime MinFault ’ } , ’RowNames ’ ,
{ ’ LG S ’ , ’ LLG S ’ , ’ LL S ’ , ’ LLL S ’ , ’ LLLG S ’ , ’ LG F ’ ,
’LLG F ’ , ’ LL F ’ , ’ LLL F ’ , ’LLLG F ’ } ) ;
end

function S e t p o i n t s= set woV woF ( I p m a t r i x , TCC matrix ,

CTI , s h e e t F a u l t 1 , s h e e t F a u l t 2 , sheetCoord1 , sheetCoord2 )
% s e t p o i n t c a l c u l a t i o n for r e l a y slow curve
% I p m a t r i x i s o r g a n i z e d as Phase and Ground
% TCC matrix & TD matrix i s o r g a n i z e d as Phase Slow ,
Ground Slow , Phase Fast , Ground Fast
c o o r d P o i n t 1= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord1 ) ;
f a u l t P o i n t 1= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 1 ) ;
c o o r d P o i n t 2= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord2 ) ;
f a u l t P o i n t 2= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 2 ) ;

f a u l t L o c= [ 1 3 5 7 9 ] ;
TD matrix =[0 0 ] ; % i n i t i a l v a l u e
TD check= 0 ;
for i= 1 : 1 : 2 % For Grid & I s l a n d c o o r d i n a t i o n
for k= 1 : 1 : 2 % For PS , GS
for j = 1 : 1 : 5

157
 for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= I p m a t r i x ( k ) ;
TCC= TCC matrix ( k ) ;
v= f a u l t L o c ( j ) ;
% Reference to navigate f a u l t current t a b l e
i f i ==1; coordMat= c o o r d P o i n t 1 ;
f a u l t M a t= f a u l t P o i n t 1 ; end % Grid c o o r d i n a t i o n p o i n t s
i f i ==2; coordMat= c o o r d P o i n t 2 ;
f a u l t M a t= f a u l t P o i n t 2 ; end % I s l a n d c o o r d i n a t i o n p o i n t s

syms TD Solve
i f ( k==1 && j >2); coordTime=
coordMat { j , p}+CTI ; TD check =
round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==1 && j <3); coordTime=
min( abs ( coordMat { j , p } ) , abs ( coordMat { j , p+2}))+CTI ;
TD check = round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==2 && j <3); coordTime=
min( abs ( coordMat { j , p } ) , abs ( coordMat { j , p+2}))+CTI ;
TD check = round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v+1,p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end

i f ( TD check ) < 0 . 5 ; TD check = 0 . 5 ; end

i f ( TD check ) >15; TD check =15;end

i f ( TD matrix ( k))< TD check ; TD matrix ( k)=TD check ; end

TD check =0; % Reset Value
end
end
end
end
S e t p o i n t s= [ I p m a t r i x ; TD matrix ; TCC matrix ] ;
end

function [ S e t p o i n t s , maxTimePS , minTimePS , maxTimeGS , minTimeGS ,

maxTimePF , minTimePF , maxTimeGF , minTimeGF]=
set woV wF ( I p m a t r i x , TCC matrix , CTI1 , CTI2 , s h e e t F a u l t 1 ,

158
s h e e t F a u l t 2 , sheetCoord1 , sheetCoord2 )
% setpoint calculation for recloser
% I p m a t r i x i s o r g a n i z e d as Phase and Ground
% TCC matrix & TD matrix i s o r g a n i z e d as
% Phase Slow , Ground Slow , Phase Fast , Ground Fast
c o o r d P o i n t 1= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord1 ) ;
f a u l t P o i n t 1= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 1 ) ;
c o o r d P o i n t 2= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord2 ) ;
f a u l t P o i n t 2= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t 2 ) ;

f a u l t L o c= [ 1 3 5 7 9 ] ;
TD matrix =[0 0 0 0 ] ; % i n i t i a l v a l u e
TD check= 0 ;
for i= 1 : 1 : 2 % For Grid & I s l a n d c o o r d i n a t i o n
for k= 1 : 1 : 4 % For PS , GS, PF, GF
for j = 1 : 1 : 5
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= I p m a t r i x ( k ) ;
TCC= TCC matrix ( k ) ;
v= f a u l t L o c ( j ) ;
% Reference to navigate f a u l t current t a b l e
i f i ==1; coordMat= c o o r d P o i n t 1 ;
f a u l t M a t= f a u l t P o i n t 1 ; CTI=CTI1 ; end
% Grid c o o r d i n a t i o n p o i n t s
i f i ==2; coordMat= c o o r d P o i n t 2 ;
f a u l t M a t= f a u l t P o i n t 2 ; CTI=CTI2 ; end
% Island coordination points

syms TD Solve
i f ( k==1 && j >2); coordTime=
coordMat { j , p}+CTI ; TD check =
round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==1 && j <3); coordTime= min( abs ( coordMat { j , p } ) ,
abs ( coordMat { j , p+2}))+CTI ; TD check =
round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==2 && j <3); coordTime=
coordMat { j , p+2}+CTI ; TD check =
round ( double ( s o l v e ( coordTime ==

159
 oT( f a u l t M a t {v+1,p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==3 && j >2); coordTime= coordMat { j +5,p}+CTI ;
TD check = round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==3 && j <3); coordTime= min( abs ( coordMat { j +5,p } ) ,
abs ( coordMat { j +5,p+2}))+CTI ; TD check =
round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v , p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end
i f ( k==4 && j <3); coordTime= coordMat { j +5,p+2}+CTI ;
TD check = round ( double ( s o l v e ( coordTime ==
oT( f a u l t M a t {v+1,p }∗1000 , Ip , TD Solve ,TCC) , TD Solve ) ) , 2 ) ;
end

i f ( TD check ) < 0 . 5 ; TD check = 0 . 5 ; end

i f ( TD check ) >15; TD check =15;end

i f ( TD matrix ( k))< TD check ; TD matrix ( k)=TD check ; end

TD check =0; % Reset Value
end
end
end
end
S e t p o i n t s= [ I p m a t r i x ; TD matrix ; TCC matrix ] ;

% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;

for k = 1 : 1 : 4
% Loop t h r o u g h Phase Slow , Ground Slow , Phase Fast , Ground Fast
for j = [ 1 3 5 7 9 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for i= 1 : 1 : 2 % Loop t h r o u g h Grid and I s l a n d Data
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
Ip= S e t p o i n t s ( 1 , k ) ;
TD= S e t p o i n t s ( 2 , k ) ;
TCC= S e t p o i n t s ( 3 , k ) ;
i f i ==1; f a u l t M a t= f a u l t P o i n t 1 ;
% Grid c o o r d i n a t i o n p o i n t s
i f o r ( k==1, k==3); timeTable P ( g , p−1)=

160
 oT( f a u l t M a t { j , p }∗1000 , Ip , TD, TCC) ; end
i f o r ( k==2, k==4); timeTable G ( t , p−1)=
oT( f a u l t M a t { j +1,p }∗1000 , Ip , TD, TCC) ; end
end

i f i ==2; f a u l t M a t= f a u l t P o i n t 2 ;
end % I s l a n d c o o r d i n a t i o n p o i n t s
i f o r ( k==1, k==3); timeTable P ( g , p−1+2)=
oT( f a u l t M a t { j , p }∗1000 , Ip , TD, TCC) ; end
i f o r ( k==2, k==4); timeTable G ( t , p−1+2)=
oT( f a u l t M a t { j +1,p }∗1000 , Ip , TD, TCC) ; end
end
end

i f o r ( k==1, k==3); g=g +1; end

i f o r ( k==2, k==4); t=t +1; end
end
end

maxTimePS= max( timeTable P ( 3 : 5 , : ) , [ ] , ” a l l ” ) ;

minTimePS= min( timeTable P ( 3 : 5 , : ) , [ ] , ” a l l ” ) ;
i f minTimePS < 0 ; maxTimePS=Inf ; end
maxTimeGS= max( timeTable G ( 1 , : ) , [ ] , ” a l l ” ) ;
minTimeGS= min( timeTable G ( 1 , : ) , [ ] , ” a l l ” ) ;
i f minTimeGS < 0 ; maxTimeGS=Inf ; end

maxTimePF= max( timeTable P ( 8 : 1 0 , : ) , [ ] , ” a l l ” ) ;

minTimePF= min( timeTable P ( 8 : 1 0 , : ) , [ ] , ” a l l ” ) ;
i f minTimePF < 0 ; maxTimePF=Inf ; end
maxTimeGF= max( timeTable G ( 6 , : ) , [ ] , ” a l l ” ) ;
minTimeGF= min( timeTable G ( 6 , : ) , [ ] , ” a l l ” ) ;
i f minTimeGF < 0 ; maxTimeGF=Inf ; end
end

function [ F i n a l S e t p o i n t , V r e f ]=
best wF ( ref Tim e , I p r e f , TD ref , TCC ref )
% Find b e s t c u r v e
[ row PS , column PS ]= find ( r e f T i m e == min( r e f T i m e ( : , 1 ) ) ) ;
[ newrow PS , newcolumn PS ]= find ( r e f T i m e == min( r e f T i m e ( row PS , 2 ) ) ) ;

[ row GS , column GS]= find ( r e f T i m e == min( r e f T i m e ( : , 3 ) ) ) ;

161
[ newrow GS , newcolumn GS]= find ( r e f T i m e == min( r e f T i m e ( row GS , 4 ) ) ) ;

[ row PF , column PF]= find ( r e f T i m e == min( r e f T i m e ( : , 5 ) ) ) ;

[ newrow PF , newcolumn PF]= find ( r e f T i m e == min( r e f T i m e ( row PF , 6 ) ) ) ;

[ row GF , column GF]= find ( r e f T i m e == min( r e f T i m e ( : , 7 ) ) ) ;

[ newrow GF , newcolumn GF]= find ( r e f T i m e == min( r e f T i m e ( row GF , 8 ) ) ) ;

F i n a l S e t p o i n t= [ I p r e f ( newrow PS ( 1 ) , 1 ) I p r e f ( newrow GS ( 1 ) , 2 )
I p r e f ( newrow PF ( 1 ) , 3 ) I p r e f ( newrow GF ( 1 ) , 4 ) ;
TD ref ( newrow PS ( 1 ) , 1 ) TD ref ( newrow GS ( 1 ) , 2 )
TD ref ( newrow PF ( 1 ) , 3 ) TD ref ( newrow GF ( 1 ) , 4 ) ;
TCC ref ( newrow PS ( 1 ) , 1 ) TCC ref ( newrow GS ( 1 ) , 2 )
TCC ref ( newrow PF ( 1 ) , 3 ) TCC ref ( newrow GF ( 1 ) , 4 ) ] ;

end

function [ F i n a l S e t p o i n t , V r e f ]= best V ( ref Tim e , I p r e f ,

TD ref , TCC ref , V comp ) % f i n d b e s t 51V s e t t i n g s
% Find b e s t c u r v e
[ row PS , column PS ]= find ( r e f T i m e == min( r e f T i m e ( : , 1 ) ) ) ;
[ newrow PS , newcolumn PS ]= find ( r e f T i m e == min( r e f T i m e ( row PS , 2 ) ) ) ;

[ row GS , column GS]= find ( r e f T i m e == min( r e f T i m e ( : , 3 ) ) ) ;

[ newrow GS , newcolumn GS]= find ( r e f T i m e == min( r e f T i m e ( row GS , 4 ) ) ) ;

[ row PF , column PF]= find ( r e f T i m e == min( r e f T i m e ( : , 5 ) ) ) ;

[ newrow PF , newcolumn PF]= find ( r e f T i m e == min( r e f T i m e ( row PF , 6 ) ) ) ;

[ row GF , column GF]= find ( r e f T i m e == min( r e f T i m e ( : , 7 ) ) ) ;

[ newrow GF , newcolumn GF]= find ( r e f T i m e == min( r e f T i m e ( row GF , 8 ) ) ) ;

F i n a l S e t p o i n t= [ I p r e f ( newrow PS ( 1 ) , 1 ) I p r e f ( newrow GS ( 1 ) , 2 )
I p r e f ( newrow PF ( 1 ) , 3 ) I p r e f ( newrow GF ( 1 ) , 4 ) ;
TD ref ( newrow PS ( 1 ) , 1 ) TD ref ( newrow GS ( 1 ) , 2 )
TD ref ( newrow PF ( 1 ) , 3 ) TD ref ( newrow GF ( 1 ) , 4 ) ;
TCC ref ( newrow PS ( 1 ) , 1 ) TCC ref ( newrow GS ( 1 ) , 2 )
TCC ref ( newrow PF ( 1 ) , 3 ) TCC ref ( newrow GF ( 1 ) , 4 ) ] ;

V r e f= [ V comp ( newrow PS ( 1 ) , 1 ) V comp ( newrow GS ( 1 ) , 1 ) ] ;

end

162
function [ S e t p o i n t s , maxTimePS , minTimePS , maxTimeGS ,
minTimeGS , maxTimePF , minTimePF , maxTimeGF ,
minTimeGF]= set wV ( V r e f , CTI , I p m a t r i x , TCC matrix ,
s h e e t F a u l t , sheetCoord ) % 51V s e t p o i n t c a l c u l a t i o n s
% I p m a t r i x i s o r g a n i z e d as Phase and Ground
% TCC matrix & TD matrix i s o r g a n i z e d as
% Phase Slow , Ground Slow , Phase Fast , Ground Fast
c o o r d P o i n t s= r e a d t a b l e ( ” TripTimes . x l s x ” , ’ Sheet ’ , sheetCoord ) ;
f a u l t P o i n t s= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t ) ;

% S o l v e f o r TD matrix f o r Phase Slow

v=1;
u=1;
for j = [ 1 4 7 10 1 3 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
V r a t i o= f a u l t P o i n t s { j +1,p}/ V r e f ( 1 ) ; % Compute v o l t a g e r a t i o
i f V r a t i o < 0 . 1 2 5 ; V r a t i o = 0 . 1 2 5 ; end
i f V r a t i o > 1 ; V r a t i o =1; end
Ip V= V r a t i o ∗ I p m a t r i x ( 1 ) ;

coordTime= c o o r d P o i n t s {v , p}+CTI ;
% Compute t h e TD n e c e s s a r y
syms TD Solve
TD Pcheck ( u ) = round ( double ( s o l v e ( coordTime
== oT( f a u l t P o i n t s { j , p }∗1000 , Ip V , TD Solve ,
TCC matrix ( 1 ) ) , TD Solve ) ) , 2 ) ;
i f ( TD Pcheck ( u ) ) > 15
TD Pcheck ( u )=15;
end
i f ( TD Pcheck ( u ) ) < 0 . 5
TD Pcheck ( u ) = 0 . 5 ;
end
u=u+1;
end
v=v+1;
end
TD matrix (1)= max( TD Pcheck , [ ] , ’ a l l ’ ) ;

% S o l v e f o r TD matrix f o r Ground Slow

v=1;
u=1;
for j = [ 1 4 ] % Loop t h r o u g h LG LLG

163
 for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
V r a t i o= f a u l t P o i n t s { j +1,p}/ V r e f ( 2 ) ;
% Compute v o l t a g e r a t i o
i f V r a t i o < 0 . 1 2 5 ; V r a t i o = 0 . 1 2 5 ; end
i f V r a t i o > 1 ; V r a t i o =1; end
Ip V= V r a t i o ∗ I p m a t r i x ( 2 ) ;

coordTime= c o o r d P o i n t s {v , p}+CTI ;
% Compute t h e TD n e c e s s a r y
syms TD Solve
TD Gcheck ( u ) = round ( double ( s o l v e ( coordTime ==
oT( f a u l t P o i n t s { j +2,p }∗1000 , Ip V , TD Solve ,
TCC matrix ( 2 ) ) , TD Solve ) ) , 2 ) ;
i f ( TD Gcheck ( u ) ) > 15
TD Gcheck ( u )=15;
end
i f ( TD Gcheck ( u ) ) < 0 . 5
TD Gcheck ( u ) = 0 . 5 ;
end
u=u+1;
end
v=v+1;
end
TD matrix (2)= max( TD Gcheck , [ ] , ’ a l l ’ ) ;
TD matrix ( 3 : 4 ) = [ 0 . 5 0 . 5 ] ; % SET FAST CURVE TD t o MINIMUM

% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;
for i= 1 : 1 : 2 % Loop t h r o u g h Slow Curve and Fast Curve
for k = 1 : 1 : 2 % Loop t h r o u g h Phase and Ground Curves
for j = [ 1 4 7 10 1 3 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
i f i==1 % Slow Curve i s 51V c o n t r o l l e d
V r a t i o= f a u l t P o i n t s { j +1,p}/ V r e f ( k ) ;
% Compute v o l t a g e r a t i o
i f V r a t i o < 0 . 1 2 5 V r a t i o = 0 . 1 2 5 ; end
i f V r a t i o > 1 V r a t i o =1; end
Ip V= V r a t i o ∗ I p m a t r i x ( k ) ;
TD= TD matrix ( k ) ;
TCC= TCC matrix ( k ) ;

164
 i f k==1; timeTable P ( g , p−1)=
oT( f a u l t P o i n t s { j , p }∗1000 , Ip V , TD, TCC) ;
end % Use phase c u r r e n t f o r phase c u r v e
i f k==2; timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +2,p }∗1000 , Ip V , TD, TCC) ;
end % Use ground c u r r e n t f o r ground c u r v e
end
i f i==2 % Fast Curve i s normal
Ip= I p m a t r i x ( k +2);
TD= TD matrix ( k +2);
TCC= TCC matrix ( k +2);

i f k==1; timeTable P ( g , p−1)=

oT( f a u l t P o i n t s { j , p }∗1000 , Ip , TD, TCC) ; end
i f k==2; timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +2,p }∗1000 , Ip , TD, TCC) ;
end
end
end
i f k==1; g=g +1; end
i f k==2; t=t +1; end
end
end
end

S e t p o i n t s= [ I p m a t r i x ; TD matrix ; TCC matrix ] ;

maxTimePS= max( timeTable P ( 3 : 5 , : ) , [ ] , ” a l l ” ) ;
minTimePS= min( timeTable P ( 3 : 5 , : ) , [ ] , ” a l l ” ) ;
i f minTimePS < 0 ; maxTimePS=Inf ; end
maxTimeGS= max( timeTable G ( 1 , : ) , [ ] , ” a l l ” ) ;
minTimeGS= min( timeTable G ( 1 , : ) , [ ] , ” a l l ” ) ;
i f minTimeGS < 0 ; maxTimeGS=Inf ; end

maxTimePF= max( timeTable P ( 8 : 1 0 , : ) , [ ] , ” a l l ” ) ;

minTimePF= min( timeTable P ( 8 : 1 0 , : ) , [ ] , ” a l l ” ) ;
i f minTimePF < 0 ; maxTimePF=Inf ; end
maxTimeGF= max( timeTable G ( 6 , : ) , [ ] , ” a l l ” ) ;
minTimeGF= min( timeTable G ( 6 , : ) , [ ] , ” a l l ” ) ;
i f minTimeGF < 0 ; maxTimeGF=Inf ; end

end

165
function f i n a l T a b l e V= oT wV( s h e e t F a u l t , S e t p o i n t s , V r e f )
% o p e r a t i n g t i m e s o f 51V
f a u l t P o i n t s= r e a d t a b l e ( ” S p e c i f i c F a u l t P o i n t s . x l s x ” , ’ Sheet ’ , s h e e t F a u l t ) ;

% S e t p o i n t s c o n t a i n Pickup , TD, TCC (PS , GS, PF, GF)

% S o l v e f o r O p e r a t i n g Times
g =1;
t =1;
for i= 1 : 1 : 2 % Loop t h r o u g h Slow Curve and Fast Curve
for j = [ 1 4 7 10 1 3 ] % Loop t h r o u g h LG LLG LL LLL LLLG
for k = 1 : 1 : 2 % Loop t h r o u g h Phase and Ground Curves
for p= 2 : 1 : 3 % Loop t h r o u g h Max and Min F a u l t s
i f i==1 % Slow Curve i s 51V c o n t r o l l e d
V r a t i o= f a u l t P o i n t s { j +1,p}/ V r e f ( k ) ;
% Compute v o l t a g e r a t i o
i f V r a t i o < 0 . 1 2 5 V r a t i o = 0 . 1 2 5 ; end
i f V r a t i o > 1 V r a t i o =1; end
Ip V= V r a t i o ∗ S e t p o i n t s ( 1 , k ) ;
TD= S e t p o i n t s ( 2 , k ) ;
TCC= S e t p o i n t s ( 3 , k ) ;

i f k==1; timeTable P ( g , p−1)=

oT( f a u l t P o i n t s { j , p }∗1000 , Ip V , TD, TCC) ;
end % Use phase c u r r e n t f o r phase c u r v e
i f k==2; timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +2,p }∗1000 , Ip V , TD, TCC) ;
end % Use ground c u r r e n t f o r ground c u r v e
end
i f i==2 % Fast Curve i s normal
Ip= S e t p o i n t s ( 1 , k +2);
TD= S e t p o i n t s ( 2 , k +2);
TCC= S e t p o i n t s ( 3 , k +2);

i f k==1; timeTable P ( g , p−1)=

oT( f a u l t P o i n t s { j , p }∗1000 , Ip , TD, TCC) ; end
i f k==2; timeTable G ( t , p−1)=
oT( f a u l t P o i n t s { j +2,p }∗1000 , Ip , TD, TCC) ;
end
end
end
i f k==1; g=g +1; end
i f k==2; t=t +1; end

166
 end
end
end

f i n a l T a b l e V= a r r a y 2 t a b l e ( [ timeTable P timeTable G ] ,
’ VariableNames ’ , { ’ PhaseTime MaxFault ’ , ’ PhaseTime MinFault ’ ,
’ GroundTime MaxFault ’ , ’ GroundTime MinFault ’ } , ’RowNames ’ ,
{ ’ LG S ’ , ’ LLG S ’ , ’ LL S ’ , ’ LLL S ’ ,
’ LLLG S ’ , ’ LG F ’ , ’LLG F ’ , ’ LL F ’ , ’ LLL F ’ , ’LLLG F ’ } ) ;
end

167
Appendix B : Custom Protection Logic for 51V
Voltage-Restrained Inverse-Time Over-
current Element in SEL Relay
# VOLTAGE RESTRAINT LOGIC
#
# CALCULATE VECTORS (VAB − VBC) , (VBC − VCA) AND (VCA − VAB)
# FIND MINIMUM MAGNITUDE OF THESE DIFFERENCE VOLTAGES

PMV01 := VAYFR − VBYFR

# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# REAL COMPONENT, PHASE AB
PMV02 := VBYFR − VCYFR
# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# REAL COMPONENT, PHASE BC
PMV03 := VCYFR − VAYFR
# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# REAL COMPONENT, PHASE CA
PMV04 := VAYFI − VBYFI
# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# IMAGINARY COMPONENT, PHASE AB
PMV05 := VBYFI − VCYFI
# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# IMAGINARY COMPONENT, PHASE BC
PMV06 := VCYFI − VAYFI
# INSTANTANEOUS FILTERED PHASE−TO−PHASE VOLTAGE,
# IMAGINARY COMPONENT, PHASE CA

# PMV07 := PMV01 − PMV02 # VABR − VBCR

# PMV08 := PMV02 − PMV03 # VBCR − VCAR
# PMV09 := PMV03 − PMV01 # VCAR − VABR
# PMV10 := PMV04 − PMV05 # VABI − VBCI
# PMV11 := PMV05 − PMV06 # VBCI − VCAI
# PMV12 := PMV06 − PMV04 # VCAI − VABI
#
PMV13 := SQRT( (PMV01 ∗ PMV01) + (PMV04 ∗ PMV04) ) # | VAB |
PMV14 := SQRT( (PMV02 ∗ PMV02) + (PMV05 ∗ PMV05) ) # | VBC |
PMV15 := SQRT( (PMV03 ∗ PMV03) + (PMV06 ∗ PMV06) ) # | VCA |
#
PSV09 := PMV13 <= PMV14

168
PSV10 := PMV13 <= PMV15
PSV11 := PMV14 <= PMV13
PSV12 := PMV14 <= PMV15
PSV13 := PMV15 <= PMV13
PSV14 := PMV15 <= PMV14
#
PMV16 := (PMV13 ∗ PSV09 ∗ PSV10 ) +
(PMV14 ∗ PSV11 ∗ PSV12 ) + (PMV15 ∗ PSV13 ∗ PSV14 )
# MIN PP VOLT
#
PMV17 := 2 . 5 0 0 0 0 0 # VNOMV ( 5KV IS THE REFERENCE)
PMV18 := PMV16 / PMV17 # MINVPU
#
PMV19 := 0 . 1 2 5 0 0 0 # NOMINAL VOLTAGE LOWER LIMIT ( 0 . 1 2 5 )
PMV20 := 1 . 0 0 0 0 0 0 # NOMINAL VOLTAGE UPPER LIMIT ( 1 . 0 )
PSV15 := PMV18 >= PMV19 # MINVPU >= 0 . 1 2 5
PSV16 := PMV18 <= PMV20 # MINVPU <= 1 . 0
PSV17 := NOT PSV15
PSV18 := NOT PSV16
#
#0.125 <= MINVPU <= 1 . 0
#
PMV21 := (PMV18 ∗ PSV15 ∗ PSV16 ) + (PMV19 ∗ PSV17 ) +
(PMV20 ∗ PSV18 ) # MIN VOLT, PU 0 . 1 2 5 <= VPU <= 1 . 0

# PMV22 i s s e t t o phase MOC

#
PMV25 := PMV22 ∗ PMV21
# UNIT 1 VOLTAGE RESTRAINED 51V PICKUP (BEFOR SETTING LIMIT CHECK)

#
# CHECK TO MAKE SURE PICKUP IS >= 0 . 2 5 A AND <= 16 A
PMV28 := 0 . 0 5 0 0 0 0 # MINIMUM 51 PICKUP
PMV29 := 1 6 . 0 0 0 0 0 0 # MAXIMUM 51 PICKUP
PSV19 := PMV25 >= PMV28 # UNIT 1 51PU >= 0 . 2 5
PSV20 := PMV25 <= PMV29 # UNIT 1 51PU <= 16
PSV21 := NOT PSV19
PSV22 := NOT PSV20
# 51P04 UNIT 1 VOLTAGE RESTRAINED 51V PICKUP (AFTER LIMIT CHECK)
PMV30 := PMV25 ∗ PSV19 ∗ PSV20 + PMV28 ∗ PSV21 + PMV29 ∗ PSV22

169
# GROUND

PMV31 := 8 . 7 5 0 0 0 0 # VNOMV ( 1 7 . 5KV IS THE REFERENCE)

PMV32 := PMV16 / PMV31 # MINVPU

PSV33 := PMV32 >= PMV19 # MINVPU >= 0 . 1 2 5

PSV34 := PMV32 <= PMV20 # MINVPU <= 1 . 0
PSV35 := NOT PSV33
PSV36 := NOT PSV34
#
#0.125 <= MINVPU <= 1 . 0
PMV37 := (PMV32 ∗ PSV33 ∗ PSV34 ) + (PMV19 ∗ PSV35 ) +
(PMV20 ∗ PSV36 ) # MIN VOLT, PU 0 . 1 2 5 <= VPU <= 1 . 0

# PMV38 i s s e t t o ground MOC

#
PMV39 := PMV38 ∗ PMV37
# UNIT 1 VOLTAGE RESTRAINED 51V PICKUP (BEFOR SETTING LIMIT CHECK)

#
# CHECK TO MAKE SURE PICKUP IS >= 0 . 2 5 A AND <= 16 A
PSV40 := PMV39 >= PMV28 # UNIT 1 51PU >= 0 . 2 5
PSV41 := PMV39 <= PMV29 # UNIT 1 51PU <= 16
PSV42 := NOT PSV40
PSV43 := NOT PSV41
# 51P04 UNIT 1 VOLTAGE RESTRAINED 51V PICKUP (AFTER LIMIT CHECK)
PMV44 := (PMV39 ∗ PSV40 ∗ PSV41 ) + (PMV28 ∗ PSV42 ) + (PMV29 ∗ PSV43 )

170
 ProQuest Number: 31014360

INFORMATION TO ALL USERS

The quality and completeness of this reproduction is dependent on the quality
and completeness of the copy made available to ProQuest.

Distributed by ProQuest LLC ( 2024 ).

Copyright of the Dissertation is held by the Author unless otherwise noted.

This work may be used in accordance with the terms of the Creative Commons license
or other rights statement, as indicated in the copyright statement or in the metadata
associated with this work. Unless otherwise specified in the copyright statement
or the metadata, all rights are reserved by the copyright holder.

This work is protected against unauthorized copying under Title 17,

United States Code and other applicable copyright laws.

Microform Edition where available © ProQuest LLC. No reproduction or digitization

of the Microform Edition is authorized without permission of ProQuest LLC.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346 USA