IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026 2377

Attribution of Responsibility for Short-Duration
Voltage Variations in Power Distribution Systems via

QGIS, OpenDSS, and Python Language
Arthur Gomes de Souza , Luiz Arthur Tarralo Passatuto , Wellington Maycon Santos Bernardes ,

Luiz Carlos Gomes Freitas , and Ênio Costa Resende

Abstract— Short-Duration Voltage Variations (SDVVs) are phe-
nomena that significantly impact power quality. Although they
typically last no longer than three minutes, such events can disrupt
load operations and cause substantial production losses. This study
presents an enhanced methodology for determining whether an
SDVV event originates upstream or downstream of the point of
common coupling between two agents interconnected through a
transformer. Building upon the work of Ferreira et al., whose origi-
nal approach was applied to a circuit using MATLAB/Simulink, this
research advances the methodology by applying it to both real and
benchmark distribution systems using open-source tools, namely
QGIS, OpenDSS, and PythonTM. The well-known IEEE 34-Bus
Test System has been used to verify the methodology’s generaliz-
ability. The method was also further validated through tests con-
ducted on two actual Brazilian distribution feeders in Uberlândia,
Minas Gerais: one supplying large industrial consumers such as a
rice mill and a carbonated beverages factory, and the other serving
a municipal wastewater treatment plant and a large photovoltaic
plant. By using real, detailed and georeferenced data, the approach
ensures an accurate representation of both the network topology
and the installed equipment. The results confirm that the proposed
methodology reliably identifies the origin of SDVV events. A key
contribution of this study is that the attribution of responsibility
remains robust regardless of variations in transformer winding

Received 4 May 2025; revised 20 August 2025; accepted 31 August 2025.
Date of publication 7 October 2025; date of current version 10 February
2026. Paper 2025-PSEC-0778.R1, presented at the 2024 International Work-
shop on Artificial Intelligence and Machine Learning for Energy Transfor-
mation (AIE), Vaasa, Finland, May 20–22, and approved for publication in
the IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS by the Power Systems
Engineering Committee of the IEEE Industry Applications Society [DOI:
10.1109/AIE61866.2024.10561325]. This work was supported in part by Minas
Gerais Research Funding Foundation (FAPEMIG) through Demanda Universal
under Grant APQ-02176-22 and Grant APQ-04929-22, in part by Conselho
Nacional de Desenvolvimento Científico e Tecnológico through CNPq under
Grant 406881/2022-7, and in part by Coordenação de Aperfeiçoamento de
Pessoal de Nível Superior - Brazil through CAPES - Finance Code 001, ROR:
00x0ma614. (Corresponding author: Wellington Maycon Santos Bernardes.)

Arthur Gomes de Souza, Luiz Arthur Tarralo Passatuto, Wellington May-
con Santos Bernardes, and Luiz Carlos Gomes Freitas are with the Depart-
ment of Electric Power Systems (DEPSEE), Faculty of Electrical Engineering
(FEELT), Federal University of Uberlândia (UFU), Uberlândia 38400-902,
Brazil (e-mail: arthurgs@ufu.br; tarralo@ufu.br; wmsbernardes@ufu.br; lcgfre-
itas@ufu.br).

Ênio Costa Resende is with the Department of Electric Power Systems
(DEPSEE), Faculty of Electrical Engineering (FEELT), Federal University of
Uberlândia (UFU), Uberlândia 38400-902, Brazil, and also with the School of
Technology and Innovations, Electrical Engineering, University of Vaasa, 65200
Vaasa, Finland (e-mail: eniocostaresende@ufu.br).

Color versions of one or more figures in this article are available at
https://doi.org/10.1109/TIA.2025.3618601.

Digital Object Identifier 10.1109/TIA.2025.3618601

configurations, fault resistance, circuit topology, load character-
istics, or the presence of distributed generation. These findings
demonstrate the accessibility, robustness and practical applicabil-
ity, offering a valuable tool for utilities and researchers aiming to
enhance power quality and accountability in distribution networks.

Index Terms—OpenDSS, phenomenon responsibility, power
quality, QGIS, short-duration voltage variation (SDVV).

NOMENCLATURE

ANEEL Brazilian Electricity Regulatory Agency.
ANN Artificial Neural Networks.
BDGD Geographic Database of the Distribution Company.
CLVR Critical Load Voltage Regulator.
DG Distributed Generation.
DVR Dynamic Voltage Restorer.
FFT Fast Fourier Transformer.
GIS Geographic Information System.
HV High Voltage.
LL Line-to-Line.
LLL Three-phase.
LLG Double-Line-to-Ground.
LG Single-Line-to-Ground.
MV Medium Voltage.
PRODIST Electric Power Distribution Procedures in the Na-

tional Electric System.
SDVV Short-Duration Voltage Variation.
STFT Short-Time Fourier Transform.
UF Unbalance Factor.

I. INTRODUCTION

A CONSUMER, whether an industry, a commercial com-
pany, or even a residential home, is exposed to challenges

associated with the quality of electrical energy that may signifi-
cantly impact their processes and operations [1], [2], [3]. One of
these problems is the SDVV. According to the Brazilian Electric-
ity Regulatory Agency (ANEEL) [4], SDVV refers to significant
deviations in the voltage rms value, that occur in time intervals of
less than three minutes. These phenomena are the main causes
of power quality loss, caused by short circuits, switching of
large loads that require high starting currents, or intermittent
disconnection of cable within the electrical system. SDVVs are
divided into voltage surges, voltage sags, and short interruptions,
depending on the duration of the occurrence [5], [6].

© 2025 The Authors. This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see
https://creativecommons.org/licenses/by/4.0/

https://orcid.org/0000-0003-2025-2988
https://orcid.org/0009-0002-5577-118X
https://orcid.org/0000-0001-7401-3478
https://orcid.org/0000-0002-1036-2801
https://orcid.org/0000-0002-5110-6791
mailto:arthurgs@ufu.br
mailto:tarralo@ufu.br
mailto:wmsbernardes@ufu.br
mailto:lcgfreitas@ufu.br
mailto:lcgfreitas@ufu.br
mailto:eniocostaresende@ufu.br
https://doi.org/10.1109/TIA.2025.3618601


2378 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

SDVV often lead to increased consumer complaints, mal-
function or shutdown of sensitive equipment across residential,
commercial, and industrial sectors. These disturbances may
also interrupt production processes, cause economic losses, and
compromise both facility and public safety [7], [8]. This article
proposes a generalized methodology for accurately assigning
responsibility in SDVV events, independent of the electrical
system’s specific configurations. The method is designed to be
broadly applicable, enhancing the precision and reliability of dis-
turbance source identification. The proposed approach enhances
electrical network reliability and facilitates the integration of
advanced analytical techniques. As a result, energy data can be
managed more efficiently and interpreted with greater clarity.

Module 8 of Electric Power Distribution Procedures in the
National Electric System (PRODIST) from Brazil evaluates the
SDVVs, and these contingencies are stratified based on their
duration and amplitude. After this stratification, the events are
divided into nine sensitivity regions. The aim is to establish
a correlation between the importance of each SDVV event
and the sensitivity levels of the different loads connected to
distribution systems, in Medium Voltage (MV) or High Voltage
(HV). Next, the Unbalance Factor (UF) indicator is calculated,
which characterizes the severity of the SDVV event [4].

It is important to highlight that the calculation of UF does not
lead to any form of penalization or allocation of responsibility
between the electricity companies and/or consumers. Instead,
it merely assesses the event. Given the frequent occurrence of
SDVV in power grids, a straightforward and practical method
for attributing responsibility is essential. The startup of large
motors may require mitigation strategies, such as the use of
soft-start devices or pre-scheduled operations, to prevent volt-
age sags and ensure compliance with power quality standards.
By defining clear responsibilities, this methodology facilitates
efficient problem-solving and ensures the reliability of the grid.

Ferreira et al. [9] introduced a methodology for allocating
responsibilities based on the analysis of voltage unbalances at
the primary and secondary sides of the service transformer.
Their original implementation was carried out using MAT-
LAB/Simulink, a proprietary platform widely used for simu-
lation in Electrical Engineering. Notably, their validation was
performed using a simplified and fully controlled laboratory
circuit, which limits its applicability to real-world conditions.

In contrast, the present study revisits and expands upon the
original methodology by applying it to both real-world dis-
tribution feeders and a standardized IEEE benchmark circuit,
using only open-source tools. This extension demonstrates the
method’s robustness and applicability in practical, large-scale
scenarios. Particularly, the study proposes adaptations and im-
provements that enhance the applicability and scalability of the
original method.

The revised methodology is initially tested on the test circuit
from the reference [9], a simplified system used to validate
foundational aspects. Following this, in sequence, it is applied
to the IEEE 34-Bus System, a widely recognized benchmark
for distribution network analysis. Subsequently, a more ro-
bust validation is conducted using two real feeders operated
by CEMIG Distribuição S/A in the city of Uberlândia, Minas

Gerais, Brazil. These feeders supply large industrial consumers
and critical infrastructure, which are especially vulnerable to
process disruptions caused by SDVV. The first feeder serves two
major industrial facilities: a carbonated beverage factory and a
rice processing plant. The second feeder supplies a municipal
wastewater treatment plant and a large photovoltaic plant, under-
scoring the growing integration of distributed generation (DG),
which has been rapidly expanding in Brazil and worldwide.

In addition, the network used for the simulations was obtained
from the Geographic Database of the Distribution Company
(BDGD), incorporating data on distribution lines (length, re-
sistance, and reactance per kilometer), transformers, and loads,
using QGIS [10] (previously known as Quantum Geographic
Information System (GIS)). It is a full-featured, free and open
source tool to perform different species of spatial analysis, al-
lowing to create, edit, visualize, analyze and publish mapping in-
formation on Windows, Mac OS, Linux and others. Previously, it
was challenging to find libraries that could automatically model
feeders into files compatible with OpenDSS, and this process
was done entirely manually. The analyzed feeders, ULAE714
and ULAU13, were fully modeled based on the BDGD using
BDGD2DSS1, a custom PythonTM library developed by two of
the authors [11]. This tool converts data from the BDGD into
files for simulation and analysis of electric distribution feeders in
the OpenDSS environment. Alternatively, the modeling can also
be performed using the library developed by Radatz et al. [12].
Once the complete network has been assembled, simulations
were carried out in OpenDSS, and a PythonTM routine was
developed for processing.

It is well-known that OpenDSS is an open-source software
widely used in both academic and commercial contexts to sim-
ulate electrical distribution networks [13], [14], [15]. Several
research studies have used OpenDSS for various purposes, such
as: (a) using chatbots to generate distribution network models,
specifically for educational applications [16]; (b) optimizing
with MATLAB the integration of Distributed Generation (DG)
systems into the distribution network [17]; and (c) evaluation
of the impact of the integration of charging stations for electric
vehicles into a real network [18]. The versatility and widespread
use of OpenDSS make it a valuable tool for energy system study.
It also has an interface to PythonTM via the py-dss-interface
package [19] enabling seamless communication between script
and software.

As is evident, a comprehensive explanation of the work’s
originality and specific contributions has been aforementioned
to the reader. Other papers summarising the related studies are
described in Table I. In general, these are studies that focus solely
on the detection and measurement of SDVV, while others com-
pare international reference standards. Only one study employs
a methodology for the attribution of a SDVV, which served as
reference for the present work.

Attention should be drawn to the fact that this article is an
updated version of Passatuto et al. [1], enhancing the explanation

1Available in https://pypi.org/project/bdgd2dss/. The user should use the
command ‘pip install bdgd2dss’ for package installation. Release on: July 31,
2025. The first commits found on Github are from September 2nd, 2024.

https://pypi.org/project/bdgd2dss/


DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2379

TABLE I
SUMMARY OF STUDIES ON SHORT-DURATION VOLTAGE VARIATION (SDVV)

of methodology, as well as incorporating additional feeders
with distinct characteristics and applying more computational
simulations. While the previous study focused on an only single
feeder, this work extends the analysis to the electrical circuit
designed by Ferreira et al. [9], to the IEEE 34-Bus Test System,
as well as two real feeders, ULAE714 and ULAU13, providing
a broader assessment of the proposed work across different
network configurations.

The paper is divided as follows: Primarily, Section II discusses
the methodology applied and explains the mathematical and
computational tools used, as well as elucidates the acquisition
and modeling of the electrical circuit with OpenDSS software
and PythonTM language. In Section III, the results are pre-
sented and critically analyzed. The validity of this investigation
is corroborated through the testing of several aforementioned
electrical networks. Finally, Section IV concludes the study and
provides recommendations for future research.

II. METHODOLOGY

The SDVV events are rms voltage changes, namely interrup-
tion, swell, and sags. These phenomena represent momentary

voltage variations in power systems: a sag is a temporary drop
below a specified threshold, typically caused by short circuits
or large load startups; a swell is a brief increase above the
nominal level, often due to load switching or grounding issues;
and an interruption occurs when the voltage falls below 5% of
the reference value, indicating a near or total loss of supply. The
duration is determined by the elapsed time interval in which the
signal exceeds a certain threshold value [4], [31]. Most SDVV
events are caused by short circuits [32].

According to Furse et al. [33], the most common causes of
short circuits in power grids are: faulty equipment, human error,
vehicle accidents, falling trees, strong winds, and storms. It is
observed that most of these events are temporary, either due to
the nature of the anomaly or due to protective measures, causing
a voltage sag in the network.

According to Bordalo et al. [34] and Martins [35], the fol-
lowing average values for the short-circuits occurrence were
determined by statistical analysis:
� Three-phase (LLL) short circuit: 3%;
� Double-Line-to-Ground (LLG) short circuit: 6%;
� Line-to-Line (LL) short circuit: 10% and



2380 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

� Single-Line-to-Ground (LG) short circuit: 81%.
In other words, 97% of short circuits are asymmetric. There-

fore, the analysis presented in this article to assign responsibility
for SDVV is based on the propagation of UF between the
primary and secondary sides of the transformer [9]. Unbalanced
three-phase voltages and currents can be decomposed into their
positive, negative, and zero sequence components, among other
analysis techniques. The relationship between the original com-
ponents A, B and C and the subsequent components is given by
(1).

⎡
⎢⎣
V̇a

V̇b

V̇c

⎤
⎥⎦×

⎡
⎢⎣

1 1 1

1 a2 a

1 a a2

⎤
⎥⎦ =

⎡
⎢⎣

˙Va0

˙Va1

˙Va2

⎤
⎥⎦ (1)

Once the sequence components of the voltages are obtained, the
UF can be calculated using (2) and (3).

UF%V2 =
Va2

Va1

× 100% (2)

UF%V0 =
Va0

Va1

× 100% (3)

The methodology is based on the evaluation of the propaga-
tion behavior of negative sequence voltage unbalance in the
transformer when a short circuit happens, and consequently a
SDVV. For the evaluation, several short circuits were applied on
the primary and secondary sides of the transformer, observing
whether the type of fault influences the results. Another variation
has been the type of connection of the primary and secondary
windings.

The zero sequence UF will be not analyzed because this com-
ponent may not be present depending on the type of connection,
for example, if one of the windings is connected in a Δ design,
making it impossible to use (3).

A further change, compared to the proposal of Ferreira et
al. [9] is that the relationship between the UF, here denoted as
URPS , of the primary and secondary sides is evaluated for the
attribution of responsibility for the SDVV (4). The values on
the primary and secondary sides are compared qualitatively in
the reference, but no explicit factor or quantitative relationship
between these values has been defined.

URPS =
UFprimary

UFsecundary
(4)

This work proposes that in case of a short circuit occurs on the
primary side of the substation, i.e. on the electric utility side,
regardless of its type or the number of phases involved in the
fault or the fault resistance value, the relationship defined by (4)
will be � 0.95 (threshold) and the responsibility for the SDVV
can be attributed to the electric utility. If, on the other hand,
the fault occurs on the secondary side, i.e. on, the consumer
side, the relationship defined by (4) is < 0.95, which means that
the responsibility for SDVV lies with consumer connected to
substation transformer.

Fig. 1. Circuit from Ferreira et al. [9].

Fig. 2. IEEE 34-Bus System [39].

A. Case Studies

The test circuit from the [9] is considered in first. This
benchmark system serves as a foundation for validating the
proposed methodology. Following this, the study analyzes a
benchmark power distribution system from IEEE, as well as two
real distribution feeders, both located in the city of Uberlândia,
Minas Gerais, Brazil. Each case presents distinct characteristics
and contributes to a comprehensive evaluation of the proposed
methodology. The analyzed systems are described as follows:
� Circuit from Ferreira et al. [9] – The test system used

in Ferreira et al. [9] consists of a simple circuit with three
transformers and two load groups, as shown in Fig. 1. The
methodology proposed in this article will initially be tested
on this electrical network, with Transformer 1 (T1) being
the primary focus of the study.

� IEEE 34-Bus System – It models a 34-node, 24.9 kV
medium-voltage radial distribution system with unbal-
anced loads and long, high-impedance lines, making it
ideal for voltage drop and power loss studies. The sys-
tem incorporates various load types, capacitor banks, and
voltage regulators, providing a realistic representation of
real-world distribution networks. Due to its complexity,
it is extensively used in studies on voltage control, fault
analysis, and the integration of distributed generation [36],
[37], [38] (Fig. 2).



DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2381

Fig. 3. Feeder ULAE714 region. A rice mill and a carbonated drink industry in highlight [40].

� Feeder ULAE714 – Originating from the Uberlândia
7 Substation, this 13.8 kV distribution circuit supplies
electricity to consumers in the Custódio Pereira neighbor-
hood. The feeder delivers power to both medium and low
voltage consumers, encompassing residential areas, com-
mercial establishments, and industrial facilities. Among
the industrial loads connected to this feeder are a rice
processing plant and a carbonated beverage factory, both
of which contribute significantly to the feeder’s demand
due to their continuous and energy-intensive operations.
Fig. 3 illustrates the layout and coverage of this feeder
within the distribution network, highlighting the substation
transformer and the loads of interest with their respective
bus numbers (note that this numbering originates from the
database in use).

� Feeder ULAU13 – Originating from the Uberlândia 1 Sub-
station, this 13.8 kV distribution circuit supplies electricity
to consumers in the Guarani and Tocantins neighborhoods.
It serves medium and low voltage loads, including resi-
dential, commercial, and institutional consumers. A key
feature of this feeder is its connection to a large-scale photo-
voltaic farm, which contributes to the local distributed gen-
eration capacity. Additionally, the feeder supplies power
to the municipal wastewater treatment plant, an essential
infrastructure facility. Fig. 4 shows the layout and area
of coverage of this feeder within the distribution system,
highlighting the substation transformer and the loads of
interest with their respective bus numbers (note that this
numbering originates from the database used).

The test circuit from the [9], along with these feeders and the
IEEE 34-Bus System, were selected to evaluate the proposed
methodology in different operational scenarios, considering
both industrial and infrastructure-critical loads, as well as a
standardized benchmark for comparison.

This approach evaluates the methodology across different
scenarios, considering load typology variations along the feeder.
Both circuits were modeled using BDGD data, a public database
with detailed electrical and geographical feeder information.
These data are processed in QGIS following Module 10 of
PRODIST, which standardizes Brazilian regulatory geographic
information, ensuring consistency in network modeling [41],
[42].

The BDGD contains comprehensive information about a vast
region served by the utility, potentially covering an entire state.
To simplify the studies, filters must be applied to delimit the
analysis area. Fig. 3 illustrates feeder ULAE714, and Fig. 4
shows feeder ULAU13, both adding a layer with a visualization
of the regions using the Google Satellite plug-in for QGIS [40].

In Figs. 3 and 4, the transmission lines responsible for
supplying electrical power to the high-voltage substation are
represented in pink. The green square indicates the substation,
where the substation transformer is located and which will be the
focus of this investigation for attributing responsibility, either
upstream or downstream, while the white line represents the
medium-voltage network. Regarding the points, red markers in-
dicate distribution transformers, whereas yellow markers denote
reclosers, which identify the connection points for consumers
supplied by the medium-voltage network. Additionally, in Fig. 3,



2382 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

Fig. 4. Feeder ULAU 13 region. Photovoltaic farm and a municipal wastewater treatment plant in highlight [40].

the green diamond highlights the rice milling industry, while
the black diamond represents the carbonated beverage industry.
In Fig. 4, the yellow symbol marks the photovoltaic farm,
and the blue diamond corresponds to the municipal wastewater
treatment plant.

For the feeder modeling, the BDGD2DSS library, which is
already publicly available [11], has been employed. A complete
description of its design and functionalities will be presented in
future works due to space constraints. This library processes the
BDGD layers and converts them into . dss files, enabling the
complete modeling of the system, including both medium- and
low-voltage sections, as well as loads and distributed generation.
The generated files can then be simulated within the OpenDSS
environment. By automating data conversion, the library facili-
tates the integration of real-world network information into com-
putational simulations, enhancing the accuracy of this analysis.
The modeled feeders are in a public repository from GitHub,
Inc© , available by the researchers in [43].

The BDGD layers used in the modeling can be grouped
according to their function:

1) Electrical Infrastructure:
� SUB – Substations of the system;
� CTMT – Medium voltage circuits, used to identify the

feeder;
� UNTRAT – Substation transformers;
� UNTRMT – Medium voltage transformers along the

circuit.
2) Electrical Network and Conductors:

� SEGCON – Detailed conductor data, including resis-
tance and reactance.

� SSDMT, SSDBT, and RAMLIG – Lengths of medium
and low voltage segments, and connection branches.

3) System Loads:
� UCMT and UCBT – Medium and low voltage consumer

units, respectively, from which the load powers are
extracted.

� PIP – Public lighting points.
� CRVCRG – Load curves associated with different types

of consumers.
4) Distributed Generation:

� UGBT and UGMT – Low and medium voltage gener-
ation units.

5) Control and Protection Equipment:
� UNSEMT and UNSEBT – Medium and low voltage

disconnect switches, essential to avoid improper inter-
ruptions in the modeling.

� EQRE and UNREMT – Voltage regulators of the net-
work.

� UNCRMT – Capacitor banks used to correct the voltage
level at certain feeder buses.

Thus, the developed model accurately represents the feeder’s
structure, ensuring that the simulation in OpenDSS properly
reflects the electrical characteristics of the studied network. The
modeling process followed the standard set by ANEEL [44],
[45], [46]. The short-circuit level at the substation input was set
to 100 MVA. The simulations were configured in daily mode,
considering a 24-hour period.

The one-line diagram presented in Fig. 5 provides a simplified
schematic representation of the feeders illustrated in Figs. 3
and 4. Transformer T1, which represents the one located at the



DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2383

Fig. 5. Single-line diagram. Transformer T1 connection type will be changed
interactively in the simulations (from Δ− Yn, to Yn −Δ, Δ−Δ and Yn −
Yn).

substation, has its data extracted from the UNTRAT layer and
will be the object of study in this analysis. Short-circuit tests
are conducted on both the primary and secondary windings of
this transformer, as illustrated in Fig. 5. During the simulation,
the connection type will be adjusted accordingly in order to
validate the proposed method and ensure accurate results for
each configuration.

Transformers T2, T3, T4, and Tn represent the medium-
voltage transformers from the UNTRMT layer, which may cor-
respond to both the distribution network transformers and those
of consumers supplied with medium voltage. Among these
consumers, those in feeders ULAE714 and ULAU13, highlighted
in Figs. 3 and 4, are of particular importance. Information about
these consumers is extracted from the UCMT and UCBT layers.

The white lines represent the medium-voltage circuit, with
data extracted from the SEGCON and SSDMT layers, while the
gray lines correspond to the low-voltage circuits from the SSDBT
layer. Additionally, the presence of distributed generation (DG)
is indicated at both medium and low voltage buses, reflecting
its distribution along the circuit, as recorded in the UGMT and
UGBT layers.

B. Computational Modeling and Analysis

To validate the proposed method, computational simula-
tions of the electrical circuits have been performed using the
OpenDSS software, also utilizing an interface script written in
the PythonTM language [47] with the py-dss-interface library.
This package was used to facilitate direct communication and
easy manipulation and to allow the export of data to more
suitable formats or the creation of diagrams as desired.

To better understand the simulation, a flowchart is shown in
Fig. 6. Basically, an instance is created between PythonTM and

Fig. 6. Flowchart of the proposed intelligent algorithm.

OpenDSS using the mentioned package, reading the . dss file
with the data of the circuit under study. Within the same script, a
list of the twenty four fault cases to be studied was created. The
four connection schemes of the transformer are then run through,
and the faults contained in the list of cases are simulated for each
of these schemes.

After compiling the circuit file with py-dss-interface, the
simulation is run in daily mode for one day, with short circuits
applied at noon, when solar incidence and DG contribution are
highest. The PythonTM script is used to extract the data of
interest. This includes the unbalances of the negative sequence
voltages in relation to the positive sequence voltages, on the
primary and secondary sides of the transformer (supplier and
load, respectively), each one in percent. By calculating the ratio
between these unbalances, the responsibility is assigned to one
of the parties involved in the system. Subsequently, the data is
organized in a data frame using Pandas library [48] for further
analysis. In the end, when all connection schemes have been run
through, the graphs are plotted using matplotlib [49] for better
visualization by the user.

The proposed flowchart can be used for different types of
analyses related to the assignment of responsibility when SDVV
occurs. The electrical system can be modelled directly with the.
dss file. Much of the extracted data have the default settings
from OpenDSS itself, and this may be checked directly via the
manual [50], following the same standard procedures in Brazil.

The proposed script simulates 24 different fault cases, con-
sidering up to four transformer connection schemes, based



2384 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

on the most commonly used configurations in the ANEEL
database [41] and the schemes available in OpenDSS. However,
not all connection schemes are tested in every case. When all
four transformer connections are considered, the total number of
simulations reaches 96 (4 connection schemes × 24 fault cases).

For each fault type, i.e., LG, LLG, or LL, simulations are
conducted with short-circuit resistances of 0 Ω, 5 Ω, 10 Ω, and
20 Ω. The inclusion of these resistance values ensures a more
comprehensive analysis, preventing extreme variations in fault
currents and aligning with numerical values used in [51].

It can be observed that the simulations in OpenDSS consider
a common reference for all elements of the circuit. Therefore,
the application of LG and LLG faults on the Δ transformer side,
will occur in one of the phases and the ground.

III. RESULTS AND DISCUSSIONS

The simulations provide the desired data, with the most
relevant being the ratio URPS between the UF on the pri-
mary (supplier) and secondary (consumer) sides of Transformer
T1 during the fault, as defined by (4). This ratio determines
which side is responsible for the SDVV, validating the proposed
methodology. Voltage plots are not presented since the studies
are conducted in DSS, a frequency-domain simulation engine
specifically designed for power distribution systems.

The tests were carried out on the circuit from [9], the IEEE
34-Bus System, and the feeders ULAE714 and ULAU13. In the
IEEE 34-Bus System, different transformer connection config-
urations were tested to assess their influence on the proposed
methodology. For the first feeder, all possible configurations
of the substation transformer (Transformer T1) were analyzed,
applying various types of short circuits to both the primary and
secondary sides of the transformer while considering different
short-circuit resistance values, as previously described. For the
second feeder, an additional analysis was conducted to evaluate
the impact of distributed generation (DG) by comparing sce-
narios with and without its presence. This comparison aimed to
determine whether the introduction of DG affects the method-
ology’s effectiveness.

A. Case 1 - Circuit From [9]

Table II presents the single-phase voltages at both the primary
and secondary sides of transformer T1 (Δ-Yn), corresponding
to buses B1 and B2 in the circuit. Five scenarios are considered:

1) Pre-fault single-phase voltages at buses B1 and B2;
2) Single-phase voltages during LG fault at bus B1 (primary

side of the transformer) with Rfault = 0 Ω;
3) Single-phase voltages during LG fault at bus B2 (sec-

ondary side of the transformer) with Rfault = 0 Ω;
4) Single-phase voltages during LL fault at bus B1 (primary

side) with Rfault = 20 Ω and;
5) Single-phase voltages during LL fault at bus B2 (sec-

ondary side) with Rfault = 20 Ω.
From these values, the unbalanced three-phase system is

decomposed into three balanced sequence components (posi-
tive, negative, and zero sequence) using (1). Subsequently, the
unbalance factors for both the primary and secondary sides

TABLE II
PRE-FAULT AND FAULT VOLTAGES AND ANGLES FOR THE CIRCUIT IN [9]

(TRANSFORMER: Δ-Yn)

are determined using (2) and (3). Finally, (4) is applied to
compute the ratio between the primary and secondary unbalance
factors, enabling the assessment of unbalance propagation. It
is important to note that Table II presents results only for the
short-circuit conditions corresponding to cases 1, 5, 20 and 24
of Table III, due to limitation of space.

Building upon this initial analysis for theΔ-Yn configuration,
the study extends to evaluate transformerT1 under four different
possible winding connections: Δ-Yn, Δ-Δ, Yn-Yn, and Yn-Δ.
Additionally, different short-circuit resistances are applied to
both the primary and secondary windings to assess their influ-
ence on system behavior. The results, presented in Table III,
illustrate how variations in transformer connections and fault
conditions impact the propagation of unbalances and voltage
deviations within the network.

Analyzing Table III, there is minimal variation due to trans-
former winding configurations or the applied short-circuit resis-
tance. This indicates that, in a smaller, simplified, and controlled
circuit, such changes do not significantly affect the results.

As a final point, the same Table III shows that primary-side
faults yield an unbalance ratio near 1, while secondary-side
faults cause a significant drop below 0.95. According to Fig. 6,
this threshold distinguishes responsibility attribution. These re-
sults confirm that the unbalance ratio reliably indicates whether
voltage disturbance originates on the primary or secondary side.

B. Case 2 - IEEE 34-Bus System

Table IV presents the pre-fault and during-fault voltages for
a LG short circuit, as well as for a LL fault applied to both the
primary and secondary sides of the transformer (SourceBus and
Bus 800), with a fault resistance of 20 Ω. As in the previous
case, these single-phase voltage values are used to calculate the
unbalance factors and analyze their propagation throughout the
system. It is important to note that Table IV includes only cases
1, 5, 20 and 24 from Table V.



DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2385

TABLE III
SIMULATION RESULTS FOR DIFFERENT CONNECTIONS T1 IN CIRCUIT FROM [9]

TABLE IV
PRE-FAULT AND FAULT VOLTAGES AND ANGLES FOR THE CASE IEEE 34-BUS

SYSTEM (TRANSFORMER: Δ-Yn)

In the IEEE 34-Bus System, the propagation of unbalance was
analyzed considering two different configurations for the supply
transformer, originally connected in Δ-Yn at bus 800 in Fig. 2.
Initially, tests were conducted with the original configuration,
and subsequently, the transformer connection was modified to

TABLE V
SIMULATION RESULTS FOR DIFFERENT FAULT TYPES CASE IEEE 34-BUS

SYSTEM

Yn-Yn to assess the impact of this change on the propagation
of unbalance throughout the system. The results obtained for
both configurations are presented in Table V, highlighting the



2386 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

TABLE VI
PRE-FAULT AND FAULT VOLTAGES AND ANGLES FOR THE ULAE714 FEEDER

(TRANSFORMER: Δ-Yn)

variations in voltages and unbalances recorded at different points
in the electrical network.

Analyzing the results in Table V, it is clear that for faults
on the primary side, the URPS values remain close to 1in both
transformer configurations. The greatest variation occurs in case
12. In this scenario, the URPS value changes from 0.9993 in the
Yn-Yn configuration to 1.0012 in the Δ-Yn configuration.

The absolute difference between these values is 0.0019, while
the relative difference, considering the Yn-Yn configuration as
a reference, is approximately 0.19%. This indicates that the
impact of the transformer connection on the fault response on
the primary side is minimal, with variations that are practically
negligible in most practical applications.

On the other hands, for faults on the secondary side, it is
observed that the unbalance ratio does not remain close to 1,
indicating that the unbalance does not propagate in the same way
as when the fault occurs on the primary side. As ratio < 0.95,
the responsibility is assigned to the consumer according to Fig. 6.

Therefore, these outcomes support the methodology’s
premise that, when the SDVV occurs on the primary side, the
unbalance ratio between the primary and secondary is close to
1. And it will be otherwise, in case the SDVV occurs on the
secondary side.

C. Case 3 - Feeder ULAE714

Table VI presents the pre-fault and during-fault voltages for
a LG short circuit, as well as for a line-to-line fault with a fault
resistance of 20Ω applied to both primary and secondary sides of
the transformer (SourceBus and Bus 126355009). Repeating the
previous process, these single-phase voltage values are used to

calculate the unbalance factors and analyze their propagation
throughout the system. It is important to note that Table VI
includes only cases 1, 5, 20 and 24 from Table VII.

Table VII presents the simulation results for the four trans-
former connection schemes:Δ-Yn,Yn-Δ,Δ-Δ, andYn-Yn. The
results include different short circuit resistance values and the
URPS ratios obtained for each configuration. It is observed that
the variation in resistance has little influence on the URPS ratio
when comparing cases with the same transformer and fault type.
The most significant difference occurs in the Δ-Yn connection
for an LG fault on the secondary side (cases 5 and 6), showing
an absolute variation of 0.1202. However, this variation does
not significantly impact the overall analysis, as the URPS ratio
remains far from unity in all cases.

When comparing different connection schemes, the results
remain consistent across configurations. Even with variations
in transformer connection, the URPS ratio stays close to unity
when the fault occurs on the primary side. The largest observed
difference appears in an LG fault on the secondary side, consid-
ering a fault resistance of 0 Ω, for the Δ-Yn and Δ-Δ connec-
tions, resulting in a value of 0.2442. However, this difference
does not affect the final conclusions, as the URPS ratio re-
mains far from unity, reinforcing the robustness of the proposed
methodology.

Thus, the study concludes that, in these cases, the URPS

ratio is minimally influenced by the type of transformer winding
connection, regardless of the presence or absence of a nonzero
fault resistance. This finding is crucial for the assignment of
responsibility in the occurrence of a SDVV, as the URPS ratio
provides an objective criterion: when it exceeds a defined thresh-
old, the fault is attributed to the primary side of the transformer
(supplier). Given that the lowest observed value was 0.9985 and
the highest was 1.0006, the established threshold of 0.95, as
defined in the script according to Fig. 6, proves to be not only
appropriate but could potentially be set even more restrictively,
reinforcing the robustness of the methodology.

It is important to highlight that when a fault occurs on the sec-
ondary side, where the consumer is connected, the measurement
of the UF at this location has minimal influence on the observed
value at the primary side. Conversely, when the fault occurs on
the primary side, the resulting unbalance directly impacts the UF
on both sides, causing their values to become nearly identical.

D. Case 4 - Feeder ULAU13

Table VIII presents the pre-fault and during-fault voltages for
a LG short circuit, as well as for a LL fault with a fault resis-
tance of 20 Ω, applied to both primary and secondary sides of
the transformer (SourceBus and Bus 126438617). Additionally,
Table VIII shows the voltage behavior at the bus (179454784) of
the photovoltaic farm. Following the same procedure as before,
these voltages serve as the basis for calculating unbalance indi-
cators and analyzing their propagation throughout the system.
Note that only cases 1, 5, 20, and 24 from Table IX are included
in Table VIII.

Observe that Table IX presents the values of the unbalance
ratio URPS for the feeder, considering scenarios with and



DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2387

TABLE VII
SIMULATION RESULTS FOR DIFFERENT CONNECTIONS FEDDER ULAE714

TABLE VIII
PRE-FAULT AND FAULT VOLTAGES AND ANGLES FOR THE ULAU13 FEEDER

(TRANSFORMER: Δ-Yn)

without DG, as well as variations in the transformer winding
connection (Δ-Yn and Yn-Yn) and short circuit resistance.

The analysis of Table IX confirms that both the variation in
short circuit resistance and the configuration of the transformer

windings do not significantly influence the unbalance ratios, as
observed in Case 1 and 2, leading to the same decisions.

Furthermore, when comparing the results with and without
the presence of DG, the largest absolute difference observed
was 0.0021 for the Δ-Yn connection in a line-to-line fault
on the primary side with RFault = 0 Ω, corresponding to a
relative variation of approximately 0.21%. This minimal impact
confirms that the presence of DG has a negligible effect on
the URPS ratios and, consequently, does not influence the
assignment of responsibility in a SDVV.

It is again confirmed, in another feeder, that when the SDVV
occurs on the transformer’s primary side, the ratio URPS is
close to a unit value, while this is not the case on the sec-
ondary side. Thus, if URPS is close to 1, the responsibility
lies with the supplier; otherwise, the responsibility is with the
consumer.

For better visualization, the script produces a scatter plot as
presented in Fig. 7, showing the ratio between primary and
secondary UF for each case considering the two connection
schemes. The data used corresponds to the values presented
in Table IX. To enhance interpretation, different markers are
used to distinguish between faults occurring on the primary
and secondary sides of the transformer. Visually, there is little
difference between the points of the two connection cases. There
is a noticeable distance between the cases where responsibility
lies with the supplier and those one where the consumer is
responsible. This observation further confirms the earlier find-
ing regarding the minimal impact of the type of transformer
connection. The threshold used for attribution is indicated in the
figure and proves to be satisfactory for distinguishing between
responsibilities.



2388 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

TABLE IX
SIMULATION RESULTS FOR DIFFERENT CONNECTIONS FEDDER ULAU13 WITH AND WITHOUT DGS

Fig. 7. Connection schemes with UF ratios between primary and secondary
sides of transformer in cases of SDVV.

This study was conducted using a computer with the following
specifications: an Intel(R) Core(TM) i5-8500 CPU @ 3.00 GHz
processor, 8 GB of RAM, a 500 GB SSD, running Windows

10 Pro (version 22H2, 64-bit), with PythonTM version 3.11.0
and installed OpenDSS version 9.6.1.3. The simulation time for
feeder ULAE714 was 793,33 seconds, while for feeder ULAU13,
it was 1275,97 seconds. Meanwhile, the IEEE 34-Bus System
simulation took less than 10 seconds.

IV. CONCLUSION

This paper demonstrates that the methodology proposed by
Ferreira et al. [9], initially applied to simplified circuits, can
also be successfully extended to complete and complex real
distribution feeders. By implementing the approach in both
actual utility networks and the IEEE 34-Bus Test System, this
study confirms its applicability in more realistic and opera-
tionally relevant scenarios. The use of OpenDSS, combined with
PythonTM, enhances the simulation and analysis process, while
offering an accessible and open-source alternative to commercial
tools. Additionally, the integration of real network data from
the BDGD strengthens the practical value and reliability of the
improved methodology.

This study contributes to the field of responsibility attribu-
tion for SDVV by analyzing various factors that may impact
unbalance propagation, including transformer winding config-
urations, fault resistance variations, different circuit types with
diverse load compositions, and the presence of DG. The re-
sults indicate that when the ratio between unbalance factors
is below 0.95, the responsibility is attributed to the consumer
side, highlighting this threshold as a key indicator. Conversely,
values equal to or greater than 0.95 suggest that the disturbance
originated on the supplier side. The analysis also revealed that
regardless of minor variations introduced by transformer con-
nections and fault resistances, the overall trend remained con-
sistent. Furthermore, despite the rapid expansion of distributed



DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2389

generation (DG) in modern distribution networks, the proposed
methodology remains robust and unaffected. The evaluation
of multiple circuit configurations with varying load conditions
further confirms its reliability and adaptability.

It is important to note that this work does not aim to pinpoint
the exact physical location of the event within the network or to
numerically allocate responsibility for it, which is a limitation
of the study and could be addressed in future research. Instead,
its purpose is to determine whether the disturbance occurred
upstream or downstream of the analysis point, providing a
methodology for responsibility attribution.

Thus, the proposed methodology contributes to regulatory
standards for events such as SDVV, relying on widely acces-
sible software and real data from electric utilities. The data
were obtained from a public repository. However, it should be
emphasized that computational simulations may not fully reflect
the current state of the grid, as the utility company frequently
makes unannounced updates to the database.

One of the main challenges faced during the research was
the accurate modeling of the circuit and the development of the
simulation code using PythonTM. The first difficulty has been
the limited availability of modeling tools, which was overcome
through the use of a custom library developed by two of the
authors and released to the public, available in [11]. This library
will be described in detail in future works. The second challenge
involved to create a script capable of simulating various types
of faults, which was greatly facilitated by the py-dss-interface
library [19]. Another significant difficulty was the potential
presence of incorrect or incomplete information in the publicly
available BDGD data, which required careful verification during
the modeling process.

The current results offer significant contributions to the en-
ergy transformation sector, providing a method that combines
computational analysis with relevant power system data to
evaluate recurring power quality events in the electrical grid.
Future work will aim to incorporate battery banks and addi-
tional consumers, refining the responsibility attribution to in-
dividual consumers or groups, rather than just identifying the
responsible side. Additionally, future investigations involving
Neural Networks [52], [53], [54], [55], [56], [57], [58] may
be explored as a means to enhance classification accuracy
and automate the attribution process, enabling more adaptive
and intelligent analysis of complex power quality patterns in
modern distribution networks. For this objective, computational
techniques such as deep models, multilayer perceptrons, or
radial basis functions may be used for training and validation.
Moreover, future studies may also include sensitivity analyses
and statistical evaluations to expand and validate the robust-
ness of the results under varying grid conditions and parameter
uncertainties.

ACKNOWLEDGMENT

The authors also thank the Programa de Pós-graduação em
Engenharia Elétrica (PPGEELT) from the Federal University of
Uberlândia (UFU).

REFERENCES

[1] L. A. T. Passatuto, A. G. d. Souza, W. M. S. Bernardes, L. C. G. Fre-
itas, and E. C. Resende, “Assignment of responsibility for short-duration
voltage variation via QGIS, OpenDSS and python,” in Proc. Int. Work-
shop Artif. Intell. Mach. Learn. Energy Transformation, 2024, pp. 1–6,
doi: 10.1109/AIE61866.2024.10561325.

[2] Dino, “Falhas no fornecimento de energia prejudicam indús-
trias,” 2023. Accessed: Sep. 6, 2025 (in Portuguese). [Online].
Available: https://valor.globo.com/patrocinado/dino/noticia/2023/09/28/
falhas-no-fornecimento-de-energia-prejudicam-industrias.ghtml

[3] D. O. Johnson and K. A. Hassan, “Issues of power quality in electrical
systems,” Int. J. Energy Power Eng., vol. 5, no. 4, pp. 148–154, 2016,
doi: 10.11648/j.ijepe.20160504.12.

[4] ANEEL, “Módulo 8 - Qualidade do fornecimento de energia
elétrica,” 2021. Accessed: Sep. 07, 2025 (in Portuguese). [On-
line]. Available: https://www.gov.br/aneel/pt-br/centrais-de-conteudos/
procedimentos-regulatorios/prodist

[5] D. O. Anggriawan, E. Wahjono, I. Sudiharto, A. A. Firdaus, D. N.
N. Putri, and A. Budikarso, “Identification of short duration voltage
variations based on short time fourier transform and artificial neu-
ral network,” in Proc. 2020 Int. Electron. Symp., 2020, pp. 43–47,
doi: 10.1109/IES50839.2020.9231815.

[6] F. R. Zaro, M. A. Abido, S. Ameenuddin, and I. M. Elamin, “Character-
ization of short-duration voltage events,” in Proc. 2012 IEEE Int. Conf.
Power Energy, 2012, pp. 650–654, doi: 10.1109/PECon.2012.6450294.

[7] G. R. d. Santos, N. C. d. Jesus, J. M. Machado, J. R. Cogo, W. S. d. Jesus,
and T. P. Franco, “Análise dos impactos associados às ocorrências de
variações de tensão de curta duração na operação de uma planta industrial
de papel e celulose,” in Proc. 2023 XV Braz. Conf. Qual. Power, 2023,
pp. 1–6, doi: 10.1109/CBQEE59548.2023.10503696.

[8] F. B. Bottura and M. Oleskovicz, “Optimal allocation of power qual-
ity monitors considering short-duration voltage variations and par-
allel harmonic resonance conditions in power distribution systems,”
Int. J. Elect. Power Energy Syst., vol. 144, 2023, Art. no. 108580,
doi: 10.1016/j.ijepes.2022.108580.

[9] A. R. Ferreira, B. M. Gianesini, J. C. Oliveira, and P. H. O. Rezende, “An
approach for assigning responsibilities of short-term voltage variations in
electrical power systems,” IEEE Access, vol. 11, pp. 7751–7758, 2023,
doi: 10.1109/ACCESS.2023.3237921.

[10] QGIS Team Development, “QGIS,” 2025. Accessed: Apr. 07, 2025.
[Online]. Available: https://qgis.org/project/overview/

[11] A. G. d. Souza and W. M. S. Bernardes, “bdgd2dss: Ferramenta para
modelagem de alimentadores da BDGD para uso com OpenDSS,” (in
Portuguese), 2025. Accessed: Sep. 07, 2025. [Online]. Available: https:
//pypi.org/project/bdgd2dss/

[12] P. Radatz and Contributors, “bdgd2opendss: A BDGD to OpenDSS con-
version tool,” 2024. Accessed: Apr. 4, 2025. [Online]. Available: https:
//github.com/pauloradatz/bdgd2opendss

[13] D. Montenegro, M. Hernandez, and G. A. Ramos, “Real time OpenDSS
framework for distribution systems simulation and analysis,” in Proc. 2012
6th IEEE/PES Transmiss. Distribution: Latin Amer. Conf. Expo., 2012,
pp. 1–5, doi: 10.1109/TDC-LA.2012.6319069.

[14] R. B. Roy, S. Alahakoon, P. J. V. Rensburg, and S. J. Arachchillage,
“Impact analysis of uncoordinated electric ferry charging on distribution
network,” in Proc. Adv. Elect. Eng., Electron. Energy, 2024, vol. 10,
doi: 10.1016/j.prime.2024.100783, Art. no. 100783.

[15] R. F. Espinoza and P. Godoy, “Opendss-based real-time RMS simulator:
Design and applications,” in Proc. 2024 Open Source Modelling Simul. En-
ergy Syst., 2024, pp. 1–7, doi: 10.1109/OSMSES62085.2024.10668999.

[16] R. S. Bonadia, F. C. L. Trindade, W. Freitas, and B. Venkatesh, “On the
potential of ChatGPT to generate distribution systems for load flow studies
using OpenDSS,” IEEE Trans. Power Syst., vol. 38, no. 6, pp. 5965–5968,
Nov. 2023, doi: 10.1109/TPWRS.2023.3315543.

[17] T. Pham, T. Nguyen, and L. Kien, “Optimal placement of photovoltaic
distributed generation units in radial unbalanced distribution systems
using MATLAB and OpenDSS-based co-simulation and a proposed meta-
heuristic algorithm,” Int. Trans. Elect. Energy Syst., vol. 2022, 2022,
Art. no. 1446479, doi: 10.1155/2022/1446479.

[18] M. Toledo-Orozco, E. Bravo-Padilla, C. Álvarez, D. B. Morales-Jadan, and
L. Gonzalez-Morales, “Methodological evaluation to integrate charging
stations for electric vehicles in a tram system using OpenDSS: A case
study in Ecuador,” Sustainability, vol. 15, no. 8, 2023, Art. no. 6382,
doi: 10.3390/su15086382.

https://dx.doi.org/10.1109/AIE61866.2024.10561325
https://valor.globo.com/patrocinado/dino/noticia/2023/09/28/falhas-no-fornecimento-de-energia-prejudicam-industrias.ghtml
https://valor.globo.com/patrocinado/dino/noticia/2023/09/28/falhas-no-fornecimento-de-energia-prejudicam-industrias.ghtml
https://dx.doi.org/10.11648/j.ijepe.20160504.12
https://www.gov.br/aneel/pt-br/centrais-de-conteudos/procedimentos-regulatorios/prodist
https://www.gov.br/aneel/pt-br/centrais-de-conteudos/procedimentos-regulatorios/prodist
https://dx.doi.org/10.1109/IES50839.2020.9231815
https://dx.doi.org/10.1109/PECon.2012.6450294
https://dx.doi.org/10.1109/CBQEE59548.2023.10503696
https://dx.doi.org/10.1016/j.ijepes.2022.108580
https://dx.doi.org/10.1109/ACCESS.2023.3237921
https://qgis.org/project/overview/
https://pypi.org/project/bdgd2dss/
https://pypi.org/project/bdgd2dss/
https://github.com/pauloradatz/bdgd2opendss
https://github.com/pauloradatz/bdgd2opendss
https://dx.doi.org/10.1109/TDC-LA.2012.6319069
https://dx.doi.org/10.1016/j.prime.2024.100783
https://dx.doi.org/10.1109/OSMSES62085.2024.10668999
https://dx.doi.org/10.1109/TPWRS.2023.3315543
https://dx.doi.org/10.1155/2022/1446479
https://dx.doi.org/10.3390/su15086382


2390 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 62, NO. 2, MARCH/APRIL 2026

[19] P. Radatz, “Py-dss-interface,” 2023. Accessed: Apr. 04, 2025. [Online].
Available: https://pypi.org/project/py-dss-interface/

[20] E d. O. Monteiro and A. L. F. Filho, “Comparative analysis of reg-
ulatory procedures for classifiying short-duration voltage variations,”
in Proc. 17th Int. Conf. Harmon. Qual. Power, 2016, pp. 569–574,
doi: 10.1109/ICHQP.2016.7783308.

[21] H. Zhang, G. Xiao, and Z. Lu, “Fast detection of short-duration grid
voltage RMS variation for compensation applications,” in Proc. 2021
IEEE 12th Energy Convers. Congr. Expo. - Asia, 2021, pp. 833–838,
doi: 10.1109/ECCE-Asia49820.2021.9479387.

[22] A. Fioravanti, A. Prudenzi, E. Fiorucci, A. Silvestri, and S. Mari, “The ef-
fect of short duration voltage sags on LED lamps,” in Proc. 2024 Int. Symp.
Power Electron., Elect. Drives, Automat. Motion, 2024, pp. 436–441,
doi: 10.1109/SPEEDAM61530.2024.10609106.

[23] L. A. d. Costa, D. d. S. Gazzana, and R. C. Leborgne, “Fuzzy logic applied
to impact assessment of short-duration RMS voltage variation,” in Proc.
54th Int. Universities Power Eng. Conf., 2019, pp. 1–6.

[24] M. Priyadarshini and M. Sushama, “Classification of short-duration volt-
age variations using wavelet decomposition based entropy criteria,” in
Proc. 2016 Int. Conf. Wireless Commun., Signal Process. Netw., 2016,
pp. 2192–2196.

[25] M. S. Priyadarshini, D. Krishna, M. B. Reddy, A. Bhatt, M. Bajaj, and
M. M. Mahmoud, “Continuous wavelet transform based visualization of
transient and short duration voltage variations,” in Proc. 4th IEEE Glob.
Conf. Advance. Technol., 2023, pp. 1–6.

[26] P. R. Kamthekar, P. V. Gautam, and R. K. Munje, “Detection, characteri-
zation and classification of short duration voltage events using DWT and
fuzzy logic,” in Proc. 2017 Int. Conf. Innov. Mechanisms Ind. Appl., 2017,
pp. 242–247.

[27] D. K. Filiu and J. C. Cebrian, “A novel stochastic methodology to
estimate the impact factor due to short-duration voltage variations in
power distribution networks,” Electric Power Syst. Res., vol. 247, 2025,
Art. no. 111773. [Online]. Available: https://www.sciencedirect.com/
science/article/pii/S0378779625003657

[28] V. Liubčuk, G. Kairaitis, V. Radziukynas, and D. Naujokaitis, “IIR shelving
filter, support vector machine and k-nearest neighbors algorithm appli-
cation for voltage transients and short-duration RMS variations analy-
sis,” Inventions, vol. 9, no. 1, 2024, Art. no. 12. [Online]. Available:
https://www.mdpi.com/2411-5134/9/1/12

[29] L. A. d. Costa, Y. Mohammadi, R. C. Leborgne, and D. d. S. Gazzana,
“Impact evaluation of the neutral-grounding resistance on short-duration
RMS voltage variations,” in Proc. 19th Int. Conf. Harmon. Qual. Power,
2020, pp. 1–6.

[30] C. Yuan et al., “A critical load voltage regulator with both voltage sag
compensation and fluctuation mitigation functions,” Int. J. Elect. Power
Energy Syst., vol. 170, 2025, Art. no. 110870. [Online]. Available: https:
//www.sciencedirect.com/science/article/pii/S0142061525004181

[31] V. A. Skolota and G. S. Zinovev, “Detecting voltage swell, inter-
ruption and sag,” in Proc. 19th Int. Conf. Young Specialists Mi-
cro/Nanotechnologies Electron Devices, 2018, pp. 6403–6408, doi:
10.1109/EDM.2018.8434940.

[32] M. Paolone et al., “Fundamentals of power systems modelling in the
presence of converter-interfaced generation,” Electric Power Syst. Res.,
vol. 189, 2020, Art. no. 106811, doi: 10.1016/j.epsr.2020.106811.

[33] C. M. Furse, M. Kafal, R. Razzaghi, and Y.-J. Shin, “Fault diagnosis for
electrical systems and power networks: A review,” IEEE Sensors J., vol. 21,
no. 2, pp. 888–906, Jan. 2021, doi: 10.1109/JSEN.2020.2987321.

[34] U. A. Bordalo, A. B. Rodrigues, and M. G. Silva, “A new methodology
for probabilistic short-circuit evaluation with applications in power quality
analysis,” IEEE Trans. Power Syst., vol. 21, no. 2, pp. 474–479, May 2006,
doi: 10.1109/TPWRS.2006.873055.

[35] D. B. Martins, “Probabilistic short-circuit analysis in power
distribution systems with distributed generation,” Master in
Electrical Engineering (Dissertation), São Carlos School Eng.,
Univ. São Paulo, São Carlos, Brazil, (in Portuguese), 2021,
doi: 10.11606/D.18.2021.tde-24032021-102152.

[36] V. Kumar, P. Jeyanthy, K. Mahesh, P. Akila, V. S. Thrisha, and B.
M. Swetha, “Fault classification & detection in IEEE 34 bus sys-
tem using convolutional neural network,” in Proc. 15th Int. Conf.
Comput. Commun. Netw. Technol., 2024, pp. 1–6, doi: 10.1109/ICC-
CNT61001.2024.10725909.

[37] W. L. Rodrigues Jr, D. M. Reis, F. A. S. Borges, F. H. D. Araújo, A.
O. d. C. Filho, and R. d. A. L. Rabelo, “Localization of voltage SAG
sources using convolutional neural network in IEEE 34-bus system,” in
Proc. 2020 IEEE Int. Conf. Syst., Man, Cybern., 2020, pp. 2836–2841,
doi: 10.1109/SMC42975.2020.9283083.

[38] I. R. A. Rodrigues and A. d. Conti, “Protection study of a microgrid
based on the 34-Bus IEEE distribution feeder,” in Proc. 2018 Simpósio
Brasileiro de Sistemas Eletricos, 2018, pp. 1–6, doi: 10.1109/SBSE.2018.
8395588.

[39] R. C. Dugan and W. H. Kersting, “Induction machine test case for the
34-bus test feeder - description,” in Proc. 2006 IEEE Power Eng. Soc.
Gen. Meeting, 2006, pp. 1–4, doi: 10.1109/PES.2006.1709506.

[40] Google, “Google Earth,” 2025. Accessed: Apr. 04, 2025. [Online]. Avail-
able: https://www.google.com/earth/

[41] ANEEL, “CEMIG-D 4950 2022-12-31,” 2022. Accessed: Apr.
04, 2025. [Online]. Available: https://dadosabertos-aneel.opendata.arcgis.
com/search?collection=Dataset

[42] ANEEL, “PRODIST - Procedimentos de distribuição de energia elétrica
no sistema elétrico nacional,” 2022. Accessed: Sep. 06, 2025 (in Por-
tuguese). [Online]. Available: https://www.gov.br/aneel/pt-br/centrais-
de-conteudos/procedimentos-regulatorios/prodist

[43] A. G. d. Souza, “Attribution-SDVV-feeders,” GitHub Repository, 2025.
Accessed: Aug. 31, 2025. [Online]. Available: https://github.com/
ArthurGS97/Attribution-SDVV-feeders

[44] ANEEL, “Module 7 - Calculation of distribution losses,” 2022. Accessed:
Sep. 02, 2025 (in Portuguese). [Online]. Available: https://www2.aneel.
gov.br/cedoc/aren2021956_2_6.pdf

[45] ANEEL, “Module 10 - Regulatory geographic information system,”
2022. Accessed: Apr. 04, 2025 (in Portuguese). [Online]. Available:
https://www2.aneel.gov.br/cedoc/aren2021956_2_9.pdf

[46] ANEEL, “Technical Report n. 0057/2014-SRD/ANEEL,” 2014.
Accessed: Apr. 04, 2025 (in Portuguese). [Online]. Available:
https://www2.aneel.gov.br/aplicacoes/audiencia/arquivo/2014/026/
documento/nota_tecnica_0057_srd.pdf

[47] Python Software Foundation, “Python-TM,” 2025. Accessed: Apr.
20, 2025. [Online]. Available: https://www.python.org/

[48] W. McKinney, “Data structures for statistical computing in python,”
in Proc. 9th Python Sci. Conf., S. van der Walt and J. Millman,
Eds., Austin, TX, USA, 2010, pp. 56–61, doi: 10.25080/Majora-
92bf1922-00a.

[49] J. D. Hunter, “Matplotlib: A 2D graphics environment,” Comput. Sci. Eng.,
vol. 9, no. 3, pp. 90–95, 2007, doi: 10.1109/MCSE.2007.55.

[50] EPRI, “OpenDSS manual,” 2020. Accessed: Apr. 04, 2025. [Online].
Available: https://www.epri.com/pages/sa/opendss

[51] A. L. S. Pessoa and M. Oleskovicz, “Fault location algorithm
for distribution systems with distributed generation based on
impedance and metaheuristic methods,” Electric Power Syst. Res.,
vol. 225, 2023, Art. no. 109871, doi: 10.1016/j.epsr.2023.109871.
[Online]. Available: https://www.sciencedirect.com/science/article/pii/
S0378779623007599?via\%3Dihub

[52] A. P. Mazzini, W. M. S. Bernardes, F. M. d. Vasconcelos, F. d. O.
Saraiva, and E. N. Asada, “Application of artificial neural networks for
discrimination of nonlinear loads,” in Proc. 2013 IEEE PES Conf. Innov.
Smart Grid Technol. (ISGT Latin Amer.), 2013, pp. 1–7, doi: 10.1109/IS-
GT-LA.2013.6554422.

[53] F. M. Vasconcelos et al., “Artificial neural network applied to prediction
of electricity generated by grid connected photovoltaic systems,” in Proc.
2013 IEEE PES Conf. Innov. Smart Grid Technol. (ISGT Latin Amer.),
2013, pp. 1–6, doi: 10.1109/ISGT-LA.2013.6554453.

[54] F. O. Saraiva, W. M. S. Bernardes, and E. N. Asada, “A framework for
classification of non-linear loads in smart grids using artificial neural net-
works and multi-agent systems,” Neurocomputing, vol. 170, pp. 328–338,
2015, doi: 10.1016/j.neucom.2015.02.090. [Online]. Available: https://
www.sciencedirect.com/science/article/abs/pii/S0925231215009285

[55] W. M. S. Bernardes and F. M. Vasconcelos, “Forecasting electric power
generation of grid-connected solar photovoltaic systems by using artificial
neural networks,” in Proc. Simpósio Brasileiro de Automação Inteligente,
2019, pp. 1–8, doi: 10.17648/sbai-2019-111138, (in Portuguese). [On-
line]. Available: https://proceedings.science/p/111138

[56] B. Masalmeh, R. Prasad, V. Nougain, and A. Lekić, “Neural networks
in RSCAD: Enhancing MMC-based HVDC simulation with advanced
machine learning,” IEEE Trans. Ind. Appl., vol. 61, no. 2, pp. 2515–2526,
Mar./Apr. 2025, doi: 10.1109/TIA.2025.3529804.

[57] W. Lin, F. Li, Q. Ran, and Z. Yang, “Training a trustworthy
DNN for optimal power flow in active distribution networks,” IEEE
Trans. Ind. Appl., vol. 61, no. 1, pp. 1716–1724, Jan./Feb. 2025,
doi: 10.1109/TIA.2024.3430270.

[58] U. Qureshi et al., “Optimizing electric vehicle integration in virtual power
plants: A stochastic optimization framework with MDNN integration,”
IEEE Trans. Ind. Appl., vol. 60, no. 6, pp. 9227–9236, Nov./Dec. 2024,
doi: 10.1109/TIA.2024.3444744.

https://pypi.org/project/py-dss-interface/
https://dx.doi.org/10.1109/ICHQP.2016.7783308
https://dx.doi.org/10.1109/ECCE-Asia49820.2021.9479387
https://dx.doi.org/10.1109/SPEEDAM61530.2024.10609106
https://www.sciencedirect.com/science/article/pii/S0378779625003657
https://www.sciencedirect.com/science/article/pii/S0378779625003657
https://www.mdpi.com/2411-5134/9/1/12
https://www.sciencedirect.com/science/article/pii/S0142061525004181
https://www.sciencedirect.com/science/article/pii/S0142061525004181
https://dx.doi.org/10.1109/EDM.2018.8434940
https://dx.doi.org/10.1016/j.epsr.2020.106811
https://dx.doi.org/10.1109/JSEN.2020.2987321
https://dx.doi.org/10.1109/TPWRS.2006.873055
https://dx.doi.org/10.11606/D.18.2021.tde-24032021-102152
https://dx.doi.org/10.1109/ICCCNT61001.2024.10725909
https://dx.doi.org/10.1109/ICCCNT61001.2024.10725909
https://dx.doi.org/10.1109/SMC42975.2020.9283083
https://dx.doi.org/10.1109/SBSE.2018.8395588
https://dx.doi.org/10.1109/SBSE.2018.8395588
https://dx.doi.org/10.1109/PES.2006.1709506
https://www.google.com/earth/
https://dadosabertos-aneel.opendata.arcgis.com/search{?}collection$=$Dataset
https://dadosabertos-aneel.opendata.arcgis.com/search{?}collection$=$Dataset
https://www.gov.br/aneel/pt-br/centrais-de-conteudos/procedimentos-regulatorios/prodist
https://www.gov.br/aneel/pt-br/centrais-de-conteudos/procedimentos-regulatorios/prodist
https://github.com/ArthurGS97/Attribution-SDVV-feeders
https://github.com/ArthurGS97/Attribution-SDVV-feeders
https://www2.aneel.gov.br/cedoc/aren2021956_2_6.pdf
https://www2.aneel.gov.br/cedoc/aren2021956_2_6.pdf
https://www2.aneel.gov.br/cedoc/aren2021956_2_9.pdf
https://www2.aneel.gov.br/aplicacoes/audiencia/arquivo/2014/026/documento/nota_tecnica_0057_srd.pdf
https://www2.aneel.gov.br/aplicacoes/audiencia/arquivo/2014/026/documento/nota_tecnica_0057_srd.pdf
https://www.python.org/
https://dx.doi.org/10.25080/Majora-92bf1922-00a
https://dx.doi.org/10.25080/Majora-92bf1922-00a
https://dx.doi.org/10.1109/MCSE.2007.55
https://www.epri.com/pages/sa/opendss
https://dx.doi.org/10.1016/j.epsr.2023.109871
https://www.sciencedirect.com/science/article/pii/S0378779623007599{?}viaLY1	extbackslash %3Dihub
https://www.sciencedirect.com/science/article/pii/S0378779623007599{?}viaLY1	extbackslash %3Dihub
https://dx.doi.org/10.1109/ISGT-LA.2013.6554422
https://dx.doi.org/10.1109/ISGT-LA.2013.6554422
https://dx.doi.org/10.1109/ISGT-LA.2013.6554453
https://dx.doi.org/10.1016/j.neucom.2015.02.090
https://www.sciencedirect.com/science/article/abs/pii/S0925231215009285
https://www.sciencedirect.com/science/article/abs/pii/S0925231215009285
https://dx.doi.org/10.17648/sbai-2019-111138
https://proceedings.science/p/111138
https://dx.doi.org/10.1109/TIA.2025.3529804
https://dx.doi.org/10.1109/TIA.2024.3430270
https://dx.doi.org/10.1109/TIA.2024.3444744


DE SOUZA et al.: ATTRIBUTION OF RESPONSIBILITY FOR SHORT-DURATION VOLTAGE VARIATIONS 2391

Arthur Gomes de Souza was born in Correntina,
Bahia, Brazil. He received the bachelor’s degree in
electrical engineering and the M.Sc. degree in elec-
trical engineering with a focus on power system pro-
tection from the Federal University of Uberlândia,
Uberlândia, Brazil, in 2021, He is currently working
toward the Ph.D. degree. His graduate studies are
financially supported by Coordination for the Im-
provement of Higher Education Personnel. As a Re-
searcher with the Laboratory of Alternative Energies
and Power System Protection, he has worked on a

range of topics, including feeder modeling and simulation, and the optimization
of protection systems in electrical networks. He is proficient in tools such
as OpenDSS, PythonTM and QGIS, with several publications demonstrating
his expertise in these areas. His research interests include electrical system
modeling, protection coordination, and the application of advanced optimization
techniques—such as evolutionary algorithms for coordinating protective devices
in distribution networks.

Luiz Arthur Tarralo Passatuto was born in Colina,
São Paulo, Brazil. He received the B.Sc. and M.Sc.
degrees in electrical engineering in 2021 and 2025, re-
spectively, from the Federal University of Uberlândia,
Uberlândia, Brazil„ where he is currently working
toward the Ph.D. degree. As a Researcher with the
Laboratory of Alternative Energies and Power System
Protection, he has worked with optimization in elec-
tric power system protection using metaheuristics,
developing software tools that use Blockchain and
power system modeling with distributed energy re-

sources. He has good knowledge in programming with PythonTM and OpenDSS
tool. His research interests include power system analysis, protection studies,
solving engineering problems with optimization techniques, and incorporation
of distributed energy resources in the electrical network.

Wellington Maycon Santos Bernardes was born
in Goiânia - Goiás, Brazil. He received the B.Sc.
degree in electrical engineering from Federal Uni-
versity of Uberlândia (UFU), Uberlândia, Brazil, in
2010, and the M.Sc. and Ph.D. degrees in electrical
engineering from São Carlos School of Engineering,
University of São Paulo, São Paulo, Brazil, in 2013
and 2018, respectively. He has undertaken a mobility
period (Sandwich Ph.D.) supported by the prestigious
scholarship of the European Commission’s Erasmus
Mundus Programme with the Faculty of Engineering,

University of Porto, Porto, Portugal. He was a Professor with UFU in 2019, a
Coordinator with the Laboratory of Alternative Energies and Power System
Protection, an ad hoc Consultant for the State Funding Agency of Distrito
Federal, and the National Institute of Educational Studies and Research Aní-
sio Teixeira, Brasília - DF. Extensive experience in academic administration,
serving as Coordinator with the Department of Electric Power Systems, and a
Undergraduate Coordinator of the Electrical Engineering Program. He also has
expertise in power systems projects, funded by Brazilian research foundations
(CNPq, CAPES, FAPEMIG, FAPESP) and utility companies. His research
interests include power system protection, alternative energies, measurement
and uncertainty analysis, optimization and applications of artificial intelligence
in electrical networks, energy efficiency, and power quality.

Luiz Carlos Gomes Freitas was born in Uberlân-
dia, Brazil, in 1976. He received an undergraduate
degree in electrical engineering with emphasis on
power systems, and the M.Sc. and Ph.D. degrees
in electrical engineering with emphasis on power
electronics from the Federal University of Uberlândia
(UFU), Uberlândia, Brazil, in 2001, 2003, and 2006,
respectively. In his doctoral thesis, he developed an
innovative topological design of a three-phase hybrid
rectifier for high power drive systems. In 2008, he
joined as the Faculty Member with the UFU, where

he is developing teaching and research activities in the area of power electronics
and power systems. Since 2010, he is the Coordinator of the Power Electronics
Research Center, Faculty of Electrical Engineering, UFU. His research interests
include electrical engineering, conversion and rectification of electric energy,
working on various topics related to power electronics, electric power quality,
and renewable energy. In 2012, he was th recipient of the Second Prize Paper
Award of the IEEE-IAS-Industrial Automation and Control Committee for his
contribution to the development of hybrid rectifier structures. Since 2013, he is a
Researcher recognized by the National Council for Scientific and Technological
Development (CNPq) with a Research Productivity Grant.

Ênio Costa Resende received the bachelor’s degree
in electrical engineering from the Federal University
of Uberlândia (UFU), Uberlândia, Brazil, including
an exchange period through the CsF / CAPES Pro-
gram with Sacramento State University (SAC-State),
Sacramento, CA, USA, in 2018, the M.Sc. degree in
electrical engineering with a specialization in power
electronics from Power Electronics Research Center,
UFU, Postgraduate Program, in 2020, and the Ph.D.
degree in electrical engineering in 2024. During his
doctoral studies, he was a Visiting Researcher with the

University of Vaasa, Vaasa, Finland, from 2023 to 2024, where he worked on
Neural Networks and Fuzzy Logic, focusing on islanding detection techniques,
synchronization strategies, and maximum power point tracking (MPPT) meth-
ods. He is currently a Postdoctoral Researcher with UFU, working on hybrid
microgrids, battery charging systems, and electric vehicle charging stations. His
research interests include islanding detection, distributed generation, renewable
energies, MPPT, control methods, power quality, microgrids, embedded control,
and dual-active-bridge converters.



<<
  /ASCII85EncodePages false
  /AllowTransparency false
  /AutoPositionEPSFiles true
  /AutoRotatePages /None
  /Binding /Left
  /CalGrayProfile (Gray Gamma 2.2)
  /CalRGBProfile (sRGB IEC61966-2.1)
  /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2)
  /sRGBProfile (sRGB IEC61966-2.1)
  /CannotEmbedFontPolicy /Warning
  /CompatibilityLevel 1.4
  /CompressObjects /Off
  /CompressPages true
  /ConvertImagesToIndexed true
  /PassThroughJPEGImages true
  /CreateJobTicket false
  /DefaultRenderingIntent /Default
  /DetectBlends true
  /DetectCurves 0.0000
  /ColorConversionStrategy /sRGB
  /DoThumbnails true
  /EmbedAllFonts true
  /EmbedOpenType false
  /ParseICCProfilesInComments true
  /EmbedJobOptions true
  /DSCReportingLevel 0
  /EmitDSCWarnings false
  /EndPage -1
  /ImageMemory 1048576
  /LockDistillerParams true
  /MaxSubsetPct 100
  /Optimize true
  /OPM 0
  /ParseDSCComments false
  /ParseDSCCommentsForDocInfo true
  /PreserveCopyPage true
  /PreserveDICMYKValues true
  /PreserveEPSInfo false
  /PreserveFlatness true
  /PreserveHalftoneInfo true
  /PreserveOPIComments false
  /PreserveOverprintSettings true
  /StartPage 1
  /SubsetFonts true
  /TransferFunctionInfo /Remove
  /UCRandBGInfo /Preserve
  /UsePrologue false
  /ColorSettingsFile ()
  /AlwaysEmbed [ true
    /Algerian
    /Arial-Black
    /Arial-BlackItalic
    /Arial-BoldItalicMT
    /Arial-BoldMT
    /Arial-ItalicMT
    /ArialMT
    /ArialNarrow
    /ArialNarrow-Bold
    /ArialNarrow-BoldItalic
    /ArialNarrow-Italic
    /ArialUnicodeMS
    /BaskOldFace
    /Batang
    /Bauhaus93
    /BellMT
    /BellMTBold
    /BellMTItalic
    /BerlinSansFB-Bold
    /BerlinSansFBDemi-Bold
    /BerlinSansFB-Reg
    /BernardMT-Condensed
    /BodoniMTPosterCompressed
    /BookAntiqua
    /BookAntiqua-Bold
    /BookAntiqua-BoldItalic
    /BookAntiqua-Italic
    /BookmanOldStyle
    /BookmanOldStyle-Bold
    /BookmanOldStyle-BoldItalic
    /BookmanOldStyle-Italic
    /BookshelfSymbolSeven
    /BritannicBold
    /Broadway
    /BrushScriptMT
    /CalifornianFB-Bold
    /CalifornianFB-Italic
    /CalifornianFB-Reg
    /Centaur
    /Century
    /CenturyGothic
    /CenturyGothic-Bold
    /CenturyGothic-BoldItalic
    /CenturyGothic-Italic
    /CenturySchoolbook
    /CenturySchoolbook-Bold
    /CenturySchoolbook-BoldItalic
    /CenturySchoolbook-Italic
    /Chiller-Regular
    /ColonnaMT
    /ComicSansMS
    /ComicSansMS-Bold
    /CooperBlack
    /CourierNewPS-BoldItalicMT
    /CourierNewPS-BoldMT
    /CourierNewPS-ItalicMT
    /CourierNewPSMT
    /EstrangeloEdessa
    /FootlightMTLight
    /FreestyleScript-Regular
    /Garamond
    /Garamond-Bold
    /Garamond-Italic
    /Georgia
    /Georgia-Bold
    /Georgia-BoldItalic
    /Georgia-Italic
    /Haettenschweiler
    /HarlowSolid
    /Harrington
    /HighTowerText-Italic
    /HighTowerText-Reg
    /Impact
    /InformalRoman-Regular
    /Jokerman-Regular
    /JuiceITC-Regular
    /KristenITC-Regular
    /KuenstlerScript-Black
    /KuenstlerScript-Medium
    /KuenstlerScript-TwoBold
    /KunstlerScript
    /LatinWide
    /LetterGothicMT
    /LetterGothicMT-Bold
    /LetterGothicMT-BoldOblique
    /LetterGothicMT-Oblique
    /LucidaBright
    /LucidaBright-Demi
    /LucidaBright-DemiItalic
    /LucidaBright-Italic
    /LucidaCalligraphy-Italic
    /LucidaConsole
    /LucidaFax
    /LucidaFax-Demi
    /LucidaFax-DemiItalic
    /LucidaFax-Italic
    /LucidaHandwriting-Italic
    /LucidaSansUnicode
    /Magneto-Bold
    /MaturaMTScriptCapitals
    /MediciScriptLTStd
    /MicrosoftSansSerif
    /Mistral
    /Modern-Regular
    /MonotypeCorsiva
    /MS-Mincho
    /MSReferenceSansSerif
    /MSReferenceSpecialty
    /NiagaraEngraved-Reg
    /NiagaraSolid-Reg
    /NuptialScript
    /OldEnglishTextMT
    /Onyx
    /PalatinoLinotype-Bold
    /PalatinoLinotype-BoldItalic
    /PalatinoLinotype-Italic
    /PalatinoLinotype-Roman
    /Parchment-Regular
    /Playbill
    /PMingLiU
    /PoorRichard-Regular
    /Ravie
    /ShowcardGothic-Reg
    /SimSun
    /SnapITC-Regular
    /Stencil
    /SymbolMT
    /Tahoma
    /Tahoma-Bold
    /TempusSansITC
    /TimesNewRomanMT-ExtraBold
    /TimesNewRomanMTStd
    /TimesNewRomanMTStd-Bold
    /TimesNewRomanMTStd-BoldCond
    /TimesNewRomanMTStd-BoldIt
    /TimesNewRomanMTStd-Cond
    /TimesNewRomanMTStd-CondIt
    /TimesNewRomanMTStd-Italic
    /TimesNewRomanPS-BoldItalicMT
    /TimesNewRomanPS-BoldMT
    /TimesNewRomanPS-ItalicMT
    /TimesNewRomanPSMT
    /Times-Roman
    /Trebuchet-BoldItalic
    /TrebuchetMS
    /TrebuchetMS-Bold
    /TrebuchetMS-Italic
    /Verdana
    /Verdana-Bold
    /Verdana-BoldItalic
    /Verdana-Italic
    /VinerHandITC
    /Vivaldii
    /VladimirScript
    /Webdings
    /Wingdings2
    /Wingdings3
    /Wingdings-Regular
    /ZapfChanceryStd-Demi
    /ZWAdobeF
  ]
  /NeverEmbed [ true
  ]
  /AntiAliasColorImages false
  /CropColorImages true
  /ColorImageMinResolution 150
  /ColorImageMinResolutionPolicy /OK
  /DownsampleColorImages false
  /ColorImageDownsampleType /Bicubic
  /ColorImageResolution 900
  /ColorImageDepth -1
  /ColorImageMinDownsampleDepth 1
  /ColorImageDownsampleThreshold 1.00111
  /EncodeColorImages true
  /ColorImageFilter /DCTEncode
  /AutoFilterColorImages true
  /ColorImageAutoFilterStrategy /JPEG
  /ColorACSImageDict <<
    /QFactor 0.76
    /HSamples [2 1 1 2] /VSamples [2 1 1 2]
  >>
  /ColorImageDict <<
    /QFactor 0.40
    /HSamples [1 1 1 1] /VSamples [1 1 1 1]
  >>
  /JPEG2000ColorACSImageDict <<
    /TileWidth 256
    /TileHeight 256
    /Quality 15
  >>
  /JPEG2000ColorImageDict <<
    /TileWidth 256
    /TileHeight 256
    /Quality 15
  >>
  /AntiAliasGrayImages false
  /CropGrayImages true
  /GrayImageMinResolution 150
  /GrayImageMinResolutionPolicy /OK
  /DownsampleGrayImages false
  /GrayImageDownsampleType /Bicubic
  /GrayImageResolution 1200
  /GrayImageDepth -1
  /GrayImageMinDownsampleDepth 2
  /GrayImageDownsampleThreshold 1.00083
  /EncodeGrayImages true
  /GrayImageFilter /DCTEncode
  /AutoFilterGrayImages true
  /GrayImageAutoFilterStrategy /JPEG
  /GrayACSImageDict <<
    /QFactor 0.76
    /HSamples [2 1 1 2] /VSamples [2 1 1 2]
  >>
  /GrayImageDict <<
    /QFactor 0.40
    /HSamples [1 1 1 1] /VSamples [1 1 1 1]
  >>
  /JPEG2000GrayACSImageDict <<
    /TileWidth 256
    /TileHeight 256
    /Quality 15
  >>
  /JPEG2000GrayImageDict <<
    /TileWidth 256
    /TileHeight 256
    /Quality 15
  >>
  /AntiAliasMonoImages false
  /CropMonoImages true
  /MonoImageMinResolution 1200
  /MonoImageMinResolutionPolicy /OK
  /DownsampleMonoImages false
  /MonoImageDownsampleType /Bicubic
  /MonoImageResolution 1600
  /MonoImageDepth -1
  /MonoImageDownsampleThreshold 1.00063
  /EncodeMonoImages true
  /MonoImageFilter /CCITTFaxEncode
  /MonoImageDict <<
    /K -1
  >>
  /AllowPSXObjects false
  /CheckCompliance [
    /None
  ]
  /PDFX1aCheck false
  /PDFX3Check false
  /PDFXCompliantPDFOnly false
  /PDFXNoTrimBoxError true
  /PDFXTrimBoxToMediaBoxOffset [
    0.00000
    0.00000
    0.00000
    0.00000
  ]
  /PDFXSetBleedBoxToMediaBox true
  /PDFXBleedBoxToTrimBoxOffset [
    0.00000
    0.00000
    0.00000
    0.00000
  ]
  /PDFXOutputIntentProfile (None)
  /PDFXOutputConditionIdentifier ()
  /PDFXOutputCondition ()
  /PDFXRegistryName ()
  /PDFXTrapped /False

  /CreateJDFFile false
  /Description <<
    /CHS <FEFF4f7f75288fd94e9b8bbe5b9a521b5efa7684002000410064006f006200650020005000440046002065876863900275284e8e55464e1a65876863768467e5770b548c62535370300260a853ef4ee54f7f75280020004100630072006f0062006100740020548c002000410064006f00620065002000520065006100640065007200200035002e003000204ee553ca66f49ad87248672c676562535f00521b5efa768400200050004400460020658768633002>
    /CHT <FEFF4f7f752890194e9b8a2d7f6e5efa7acb7684002000410064006f006200650020005000440046002065874ef69069752865bc666e901a554652d965874ef6768467e5770b548c52175370300260a853ef4ee54f7f75280020004100630072006f0062006100740020548c002000410064006f00620065002000520065006100640065007200200035002e003000204ee553ca66f49ad87248672c4f86958b555f5df25efa7acb76840020005000440046002065874ef63002>
    /DAN <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>
    /DEU <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>
    /ESP <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>
    /FRA <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>
    /ITA (Utilizzare queste impostazioni per creare documenti Adobe PDF adatti per visualizzare e stampare documenti aziendali in modo affidabile. I documenti PDF creati possono essere aperti con Acrobat e Adobe Reader 5.0 e versioni successive.)
    /JPN <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>
    /KOR <FEFFc7740020c124c815c7440020c0acc6a9d558c5ec0020be44c988b2c8c2a40020bb38c11cb97c0020c548c815c801c73cb85c0020bcf4ace00020c778c1c4d558b2940020b3700020ac00c7a50020c801d569d55c002000410064006f0062006500200050004400460020bb38c11cb97c0020c791c131d569b2c8b2e4002e0020c774b807ac8c0020c791c131b41c00200050004400460020bb38c11cb2940020004100630072006f0062006100740020bc0f002000410064006f00620065002000520065006100640065007200200035002e00300020c774c0c1c5d0c11c0020c5f40020c2180020c788c2b5b2c8b2e4002e>
    /NLD (Gebruik deze instellingen om Adobe PDF-documenten te maken waarmee zakelijke documenten betrouwbaar kunnen worden weergegeven en afgedrukt. De gemaakte PDF-documenten kunnen worden geopend met Acrobat en Adobe Reader 5.0 en hoger.)
    /NOR <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>
    /PTB <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>
    /SUO <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>
    /SVE <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>
    /ENU (Use these settings to create PDFs that match the "Suggested"  settings for PDF Specification 4.0)
  >>
>> setdistillerparams
<<
  /HWResolution [600 600]
  /PageSize [612.000 792.000]
>> setpagedevice