Received 30 December 2022; revised 1 May 2023; accepted 11 June 2023.

Date of publication 14 June 2023; date of current version 23 June 2023.

Digital Object Identifier 10.1109/OAJPE.2023.3285888

The Impact of Detailed Distribution System

Representation in Dynamic
Sub-Transmission-Distribution Co-Simulation
SUSHRUT THAKAR1 (Member, IEEE), VIJAY VITTAL 2 (Life Fellow, IEEE),
RAJA AYYANAR 2 (Fellow, IEEE), AND CYNTHIA ROJAS3 (Member, IEEE)
1 Electric Power Research Institute, Knoxville, TN 37932 USA
2 School of Electrical, Computer and Energy engineering, Arizona State University, Tempe, AZ 85281 USA
3 Arizona Public Service, Phoenix, AZ 85072 USA

CORRESPONDING AUTHOR: V. VITTAL ([email protected])

This work was supported in part by the U.S. Department of Energy, Solar Energy Technology Office, under Grant DE-EE0008773.

ABSTRACT There has been a significant growth in the distributed energy resources (DERs) connected
to the distribution networks in recent years, increasing the need for modeling the distribution networks in
detail in conjunction with the sub-transmission/transmission networks. This paper models a real distribution/
sub-transmission network using a three-phase/three-sequence co-simulation. One of the modeled distribution
feeders has a high penetration of DERs with significant reverse power flow and is modeled including the
secondary network. Custom user-defined models are used to represent the solar photovoltaic (PV) units on
the feeder including advanced controls and abnormal voltage responses from IEEE 1547-2018 standard. The
co-simulation framework developed supports power flow/steady state as well as dynamic analysis. Using
this developed framework, this paper studies the impact of balanced and unbalanced faults applied to the
distribution and sub-transmission networks. The impacts of the faults on the feeder with the high penetration
of DERs are studied in terms of the solar PV units tripping due to under/overvoltages and the resulting
change in the feeder-head flow. It is seen that the detailed modeling of the distribution network is needed
for accurately capturing the response from the distribution-connected DERs during fault events both on the
distribution as well as sub-transmission networks.
INDEX TERMS Co-simulation, distributed energy resources, distribution systems, integrated transmission-
distribution analysis, power system dynamic simulation.

I. INTRODUCTION Due to the increased penetration of distribution-connected

D ISTRIBUTION systems are one of the key elements

of the overall power system network. Historically,
several approximations have been invoked when studying
DERs and smart devices, there has been a critical
need for studies involving both distribution and sub-
transmission/transmission networks to examine the impact of
their behavior. Such approximations can be found in the the DERs on the sub-transmission system and transmission
distribution system modeling textbooks such as [1]. These system. This need is highlighted by the differences observed
approximations may not be appropriate with the increasing between integrated transmission-distribution system simula-
proliferation of distributed energy resources (DERs) con- tions in terms of aspects such as voltage regulator opera-
nected to the distribution systems in the recent years. Some tions [3], exchanged power at the transmission/distribution
smaller grids like Hawaii can have an instantaneous penetra- boundary in presence of voltage dependent loads [4], or time
tion of DERs of more than 71% of their daytime demand, series voltage excursions [5]. The authors of [6] show that
and supply 11% of annual energy [2]. Furthermore, even accurately capturing the response of DERs can impact bulk
larger grids across the United States of America and the world system events such as fault-induced delayed voltage recov-
expect the share of DERs to grow significantly in the near ery. Hence, for distribution networks, especially those with
future. a high penetration of DERs, it is important to consider
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
490 VOLUME 10, 2023
Thakar et al.: Impact of Detailed Distribution System Representation

transmission/sub-transmission and distribution networks in The authors of [15] model the combined T&D system
an integrated fashion. using MATLAB/Simulink and study the impact of aggregate
The authors of [6] adopt a coupled simulation approach DERs including protection settings on dynamic stability for
for studying the impact of DERs on the fault induced delayed load changes and balanced/unbalanced faults using small test
voltage recovery. The coupling approach entails recording the networks. The authors of [16] simulate the combined T&D
responses from one domain/simulation to update the other system using DIgSILENT PowerFactory to study the bulk
domain/simulation. The drawback of the coupled simulation system impact of DERs and dynamic motor loads and use
approach is that due to the coupling process, each domain the simulation to parameterize the equivalent models to be
simulation has to be run multiple times before reaching used instead of a detailed distribution feeder representation.
convergence. Another approach for integrating the distribu- They model the transmission system in positive sequence
tion and sub-transmission/transmission is to model the entire and use a small test system to represent each feeder. The
system using a single modeling framework [7] or using authors of [17] consider the impact of solar PV units on
a single commercial software, such as ePHASORSIM [8]. stability using T&D co-simulation for different disturbances
However, a large number of available tools for transmission and different PV penetration levels. They consider a lim-
and distribution systems do not support integrated modeling ited number of large (∼100kW) DERs - this may not be
of both systems in the requisite detail, hence, co-simulation the case in a more residential area with a high number of
is seen as a suitable approach which allows the use of exist- small rooftop solar PV units. In [18], the load tap opera-
ing established software with the domain-specific modeling tions due to DER generation intermittency and the voltage
capabilities, features and algorithms while analyzing the dif- recovery after a fault event in presence of DER ride-through
ferent systems in an integrated fashion. are studied using T&D co-simulation. The authors rep-
Using either the co-simulation approach or the integrated resent a real transmission network in positive sequence
modeling approach, several works have been published and the distribution network in three phase detail, but the
in recent years discussing dynamic simulations involving DER units are represented using an aggregate model. The
models of both transmission/sub-transmission and distri- authors of [19] develop a two-level Schur-complement based
bution grids. The authors of [9] model the entire net- domain decomposition method including parallel processing
work (transmission/sub-transmission and distribution) using to simulate the differential-algebraic equations corresponding
dynamic phasors and validate the simulation results with to the combined transmission-distribution system. A case
detailed MATLAB/Simulink models for a small synthetic study on the expanded Nordic system showing a long term
test case. On the other hand, the authors of [8] create a voltage instability after a fault event is presented using
combined transmission-distribution (T&D) model for power this algorithm. In [20], dynamic equivalents of distribu-
flow and dynamic studies using ePHASORSIM for a real tion networks are formed though Monte-Carlo simulations
system. In [10] a scalable multi-timescale dynamic T&D using multiple disturbances to account for the uncertainty
co-simulation formulation is presented. The authors of [11] in the inverter based generation parameters. The perfor-
describe dynamic T&D co-simulation approaches with series mance of dynamic equivalents is compared with the detailed
and parallel solution schemes, and consider the impact of distribution system representation both with and without
the integration time step on the T&D co-simulation for both considering disconnections from inverter based generation.
approaches. The focus in these papers is on developing the These inverter based resources are shown to provide trans-
integrated modeling framework rather than the examination mission voltage support during and after clearing the fault.
of system response for specific disturbances or for examining Here, the residential rooftop installations are considered to
the impact of DERs on system performance. trip immediately after the inception of a fault, and sustained
One application of T&D dynamic co-simulation is to response from only larger installations is considered. The
enhance the load model representation of the distribu- authors of [21] consider a dynamic T&D co-simulation using
tion feeders. The authors of [12] use T&D dynamic HELICS and study frequency regulation/response from the
co-simulation to estimate static load parameters represent- DERs including application to a large synthetic network. The
ing the distribution feeders. The co-simulation formulation considered disturbances do not lead to low voltages where
from [11] is utilized in [13] to determine the composite load DER over/undervoltage trip/no trip characteristics would be
model parameters for capturing response from a specific important.
feeder. In [14] a three-phase/three-sequence dynamic T&D This study aims to model a real sub-transmission and dis-
co-simulation algorithm is developed using multi-area tribution network in detail (including the secondary network
Thévenin equivalent approach. This paper considers balanced and detailed models for rooftop solar PV units existing at
as well as unbalanced faults. The authors of [7] model the each residential location) using a three-phase/three-sequence
entire transmission-distribution network using a three-phase co-simulation framework and study the behavior of the mod-
formulation for a utility network and consider an unbalanced eled network for different balanced and unbalanced faults,
fault. However, these references do not consider the impacts considering the solar PV trip/no trip characteristics and
of DERs connected to the distribution feeders. advanced controls. The study establishes the importance of

VOLUME 10, 2023 491

modeling the distribution network in detail to capture the negative and zero sequence circuits during the power flow,
response of the T&D network to faults accurately. The con- while during the dynamic simulation for all three sequence
tributions of this work can be summarized as follows: circuits, the distribution systems are represented using equiv-
• A detailed model of a real sub-transmission and dis- alent current injection. The three sequence networks are con-
tribution network is created using the proposed sub- sidered to be decoupled in the InterPSS network model. The
transmission/distribution co-simulation formulation in positive sequence dynamic simulation is performed using a
three-sequence and three-phase domains respectively. differential-algebraic equation formulation, whereas the neg-
The control features and the dynamic characteristics of ative and zero sequence effects are represented by sequence
solar PV units are represented in this framework using impedances. More details of the three-sequence dynamic
custom phasor domain models. simulation approach followed are available in [14] and [23].
• The impact of the advanced control and trip/no-trip con- A three-sequence model allows for applying balanced as
ditions from the IEEE 1547-2018 standard on the feeder well as unbalanced faults on the sub-transmission system to
response is studied for different faults. examine the impact on the integrated sub-transmission and
• For an unbalanced fault on the distribution feeder with distribution system.
a very high penetration above 400% when comparing The distribution system is modeled in three-phase unbal-
the load and solar PV generation, it is shown that the anced detail using OpenDSS [24]. OpenDSS is selected
post-fault clearance current may trigger the ground pro- as a well-supported open-source software which allows for
tection installed at the feeder-head. flexible and detailed modeling of distribution networks. The
• It is shown that different feeders connected to the same distribution networks modeled include the substation trans-
substation can behave differently for a sub-transmission former as well as several feeders. One of the feeders is
fault. modeled in detail including the primary as well as sec-
• It is shown that a detailed distribution feeder modeling ondary networks, described further in Section IV. The loads
including the secondary network is needed to accurately and solar PV inverter units for this feeder are modeled at
capture the response from the solar PV units. their residence/user location on the secondary network. The
solar PV inverters are modeled as a custom user model in
The rest of the paper is organized as follows: Section II
OpenDSS, described in Section III. The sub-transmission
describes the distribution-sub-transmission co-simulation
system is represented using an unbalanced voltage source for
framework used in this work, and Section III describes the
both power flow and dynamic simulation modes.
custom model created to represent the solar PV units installed
Helics [25] is selected for the time synchronization and
in the distribution network. The network data and model
the data coordination/information exchange between the
are discussed in Section IV. The behavior of the distribu-
two software packages. Helics is an open-source software
tion feeder with high penetration of solar PV generation
designed to facilitate co-simulation between different soft-
under balanced and unbalanced faults on the distribution
ware and supports connections to different software. For this
and sub-transmission network is studied in Section V. This
work, the Java and Python language bindings of Helics were
section also discusses the impact of different controls and
used while interfacing Helics software with InterPSS and
abnormal voltage responses on the feeder behavior during
OpenDSS, respectively.
these faults. Section VI establishes the importance of mod-
The overall schematic of the different models and data
eling the secondary network by comparing the simulations
exchange is shown in Fig. 1. In each iteration, the voltages for
with and without having a detailed secondary network model.
the boundary buses from InterPSS are sent to the distribution
Section VII concludes the paper.
systems and the currents injected at the boundary buses from
the distribution network are measured and sent back to the
II. CO-SIMULATION FRAMEWORK
sub-transmission system. Here, the sub-transmission model is
There are four aspects of the dynamic co-simulation
in three-sequence domain and the distribution systems are in
framework:
three-phase domain. The currents in three-phase domain from
• The transmission/sub-transmission network model the distribution networks are converted to three-sequence
• The distribution network model values before being sent to the sub-transmission systems,
• Representation of the sub-transmission and distribution while the sub-transmission boundary voltages are sent to the
systems in the other network distribution network in three-sequence domain and converted
• The coordination and information exchange during the to three-phase values before being updated in the network
co-simulation. model. Note, since the sub-transmission system is modeled
The sub-transmission system for this project is modeled using all three sequence networks, unbalances in voltages and
in InterPSS [22]. InterPSS, an open-source software, was currents are allowed/retained across the boundary.
selected because it supports modeling in three-sequence At the start of the simulation, the software packages, and
detail for power flow as well as dynamic simulations. The the Helics connections are initialized and the networks are
distribution systems are represented as equivalent loads loaded. During the power flow, the boundary values are
in positive sequence and equivalent current injections in iterated between the software packages until convergence

492 VOLUME 10, 2023

Thakar et al.: Impact of Detailed Distribution System Representation

FIGURE 1. Schematic of the co-simulation models and software.

FIGURE 3. Schematic of the controls for the solar PV inverter
model.

TABLE 1. Abnormal voltage trip settings from IEEE 1547-2018

standard [26].

the dc-side of the unit is represented as a constant dc voltage

source. The pulse-width modulation is represented using an
FIGURE 2. The serial solution process in the dynamic
average model. However, via a user defined model, the phase
co-simulation. locked loop and proportional resonant current control as well
as the active/reactive power control are modeled in detail.
Different active/reactive power control modes as defined in
is attained in the distribution and sub-transmission sys- the IEEE 1547-2018 Standard [26] are supported by the user-
tems as well as for the boundary bus values. Once the defined model. In addition, the abnormal voltage trip settings
steady-state power flow is successfully solved, the dynamic according to the inverter category as defined in [26] are also
models are initialized. The loads in the distribution system supported.
are represented using a constant impedance model during the For the active/reactive power modes, the setpoints are
dynamic simulation, while the solar PV units are represented assumed to be equal to the default settings given in [26]
using custom user models. In each time step, the dynamic for Category B inverters. The abnormal voltage trip/no trip
co-simulation is performed in a serial manner: the different settings used are given in Table 1. These settings represent
subsystems are solved sequentially, providing updated values the most conservative settings for each of the three categories
to the next subsystem, and moving to the next time step once from [26] (i.e. the mandatory ride through settings are consid-
all subsystems have been solved. A schematic of this serial ered and the permissive ride through is not considered). Note,
solution process during the dynamic simulation is shown in the clearing times given in the table are maximum values,
Fig. 2. for the results presented in Sections V and VI it is assumed
that the inverters trip within one cycle if the corresponding
III. SOLAR PHOTOVOLTAIC MODEL clearing time is 0.16 s (∼9.6 cycles).
In distribution feeders with a high penetration of solar pho- The user model represents the solar PV unit as a system
tovoltaic generation, the solar PV units in the system need to of differential-algebraic equations. Using these equations,
be represented in detail to capture the accurate behavior of the user-defined model requires the user to define various
the distribution system under various disturbances. A user- functions as a part of solving the power flow and performing
defined model for OpenDSS is created to represent the solar dynamic simulations. The dynamic equations corresponding
PV units including the controls. A schematic of the modeled to the filter as well as the current control and phase locked
controls is shown in Fig. 3. Here, without loss of generality loop are modeled as phasor dynamics equations using the

VOLUME 10, 2023 493

 FIGURE 5. Single line diagram of the five modeled feeders with
some key features.

TABLE 2. Key characteristics of the modeled feeders.

FIGURE 4. Schematic of the sub-transmission and distribution

networks modeled.

procedure from [27]. The created model is previously val-

idated in [28]. Different advanced controls and abnormal
voltage responses described in this section are implemented
building on the version from [28].

IV. NETWORK MODEL

The network modeled is based on a real substation and the the secondary network included. Utilizing the detailed model
connected feeders in Arizona. The selected substation from of Feeder 1, the dynamic co-simulations and results discuss
the utility feeds five feeders via two substation transformers. the impact of various fault events on it.
The substation transformers and the five feeders are modeled The initial feeder models are created by converting the
as two distribution systems, divided between two substation feeder models obtained in CYME format from the util-
transformers. The high voltage bus at this distribution substa- ity to OpenDSS circuit models using a tool developed for
tion is considered to be the boundary between the distribution this purpose [29]. The Feeder 1 model is further tuned
and sub-transmission systems. This substation is connected by combining other data from various sources such as the
to two other substations via 69 kV level lines. These two geographic information system (GIS) data and advanced
substations in turn connect to the bulk transmission network. metering infrastructure (AMI) meters - the procedure fol-
Hence, the bulk transmission system and the rest of the net- lowed for creating the feeder models is described in previous
work is represented at these two substations using equivalent publications [30], [31]. The sub-transmission model contains
Thévenin source model including the transfer impedance. two voltage sources - the active power and voltage references
A schematic diagram of the modeled network is given in are created based on the sub-transmission active power and
Fig. 4. The figure also shows the location of the faults applied voltage magnitude measurements and feeder-head measure-
to the sub-transmission network as discussed in Section V. ments of active and reactive powers from the feeders. The
The single line diagrams with some key elements of the operating point is selected to correspond to a high reverse
feeders are shown in Fig. 5, and some of the key character- active power flow case, at 1 PM on March 15, 2019. This
istics for these feeders are given in Table 2. From this table, operating point corresponds to light load and high solar PV
it is observed that Feeder 1 has a high level of penetration generation, resulting in a reverse power flow of over 2 MW
of the solar PV generation (and hence peak reverse power from Feeder 1. It is found that the feeder-head active and reac-
flow) compared to the other feeders - Feeder 1 also has one of tive powers for all five feeders and the active power flows in
the higher penetrations of solar PV generation for the feeders the two sub-transmission lines have errors less than 5% com-
controlled by the partner utility. Feeders 2-5 are modeled in pared to the field measurements. The voltages obtained for
detail, but the secondary networks are not modeled for these Feeder 1 also match closely with the voltage measurements
feeders. On the other hand, Feeder 1 is modeled in detail with from AMI meters connected to this feeder, as shown in Fig. 6.

494 VOLUME 10, 2023

Thakar et al.: Impact of Detailed Distribution System Representation

FIGURE 6. Histogram for the errors between the measured and

co-simulation voltages for AMI meters on Feeder 1.

TABLE 3. Magnitudes of the boundary values for the snapshot

FIGURE 7. Fault location for the fault on the distribution Feeder 1.
T&D co-simulation.

Feeder 1 is studied closely when applying various faults since

Feeder 1 is modeled in detail and has a high penetration of
solar PV generation.
For all the studies in this section, the selected base operat-
ing point is described in Section IV. In each case, the chosen
fault is applied for five cycles (fault on the sub-transmission
network) or fifteen cycles (fault on the distribution network)
and cleared after the fault duration. The fault locations on
the distribution and sub-transmission networks are shown
The boundary values in the different systems are provided in in Fig. 4 and Fig. 7. For the sub-transmission faults, the
Table 3 as an illustration of the unbalance across the T&D fault location is selected to be at the middle of one of
boundary during the co-simulation. It is observed that distri- the sub-transmission lines. For distribution system faults, the
bution system D2 comprised of Feeder 2 and Feeder 3 has a fault location is selected near the middle of Feeder 1 on a
very high ratio of negative to positive sequence currents – this three-phase bus on the main three-phase trunk of Feeder 1.
corresponds to a very low power exchanged for these feeders The solar PV inverters are assumed to have the trip/
at the feeder-head at the peak reverse power flow scenario no-trip response corresponding to the Category I inverters
selected (643 kW and 333 kVAr). from Table 1 and are assumed to trip for voltages below
The present penetration of solar PV generation in Feeder 1 0.45 p.u. and above 1.2 p.u. within one cycle. The tripped
(approx. 237 % when comparing the hourly load and solar PV inverters are assumed to not automatically reconnect after the
generation at the chosen operating point) does show impacts fault is cleared. Furthermore, unless otherwise mentioned, the
on the feeder fault responses, however, these impacts are results correspond to solar PV inverters operating at unity
expected to be even more prominent for higher penetrations power factor mode. However, the impact of different inverter
of solar PV generation. Hence, several cases of higher solar abnormal categories as well as advanced inverter controls
PV penetration of up to 400% penetration level are created (volt-VAr, with the reactive power-voltage curve taken from
for Feeder 1 by random allocation. The procedure for this the default values given in IEEE 1547-2018 standard) is also
allocation was developed by the National Renewable Energy discussed for these faults. The inverter current is limited to
Laboratory during their LA100 study [32]. twice the rated current when experiencing low voltage mag-
nitude. The inverter model scales down the original active and
V. RESPONSE OF THE T&D SYSTEM TO FAULTS reactive power references according to the selected control
With the accurate network model and the T&D dynamic mode to satisfy the current limit while retaining the ratio
co-simulation formulation described in this paper, this section of the active/reactive power references. For example, if the
discusses the response of the combined T&D system for bal- current limit is enforced for an inverter following unity power
anced (three-phase to ground - 3pG) and unbalanced (single factor control, then the active power reference will be scaled
line to ground - SLG) faults on the distribution and sub- down to satisfy the current magnitude limit, but for an inverter
transmission network. These responses highlight the impact following volt-VAr control, both the active and reactive power
of the faults on the combined system and the need for mod- references will be scaled down to satisfy the current magni-
eling the distribution network in detail. The response of tude limit.

VOLUME 10, 2023 495

FIGURE 8. The zero (0), positive (+) and negative (−) sequence FIGURE 10. The voltage profile for Feeder 4 and Feeder 5 with
boundary currents in the distribution systems (added) and single line to ground fault on Phase A of Feeder 1.
sub-transmission system.

FIGURE 11. Feeder-head active power for the faulted feeder,

Feeder 1, for single line to ground fault on Phase A of Feeder 1.
FIGURE 9. The voltage profile for Feeder 1 with single line to
ground fault on Phase A of Feeder 1.
As a result, within one cycle after the fault is applied,
A. UNBALANCED FAULT ON THE DISTRIBUTION a large number of solar PV units trip for Phases A and C.
NETWORK This results in a loss of approximately 1099 kW generation
When a SLG fault is applied on Feeder 1, the boundary for Phase A and 1314 kW generation for Phase C. It was
currents in the distribution system (added together for both previously discussed in [33] that even with the volt-VAr con-
systems D1 and D2) and the sub-transmission system are trol the voltage profiles (and the loss of generation) remain
plotted in Fig. 8 as a validation for a successful dynamic T&D very similar. However, it should be noted that instead of
co-simulation. Fig. 9 and Fig. 10 show the voltage profile for inverter abnormal response Category I from IEEE 1547-2018,
the feeders 1,4, and 5 connected to the substation transformer a more aggressive Category III implementation would result
for the faulted feeder. in the Phase A inverters remaining connected though the fault.
From Fig. 10, it is observed that for the other feeders Hence, there would be a reduction in the solar PV generation
connected to the substation transformer, only a small voltage lost and no generation would be lost for Phase A. However,
dip at the feeder-head is experienced for faulted Phase A, even the most aggressive Category III mandates the solar PV
corresponding to the voltage drop across the substation trans- units disconnect for voltages of more than 1.2 p.u., and the
former. Hence, the voltage profiles for the other feeders do loss of generation from Phase C would be the same even
not indicate a high voltage drop or swell similar to the faulted for Category III inverters. The disconnection of the solar
feeder. However, it is seen that for Feeder 1 the faulted PV generation for only the Phases A and C is reflected by
Phase A voltages are very low while Phase C experiences the feeder-head active power flow for Feeder 1, as shown in
a high voltage swell. For this feeder, the highly unbalanced Fig. 11, and the corresponding feeder-head current is plotted
fault current flowing through underground cables with mutual in Fig. 12.
coupling and the effective phase/ground impedances at the When the fault is cleared, this results in Phases A and
fault location result in such a voltage swell. This aspect of C with forward active power flow while the Phase B still
voltage swell is discussed in [33]. has a reverse active power flow because the solar PV units

496 VOLUME 10, 2023

Thakar et al.: Impact of Detailed Distribution System Representation

FIGURE 14. The voltage profile for Feeder 1 with a balanced

FIGURE 12. Feeder-head current magnitude for the faulted feeder, three-phase to ground fault on Feeder 1.
Feeder 1, for single line to ground fault on Phase A of Feeder 1.

FIGURE 15. The voltage profile for Feeder 4 and Feeder 5 with a
balanced three-phase to ground fault on Feeder 1.
FIGURE 13. The post-fault clearance ground current for Feeder 1
after applying a SLG fault for different solar PV penetration
cases.

connected to Phase B are not tripped – thus creating an unbal-

ance in the post-fault clearance condition. This unbalance
is noteworthy since this is a steady-state unbalance in the
post-fault clearance condition. When calculated, the ground
current measured at the feeder-head for this state would be
approximately 190.45 A. For this feeder, there are phase and
ground protection relays installed at the feeder-head. The
ground current in the post-fault clearance condition being
comparable to the ground relay pickup current of 300 A is
a significant concern, since it will persist for a much longer
period compared to the fault duration as long as no solar FIGURE 16. Feeder-head active power for the faulted feeder,
PV units connect back automatically. A similar SLG fault Feeder 1, for balanced three-phase to ground fault on Feeder 1.
was also applied to the modeled T&D network by increasing
the penetration of solar PV resources on Feeder 1. It is seen for Feeder 4 and Feeder 5 is shown in Fig. 15. It is seen that
from Fig. 13 that at 418% penetration level the post-fault all three phases for the faulted Feeder 1 experience very low
clearance ground current goes above the pickup current and voltages, with the voltage near the fault location being nearly
would result in the relay tripping the feeder in steady-state zero (Fig. 14) while the other feeders do not experience such
even when the fault is already successfully cleared, resulting low voltages (Fig. 15).
in a loss of connection for all the customers on this feeder. This results in most of the solar PV units on the faulted
feeder experiencing very low voltages at the terminal, and
B. BALANCED FAULT ON THE DISTRIBUTION NETWORK hence a large amount of solar PV generation trips within one
For the balanced fault on the distribution system, the voltage cycle for this fault. This can also be seen in Fig. 16, which
profile for Feeder 1 is shown in Fig. 14, and the voltage profile shows the feeder-head active power for the faulted feeder and

VOLUME 10, 2023 497

FIGURE 17. Feeder-head current magnitude for the faulted feeder,
Feeder 1, for balanced three-phase to ground fault on Feeder 1.
FIGURE 18. The voltages profiles for Feeder 1 under a single line
to ground fault on the sub-transmission system. Some of the
voltage rise/drops across the secondary network are
Fig. 17, which shows the feeder-head current for the faulted
highlighted by the orange circles.
feeder. In these figures, before the fault was applied there is
substantial reverse active power flow at the feeder-head for
all three phases. However, assuming that the tripped solar
PV units do not automatically reconnect after the fault is
cleared, there is a forward active power demand seen from
the feeder in the post-fault clearance condition due to the loss
of generation from the solar PV units. Overall, there is a loss
of approximately 1171 kW generation on Phase A, 1341 kW
on Phase B and 1314 kW on Phase C for Feeder 1.
While for the selected operating point, the resultant extra
net demand may not be a significant issue since the oper-
ating point corresponds to a low load scenario, a similar
increase in demand would be more significant for a near-peak
demand operating point. For this fault, even if volt-VAr con-
trol is implemented instead of unity power factor for the
solar PV units, the loss of generation is not significantly FIGURE 19. The voltages profiles for Feeder 4 and Feeder 5 under
impacted. Note, however, that using category III inverters a single line to ground fault on the sub-transmission system.
instead of category I inverters would result in the inverters
not disconnecting due to the low voltages caused during the are on the high voltage/delta side of the substation trans-
short-duration fault considered and no generation would be former. Due to the delta-wye transformer at the substation,
lost in such a case. two of the phases (A and B) on the feeder experience low
Another important aspect observed for the balanced (this voltages between 0.75 and 0.85 p.u. while the Phase C voltage
Subsection) as well as unbalanced (previous Subsection V-A) is between 1 and 1.1 p.u., as shown in Fig. 18 and Fig. 19.
faults on the distribution feeder is that the other distribution These voltages are not so low or high as to cause any gen-
feeders connected to the same substation transformer do not eration to trip. However, due to the low voltages on Phases A
experience similar abnormal voltage values as the faulted and B of the feeder, there is an increased reverse active power
feeder, indicating that modeling the feeders separately is flow at the feeder-head for these phases, as shown in Fig. 20.
important to accurately capture the fault response for the The corresponding feeder-head current is plotted in Fig. 21.
distribution faults. Further, modeling the faulted feeder in Fig. 18 also shows some locations with a significant voltage
sufficient detail is needed to appropriately apply the fault and rise/drop across the secondary network, indicating the need to
capture the voltages at the DER and load terminals. model the distribution feeders in detail to accurately capture
the voltage at the loads/DERs.
C. UNBALANCED FAULT ON THE SUB-TRANSMISSION
NETWORK D. BALANCED FAULT ON THE SUB-TRANSMISSION
When a SLG fault is applied on the sub-transmission system, NETWORK
assuming the fault is applied on Phase A, it is seen that at the When a balanced fault is applied at the middle of the sub-
boundary bus, the voltage for the faulted Phase A is approx- transmission network, the boundary bus experiences a low
imately 0.56 p.u. voltage, whereas the other two phases have voltage of approximately 0.41 p.u. This results in all three
voltages between 0.95 and 1.06 p.u. However, these voltages phases on all the feeders experiencing a low voltage. The

498 VOLUME 10, 2023

Thakar et al.: Impact of Detailed Distribution System Representation

FIGURE 20. Feeder-head active power for the faulted feeder,

Feeder 1 when a single line to ground fault is applied on the FIGURE 23. The voltage profile for Feeder 4 and Feeder 5 with
sub-transmission system. balanced three-phase to ground fault on the sub-transmission
system.

FIGURE 21. Feeder-head current magnitude for the faulted

feeder, Feeder 1 when a single line to ground fault is applied on FIGURE 24. Feeder-head current magnitude for the faulted
the sub-transmission system. feeder, Feeder 1 when a balanced three-phase to ground fault is
applied on the sub-transmission system.

Phases A, B and C respectively for Feeder 1. The feeder-head

current magnitude for Feeder 1 is plotted in Fig. 24.
After clearing the fault, the active power flow is still in
the reverse direction, the magnitude is somewhat less due to
the loss of some solar PV generation. However, it is seen
from Fig. 16 that while several solar PV units experience
a voltage above 0.45 p.u., the voltage at the feeder-head is
approximately 0.413 p.u. and there is a significant voltage rise
along the feeder which allows the solar PV units connected
to have voltages higher than 0.45 p.u. However, the other
feeders do not have such a high reverse active power flow
and corresponding voltage rise, hence, the solar PV units
connected to the other feeders would all trip for this case.
FIGURE 22. The voltage profile for Feeder 1 with balanced Hence, for this case it is seen that different feeders may
three-phase to ground fault on the sub-transmission system. behave differently – this behavior would not be captured by
aggregating all the feeders together as a single load while
voltage profile for the feeders Feeder 1, Feeder 4 and Feeder 5 studying this sub-transmission fault.
(fed by substation transformer T1) during this fault is shown For this fault, it is observed that the voltages for Feeder 1
in Fig. 22 and Fig. 23. As seen from the figure, while not are close to 0.45 p.u. threshold below which the category I
all solar PV units experience a voltage below 0.45 p.u. and inverters would trip. In this case, if all the solar PV units
trip, there is still a loss of 1065 kW, 751 kW and 742 kW on on Feeder 1 operate in volt-VAr mode, there is a larger

VOLUME 10, 2023 499

FIGURE 25. The voltage profile for Feeder 1 with balanced FIGURE 26. The feeder-head current for Feeder 1 with balanced
three-phase to ground fault in the sub-transmission system, three-phase to ground fault in the sub-transmission system,
with Feeder 1 solar PV units operating in volt-VAr control. with Feeder 1 solar PV units operating in volt-VAr control.

voltage rise owing to the extra reverse reactive power from

the inverters, as seen in Fig. 25. This results in a majority
of the solar PV units that would experience a voltage below
the threshold of 0.45 p.u. for unity power factor control not
experiencing such voltages below the threshold. Hence, the
solar PV generation lost in this case is reduced to 137 kW
FIGURE 27. The schematic showing considered ‘‘primary’’ and
for Phase A, 81 kW for Phase B and 40 kW for Phase C. ‘‘secondary’’ models.
This example shows that the advanced control modes such as
volt-VAr mode could have an impact in terms of generation
tripped due to certain sub-transmission events. For this fault,
it is also observed that the response to the fault without the
detailed distribution feeder model can be inaccurate – this is
discussed in Section VI.

VI. IMPACT OF SECONDARY NETWORK MODEL

It is seen throughout the voltage profiles shown in the pre-
vious section that there is a significant voltage rise across
the network due to the reverse active power flow, and some
of this voltage rise is across the secondary network. In this
section, the importance of modeling the secondary network
is assessed. For this purpose, two models of Feeder 1 are
considered. The ‘‘secondary’’ model is described and used
so far, where the secondary network is modeled in detail for FIGURE 28. Feeder voltages for Feeder 1 for the ‘‘primary’’ and
Feeder 1, and the loads and the solar PV units are located ‘‘secondary’’ models in steady state with inverters in volt-VAr
at the household/user locations. Another model, called ‘‘pri- mode.
mary’’ model in this section, models Feeder 1 similar to the
models of other four feeders, with the load and solar PV head of the feeder for the feeder modeled with the secondary
units represented in an aggregate manner at the secondary networks were similar between the primary and secondary
terminal of the distribution transformers, and there is no models, however, it is noted that there are certain overvoltages
representation of the secondary network. For the aggregated observed in the secondary model not observed in the primary
solar PV units, the voltage at the secondary terminal of the model because the voltage rise across the secondary network
distribution transformer is used for the volt-VAr curve, and is not captured, as seen in Fig. 28.
the reactive power limits are equal to the summation of the Due to this voltage difference, the reactive power from
corresponding limits of the connected inverters. A schematic the inverters absorbed/injected at the transformer can have
describing these models can be found in Fig. 27. a significant difference for volt-VAr control (up to 44%
It was found that at steady state, for both unity power difference for certain transformers) when comparing the
factor control and volt-VAr control, the voltages along the two models. This difference between the reactive pow-
feeders as well as the active and reactive powers seen at the ers injected/absorbed by the inverters connected to that

500 VOLUME 10, 2023

Thakar et al.: Impact of Detailed Distribution System Representation

FIGURE 29. The reactive power injected into the network by the
inverters at three distribution transformers for ‘‘primary’’ and
FIGURE 31. Feeder-head active power for ‘‘primary’’ and
‘‘secondary’’ models.
‘‘secondary’’ model during a balanced fault on the
sub-transmission model.

1065 kW, 751 kW and 742 kW respectively. Hence, it is

important to model the secondary network in detail and not
modeling the secondary network can lead to an inaccurate
response from the distribution model.

VII. CONCLUSION
This paper models a real sub-transmission-distribution net-
work in three-sequence/three-phase detail using a T&D
co-simulation formulation. The formulation is utilized to
study the impacts of different faults on the network. The
modeled network has a high penetration level of residential
rooftop solar PV generation. It is seen that under faults,
a large portion of these DERs may disconnect due to low or
FIGURE 30. Feeder 1 voltages for ‘‘primary’’ and ‘‘secondary’’ high voltage trips. Such trips may result in the active power
model during a balanced fault on the sub-transmission model.
flow at the feeder-head changing from reverse direction to
forward direction. If two of the three phases experience
transformer is shown as a bar chart in Fig. 29 for three trans- low/high voltages that cause solar PV generation trips, it is
formers connected to Feeder 1. Considering all the solar PV seen that at higher penetration levels the feeder may trip even
units, the combined reactive power injected from all the solar after the fault is cleared with the ground protection at the
PV units is different by 68 kVAr for the selected operating feeder-head with the present settings. Further, it is seen that
point (∼10% of the total reactive power from all the inverters advanced controls (such as volt-VAr) and advanced abnormal
or 14% of the feeder-head reactive power for this operating voltage responses as defined in the IEEE 1547-2018 standard
point). may alleviate some of the loss of solar PV generation. It is
For several balanced/unbalanced faults applied on the dis- observed that different feeders at the same substation may
tribution and sub-transmission systems, the response for both have very different responses for the same fault, indicating
the models is similar. However, for certain faults, the voltage the need to model different feeders at a substation sepa-
at the inverter terminal is above 0.45 p.u. after capturing the rately. By comparing two different network models, with
voltage rise across the secondary network whereas it is below and without the secondary network modeled in detail, it is
0.45 p.u. for the primary model – resulting in a difference in found that without modeling the secondary network, some
the number of solar PV units which are registered as tripped in overvoltages may not be captured in steady state, and that
both models. For one such case, where a balanced three-phase there may be an error in the estimated the reactive power from
to ground fault on the sub-transmission system is applied in the inverters operating in advance control modes such as volt-
the middle of one of the sub-transmission lines, the Feeder 1 VAr. Furthermore, by comparing the fault response for the
voltage profile and the feeder-head active power are shown in distribution feeder with and without modeling the secondary
Fig. 30 and Fig. 31, respectively. network, it is shown that the loss of solar PV generation may
Note that for this study the inverters are kept at unity be estimated incorrectly if the secondary network is ignored.
power factor mode. For this fault, in the primary model,
the solar PV generation tripped in Phases A, B and C are ACKNOWLEDGMENT
1219 kW, 1335 kW and 1175 kW respectively, while in The authors acknowledge the team at Arizona Public Service
the secondary model, the tripped generation are much less, for providing various inputs and for their work towards the

VOLUME 10, 2023 501

project, and the team at the National Renewable Energy Lab- [17] Q. Li and S. Abhyankar, ‘‘Evaluation of high solar penetration
oratory for their contributions in creating the high penetration impact on bulk system stability through a transmission-distribution
dynamics co-simulation,’’ in Proc. IEEE Power Energy Soc. Gen.
scenarios. Meeting (PESGM), Atlanta, GA, USA, Aug. 2019, pp. 1–5, doi:
10.1109/PESGM40551.2019.9119339.
REFERENCES [18] M. M. Rezvani, S. Mehraeen, J. R. Ramamurthy, and T. Field,
‘‘Dynamic interaction of distribution-connected DER_A with trans-
[1] W. H. Kersting, Distribution System Modeling and Analysis, 4th ed.
mission system via co-simulation analysis,’’ IEEE Trans. Ind. Appl.,
Boca Raton, FL, USA: CRC Press, 2018.
vol. 58, no. 2, pp. 1502–1511, Mar. 2022, doi: 10.1109/TIA.2021.
[2] R. Quint et al., ‘‘Transformation of the grid: The impact of distributed
3137087.
energy resources on bulk power systems,’’ IEEE Power Energy Mag., [19] P. Aristidou, S. Lebeau, and T. Van Cutsem, ‘‘Power system dynamic
vol. 17, no. 6, pp. 35–45, Nov. 2019, doi: 10.1109/MPE.2019.2933071. simulations using a parallel two-level Schur-complement decomposition,’’
[3] H. Jain, B. Palmintier, I. Krad, and D. Krishnamurthy, ‘‘Studying the IEEE Trans. Power Syst., vol. 31, no. 5, pp. 3984–3995, Sep. 2016, doi:
impact of distributed solar PV on power systems using integrated transmis- 10.1109/TPWRS.2015.2509023.
sion and distribution models,’’ in Proc. IEEE/PES Transmiss. Distribution [20] G. Chaspierre, G. Denis, P. Panciatici, and T. Van Cutsem, ‘‘A dynamic
Conf. Expo., Apr. 2018, pp. 1–5, doi: 10.1109/TDC.2018.8440457. equivalent of active distribution network: Derivation, update, validation
[4] M. Gupta and A. R. Abhyankar, ‘‘Coordinated load flow solution for and use cases,’’ IEEE Open Access J. Power Energy, vol. 8, pp. 497–509,
coupled transmission-distribution system incorporating load modeling,’’ 2021, doi: 10.1109/OAJPE.2021.3102499.
in Proc. 20th Nat. Power Syst. Conf. (NPSC), Dec. 2018, pp. 1–6, doi: [21] W. Wang, X. Fang, H. Cui, F. Li, Y. Liu, and T. J. Overbye, ‘‘Transmission-
10.1109/NPSC.2018.8771708. and-distribution dynamic co-simulation framework for distributed energy
[5] N. Panossian, T. Elgindy, B. Palmintier, and D. Wallison, ‘‘Syn- resource frequency response,’’ IEEE Trans. Smart Grid, vol. 13, no. 1,
thetic, realistic transmission and distribution co-simulation for volt- pp. 482–495, Jan. 2022, doi: 10.1109/TSG.2021.3118292.
age control benchmarking,’’ in Proc. IEEE Texas Power Energy [22] M. Zhou and S. Zhou, ‘‘Internet, open-source and power system simula-
Conf. (TPEC), College Station, TX, USA, Feb. 2021, pp. 1–5, doi: tion,’’ in Proc. IEEE Power Eng. Society General Meeting, Tampa, FL,
10.1109/TPEC51183.2021.9384935. USA, Jun. 2007, pp. 1–5, doi: 10.1109/PES.2007.385492.
[6] R. W. Kenyon, B. Mather, and B.-M. Hodge, ‘‘Coupled transmission [23] Q. Huang and V. Vittal, ‘‘Application of electromagnetic transient-
and distribution simulations to assess distributed generation response to transient stability hybrid simulation to FIDVR study,’’ IEEE
power system faults,’’ Electric Power Syst. Res., vol. 189, Dec. 2020, Trans. Power Syst., vol. 31, no. 4, pp. 2634–2646, Jul. 2016, doi:
Art. no. 106746, doi: 10.1016/j.epsr.2020.106746. 10.1109/TPWRS.2015.2479588.
[7] H. Jain, A. Parchure, R. P. Broadwater, M. Dilek, and J. Woyak, ‘‘Three- [24] EPRI. (2019). OpenDSS—EPRI Distribution System Simulator. [Online].
phase dynamic simulation of power systems using combined transmission Available: https://smartgrid.epri.com/SimulationTool.aspx
and distribution system models,’’ IEEE Trans. Power Syst., vol. 31, no. 6, [25] B. Palmintier, D. Krishnamurthy, P. Top, S. Smith, J. Daily, and
pp. 4517–4524, Nov. 2016, doi: 10.1109/TPWRS.2016.2535297. J. Fuller, ‘‘Design of the HELICS high-performance transmission-
[8] G. R. Bharati, S. Chakraborty, C. Duan, and T. Nishikawa, ‘‘An integrated distribution-communication-market co-simulation framework,’’ in Proc.
transmission-distribution modeling for phasor-domain dynamic analysis Workshop Model. Simul. Cyber-Physical Energy Syst. (MSCPES), Pitts-
in real-time,’’ in Proc. IEEE Power Energy Soc. Innov. Smart Grid burgh, PA, USA, Apr. 2017, pp. 1–6, doi: 10.1109/MSCPES.2017.
Technol. Conf. (ISGT), Washington, DC, USA, Feb. 2020, pp. 1–5, doi: 8064542.
10.1109/ISGT45199.2020.9087705. [26] IEEE Standard for Interconnection and Interoperability of Distributed
[9] J. Gao and Y. Xu, ‘‘Co-simulation of power transmission and distribution Energy Resources With Associated Electric Power Systems Interfaces,
networks using dynamic phasors,’’ in Proc. IEEE 9th Annu. Int. Conf. Standard 1547, 2018, doi: 10.1109/IEEESTD.2018.8332112.
CYBER Technol. Autom., Control, Intell. Syst. (CYBER), Suzhou, China, [27] Z. Yu, Y. Tang, T. Yao, and R. Ayyanar, ‘‘Dynamic simulation of CIG
Jul. 2019, pp. 1646–1651, doi: 10.1109/CYBER46603.2019.9066463. in large unbalanced distribution systems using an open source tool,’’ IET
[10] A. K. Bharati and V. Ajjarapu, ‘‘SMTD co-simulation framework with Gener., Transmiss. Distrib., vol. 13, no. 9, pp. 1638–1645, May 2019, doi:
HELICS for future-grid analysis and synthetic measurement-data gener- 10.1049/iet-gtd.2018.6089.
ation,’’ IEEE Trans. Ind. Appl., vol. 58, no. 1, pp. 131–141, Jan. 2022, doi: [28] S. Thakar, V. Vittal, and R. Ayyanar, ‘‘An integrated transmission-
10.1109/TIA.2021.3123925. distribution co-simulation for a distribution system with high renew-
[11] R. Venkatraman, S. K. Khaitan, and V. Ajjarapu, ‘‘Dynamic co-simulation able penetration,’’ in Proc. IEEE 48th Photovoltaic Specialists Conf.
methods for combined transmission-distribution system with integration (PVSC), Fort Lauderdale, FL, USA, Jun. 2021, pp. 0672–0679, doi:
time step impact on convergence,’’ IEEE Trans. Power Syst., vol. 34, no. 2, 10.1109/PVSC43889.2021.9518452.
pp. 1171–1181, Mar. 2019, doi: 10.1109/TPWRS.2018.2874807. [29] K. Montano and S. Thakar. (2020). DISMOTT: Distribution System Model
[12] S. Abhyankar, K. Balasubramaniam, and B. Cui, ‘‘Load model parameter Transformation Tool. School of Electrical, Computer and Energy Engi-
estimation by transmission-distribution co-simulation,’’ in Proc. Power neering, Arizona State University, Tempe, AZ, USA. [Online]. Available:
Syst. Comput. Conf. (PSCC), Dublin, Ireland, Jun. 2018, pp. 1–7, doi: https://github.com/thakars/DISMOTT
10.23919/PSCC.2018.8442939. [30] K. Montano-Martinez, S. Thakar, V. Vittal, R. Ayyanar, and
[13] R. Venkatraman, S. K. Khaitan, and V. Ajjarapu, ‘‘Application of com- C. Rojas, ‘‘Detailed primary and secondary distribution system
bined transmission-distribution system modeling to WECC composite load feeder modeling based on AMI data,’’ in Proc. 52nd North Amer.
model,’’ in Proc. IEEE Power Energy Soc. Gen. Meeting (PESGM), Port- Power Symp. (NAPS), Tempe, AZ, USA, Apr. 2021, pp. 1–6, doi:
land, OR, USA, Aug. 2018, pp. 1–5, doi: 10.1109/PESGM.2018.8585910. 10.1109/NAPS50074.2021.9449779.
[14] Q. Huang and V. Vittal, ‘‘Integrated transmission and distribution system [31] K. Montano-Martinez et al., ‘‘Detailed primary and secondary distribu-
power flow and dynamic simulation using mixed three-sequence/three- tion system model enhancement using AMI data,’’ IEEE Open Access
phase modeling,’’ IEEE Trans. Power Syst., vol. 32, no. 5, pp. 3704–3714, J. Power Energy, vol. 9, pp. 2–15, 2022, doi: 10.1109/OAJPE.2021.
Sep. 2017, doi: 10.1109/TPWRS.2016.2638910. 3125900.
[15] X. Lu, J. Wang, Z. Li, and M. Yue, ‘‘Dynamic stability assessment for [32] B. Palmintier et al., ‘‘Chapter 7: Distribution system analysis,’’ Nat. Renew.
integrated transmission-distribution system considering distributed energy Energy Lab., Golden, CO, USA, Tech. Rep., NREL/TP-6A20-79444-
resources,’’ in Proc. Int. Conf. Smart Energy Syst. Technol. (SEST), Porto, 7, 2021. [Online]. Available: https://www.nrel.gov/docs/fy21osti/79444-
Portugal, Sep. 2019, pp. 1–6, doi: 10.1109/SEST.2019.8849037. 7.pdf
[16] P. Dattaray, D. Ramasubramanian, P. Mitra, J. C. Boemer, M. Bello, [33] S. Thakar, V. Vittal, and R. Ayyanar, ‘‘Mitigation of fault related voltage
and A. Gaikwad, ‘‘Bulk system impact of DER and loads using T & D swell on distribution feeders using DER-based advanced inverter con-
cosimulation and aggregate models,’’ in Proc. IEEE PES Innov. Smart Grid trols,’’ in Proc. IEEE Power Energy Soc. Gen. Meeting (PESGM), Denver,
Technol. Europe, Espoo, Finland, Oct. 2021, pp. 1–5, doi: 10.1109/ISG- CO, USA, Jul. 2022, pp. 1–5.
TEurope52324.2021.9640042.

502 VOLUME 10, 2023