energies

Article
Voltage Estimation Method for Power Distribution
Networks Using High-Precision Measurements
Chan-Hyeok Oh , Seok-Il Go , Joon-Ho Choi , Seon-Ju Ahn * and Sang-Yun Yun *
Department of Electrical Engineering in Chonnam National University, Gwangju 61186, Korea;
dhcksgur@naver.com (C.-H.O.); riseisgood@nate.com (S.-I.G.); joono@chonnam.ac.kr (J.-H.C.)
* Correspondence: sjahn@chonnam.ac.kr (S.-J.A.); drk9034@jnu.ac.kr (S.-Y.Y.);
Tel.: +82-62-530-1738 (S.-J.A.); +82-62-530-1745 (S.-Y.Y.)




Received: 9 March 2020; Accepted: 6 May 2020; Published: 10 May 2020


Abstract: In this study, we propose a voltage estimation method for the radial distribution network
with distributed generators (DGs) using high-precision measurements (HPMs). The proposed method
uses the section loads center for voltage estimation because individual loads are not measured in the
distribution system. The bus voltage was estimated through correction of the section load center by
using an HPM at the end of the main feeder. The correction parameter of the section load center was
calculated by comparing the initial voltage estimates and the measurements of the HPMs. After that,
the voltage of the main feeder was re-estimated. Finally, the bus voltage in the lateral feeder was
estimated based on the voltage estimates in the main feeder and the current measurements in the
lateral feeder. The accuracy of the proposed algorithm was verified through case studies by using
test systems implemented in MATLAB, Simulink, and Python environments. In order to verify the
utilization of the proposed method to the practical system, a test with injection of approximately
5% of normally distributed random noise was performed. Through the results of the case studies,
when an HPM is installed at the end of the main feeder, it demonstrated that the voltage estimation
accuracy can be greatly improved by the proposed method. Compared with the existing methods,
the proposed method was less affected by PV and showed robustness to measurement noise.

Keywords: voltage estimation; section load center; high-precision measurement; radial distribution
network

1. Introduction
In recent years, the increasing number of distributed generators (DGs) and reduction of the
profitability for power companies have resulted in a change in the existing operation method that
overcomes the variability of the distribution network by sufficient infrastructural investments at the
planning stage. Voltage and overload problems that are caused by loads and DGs occur during
relatively short durations. Real-time monitoring and active control are considerably required to replace
conventional excessive infrastructure investments as a solution to improve operational efficiency.
Therefore, a new alternative is arising, which is known as the active distribution network in contradiction
to the passive distribution network [1]. This change is cost-effective and improves the hosting capacity
of DGs. However, the active distribution system implies a certain degree of risk in comparison with
the conventional system.
To overcome this risk, various distribution management systems (DMSs) that adopt advanced
control technologies, such as voltage-VAR optimization (VVO), fault location, isolation and service
restoration (FLISR), and real-time network reconfiguration (RTNR), have been developed and
applied [2]. In order for these control technologies, the accuracy of the voltage, current, and load data
are crucial. Accurate measurements are required to derive accurate solutions. However, there are

Energies 2020, 13, 2385; doi:10.3390/en13092385 www.mdpi.com/journal/energies

Energies 2020, 13, 2385 2 of 23

errors in the measurements and difficulty in measuring all actual loads. Especially, the high voltage
distribution network has a very narrow voltage control range [3], and the accuracy of the voltage
measurements generally has an error rate of 2% and above [4]. Consequently, calculating control
recommendations using voltage measurements may not be useful for the operation of an actual network.
Moreover, all of the actual loads cannot be measured in real time but can only be modeled as section
loads between the measurement devices. Accordingly, the calculation of control recommendations is
also affected by the method used for section load models [5]. Recently, high-precision measurements
(HPMs), such as power quality meter (PQ meter) and phasor measurement unit (PMU), have been
applied more for power quality monitoring and fault detection [6–8]. By using HPMs, the estimation
error at the installation point of the HPM can be known, so it can be corrected to increase the estimation
accuracy. Therefore, it is necessary to improve the utilization of the advanced controls at the actual
distribution network.
Many studies have been conducted on the voltage estimations in a distribution network.
These studies can be classified into two groups. One group is to deal with state estimation (SE)
using the weighted least square (WLS) method, and the other group is to estimate voltages and section
loads considering the characteristics of the distributed loads [9–14]. SE is a method for estimating
voltages through the present measured value. Ju et al. proposed a fast decoupled SE using current
measurements [9]. In order to prevent a singular matrix, when the Jacobian matrix was created in
the fast decoupled method, each measured value was transformed to formulas of active and reactive
powers. Kong et al. suggested a SE using measurement data of supervisory control and data acquisition
(SCADA) and PMUs [10]. As SCADA and PMUs have different measurement intervals, a hybrid
method that combines WLS-based and linear SE is proposed. The former was used at intervals when
both SCADA and PMU data were measured. The latter was used at other intervals when only the
PMU data was measured. Sulis et al. proposed a three-phase SE, which considered the unbalanced
characteristics of the distribution network, and a WLS-based SE using the existing measurement devices
and PMU [11,12]. The estimation accuracy was improved by single and three-phase SE using PMU
measurement data. Moreover, a comparative analysis was conducted for the accuracies of different
methods, which utilized existing measurements, PMU measurement data, and both, respectively.
These methods were conducted under the assumption that loads are concentrated at nodes, and their
magnitudes were known values. However, the actual distribution networks do not have concentered
loads but distributed ones. Additionally, it is not possible to accurately identify the magnitude of loads.
Accordingly, it is difficult to apply these methods to actual networks. To handle these difficulties,
Yun et al. suggested a SE that calculated section loads by using current measurements and nominal
voltages and considered them as pseudo measurements [13]. The network reduction can improve
the computing speed for the convergence of SE. However, as a calculated section load is used as
an input for SE, the voltage estimation tends to depend on input values. Wakeel et al. proposed a
voltage estimation method based on iteratively re-weighted least square (IRWLS) using actual load
data and the voltage and active/reactive power measurement of the substation [14]. The load data were
measured using a smart meter, and a pseudo measurement was generated using a K-mean clustering
method if the measurement value was not acquired. However, implementing this method requires
construction of an advanced metering infrastructure (AMI). The other group estimated voltages and
section loads considering the characteristics of the distributed loads in a distribution network [15,16].
Park et al. proposed a method of estimating section loads and voltage magnitude of the remote
terminal unit (RTU) from the voltage measured at a substation and currents measured at the RTU [15].
Park et al. proposed a voltage estimation method for the RTU, which calculates section load centers by
using the contract power for high voltage loads and the capacity of a low voltage transformer [16].
These two methods estimated the section loads and voltages of the RTU. However, as the voltage
measurements of the RTU are not reliable, it is not possible to determine the error in the estimated
values. Moreover, the change in the load center, which is caused by the change of loads with different
patterns, cannot be considered.
Energies 2020, 13, 2385 3 of 23

In this paper, we proposed a method that greatly improves the accuracy of voltage and load
estimation by adding minimal HPMs in a distribution network that does not measure individual
loads and has only the measurements of switches. There are superiorities of the proposed method:
First, it is robust to measurement noise, and therefore has high applicability to an actual distribution
network. Second, it can accurately estimate even for recent situations that increase the interconnection
of renewable energy resources in distribution networks. Through various case studies, the superiorities
of the proposed method were demonstrated. This paper is organized as follows: In Section 2,
the conventional methods of voltage estimation are analyzed. Section 3 presents the proposed method.
In this section, six section types are distinguished to consider the impacts of the DGs. A method of
identifying each section through the measured values is introduced. A correction parameter of the
section load center is proposed based on the voltage measurements by an HPM. Section 4 presents case
studies that were implemented using test systems in MATLAB, Simulink, and Python environments.
The case studies consider measurement noise. The results of the case studies proved that the proposed
method has a higher accuracy of voltage estimation than the conventional method.

2. Existing Voltage Estimation Methods and Motivation of This Research

2.1. Voltage Measurement Methods and the Importance of Its Accuracy

The voltage measurement methods utilize either a capacitor voltage transformer (CVT) or a
bushing capacitance potential device (BCPD). A CVT measures voltages using capacitors arranged in
series. Since capacitors with constant capacitance are used, the measurement error is low. However,
since the CVT requires large installation space and expensive cost, it is usually installed at a substation.
The BCPD-based method utilizes the installation of a conductor ring in an insulated bushing and
measures a capacitance-induced voltage. Since this method is inexpensive and requires a smaller
space, it is widely used for voltage measurement at the remote switch points. However, the voltage
measurements are not accurate at an error rate of 1%~2% or above since the capacitance of the conductor
ring is not uniform [16].
The voltage control ranges of the primary distribution lines are based on the allowable range of the
low voltage (LV) line supplied to customers. Nominal voltage and variation ranges of some countries
are presented on the left side of Table 1 [17]. If this range is converted for the primary side by considering
the voltage drop in a LV line and the tap ratio of a transformer, the result is narrower than the low
voltage range. The allowable voltage drop ranges of American national standards institute(ANSI),
consumer’s electrical installation guide(CEIG), Canadian standards association(CSA) and national fire
protection association(NFPA) are presented on the right side of Table 1. The voltage drops of only
2%~3% were allowed for the reference voltage on the primary side. Taking into consideration the
narrow voltage regulation range, it is crucial to accurately measure the line voltages for efficient and
reliable control.

Table 1. Regulation of voltage control range in each country and standards of voltage drop.

Maximum Voltage Drop

Nominal Variation Range
Country Standard Primary Secondary
Voltage (V) (min./max. limit)
(High Volt) (Low Volt)
Australia 230 −6.1/+10.0%
ANSI C84.1 [3] 0.03 pu 0.05 pu
Canada 120 −8.3/+4.2%
Germany 230 ±10.0%
CEIG (Korea) [18] 0.02 pu 0.04 pu
Japan 100 ±6.0%
Korea 220 ±5.9%
CSA C22.1-15 [19] 0.03 pu 0.05 pu
U.K. 230 −6.0/+10.0%
U.S. 120 ±5.0% NFPA 70 [20] 0.03 pu 0.05 pu
 Energies 2020, 13, x FOR PEER REVIEW 4 of 23
Energies 2020, 13, 2385 4 of 23
2.2. Review of the Existing Studies on Voltage Estimation
2.2. Review of the Existing Studies on Voltage Estimation
2.2.1. Studies on SE Based on WLS
2.2.1. Studies on SE Based
Most WLS-based SEon WLS that loads are concentered on nodes and the magnitudes of loads
assume
are Most
known [9–12]. However,
WLS-based SE assume the that
actual loads
loads areare not concentered
concentered on nodes on nodes
and thebut distributed
magnitudes ofin sections.
loads are
The upper part of Figure 1 shows a typical section of the distribution
known [9–12]. However, the actual loads are not concentered on nodes but distributed in sections. line, where black arrows present
theupper
The load concentrated
part of Figureon nodes,a while
1 shows typicalthe red arrows
section present the line,
of the distribution nodes distributed
where black arrows in thepresent
section.
The
the lower
load graph of Figure
concentrated on nodes, 1 showswhilevoltage
the redprofiles according
arrows present thetonodes
the different
distributed loadintypes. As the
the section.
voltage drop by current flowing in line is expressed by
The lower graph of Figure 1 shows voltage profiles according to the different load types. As the 𝑉 = 𝑍 × 𝐼 , different current
behavior
voltage dropbetween
by currentloadsflowing
concentratedin line on node and by
is expressed load distributed
Vdrop = Zline ×inIlinesection causecurrent
, different different voltage
behavior
between loads concentrated on node and load distributed in section cause different voltage drops.of
drops. The difference of voltage drops increases as the line becomes longer and the magnitude
load
The becomesoflarger.
difference voltageTherefore,
drops increases voltage as estimation
the line becomes usinglonger
the concentrated
and the magnitudeload may cause
of load a larger
becomes
estimation
larger. error than
Therefore, voltagethatestimation
using the distributed load. Accordingly,
using the concentrated load may thecause
characteristic of the distributed
a larger estimation error
load needs to be considered for accurate voltage estimation.
than that using the distributed load. Accordingly, the characteristic of the distributed load In addition, all the loads cannot
needs to bebe
measured,for
considered andaccurate
the exactvoltage
magnitudes of the loads
estimation. cannot be
In addition, alldetermined.
the loads cannot Recently, AMI, suchand
be measured, as smart
the
meter, has been utilized for load measurement. In another voltage
exact magnitudes of the loads cannot be determined. Recently, AMI, such as smart meter, has been estimation study that uses the state
estimation
utilized method,
for load individual loads
measurement. are measured
In another via AMI using
voltage estimation studysmartthatmeter,
uses the andstatethe sum of these
estimation
loads are
method, assumed loads
individual as theare concentrated
measured via loadAMI measurements
using smartofmeter, the medium
and thevoltage
sum of (MV) these side
loads buses.
are
It is a method of estimating the voltage by IRWLS using MV
assumed as the concentrated load measurements of the medium voltage (MV) side buses. It is a side load measurement and the voltage
and active/reactive
method of estimating power measurements
the voltage by IRWLS of using
the substation
MV side[14]. loadInmeasurement
that study, a and pseudo the measurement
voltage and
was generated using a cluster center obtained by the K-mean clustering
active/reactive power measurements of the substation [14]. In that study, a pseudo measurement was method if the measurement
value was
generated not aacquired.
using cluster centerHowever, obtained thereby are problems
the K-mean experienced
clustering method in using the loads measured
if the measurement value
through
was AMIs, such
not acquired. as a smart
However, theremeter, for real-time
are problems operation:
experienced First, the
in using the AMIs, which measure
loads measured through the
load on the low-voltage (LV) side, are only available in limited
AMIs, such as a smart meter, for real-time operation: First, the AMIs, which measure the load on theareas and some countries. Secondly,
real-time operation
low-voltage (LV) side, requires
are only at least a fewinminutes
available limitedofareas voltage
andestimation.
some countries.If the interconnection
Secondly, real-time of the
distributed
operation generation
requires at leastincreases
a few minutesthe network congestion,
of voltage estimation. the period of voltage estimation
If the interconnection will require
of the distributed
shortening and the measurement synchronization is also very important.
generation increases the network congestion, the period of voltage estimation will require shortening However, in general, the
measurement interval of the AMI is longer than 15 min, and
and the measurement synchronization is also very important. However, in general, the measurement the measurement synchronization
between
interval of AMI
the AMI devices is not
is longer than guaranteed
15 min, and [21].
theThirdly,
measurement data measured
synchronizationthrough AMIs AMI
between are managed
devices
through a meter data management system (MDMS), which
is not guaranteed [21]. Thirdly, data measured through AMIs are managed through a meter data is typically configured separately from
the operating
management system
system of the distribution
(MDMS), network.
which is typically Therefore,
configured it is difficult
separately fromtothe load the measured
operating systemdata of
through the AMI into the operating system in real time and use
the distribution network. Therefore, it is difficult to load the measured data through the AMI into thethem in the operating algorithms.
Moreover,
operating it is common
system in real time to utilize
and use thethemload indata
thepatterns
operating stored in the MDMS.
algorithms. Moreover, Load patterns
it is common basedto
on data accumulated by AMI are used for SE. Unfortunately, this
utilize the load data patterns stored in the MDMS. Load patterns based on data accumulated by AMI method has not ensured practicality
as used
are AMI for is not
SE.widely used. The
Unfortunately, thisdata accumulated
method over 2–3practicality
has not ensured years and patterns
as AMI is based on them
not widely used.also
have errors [22–24].
The data accumulated over 2–3 years and patterns based on them also have errors [22–24].

Figure 1. Voltage profile characteristics of concentrated and distributed loads.

Figure 1. Voltage profile characteristics of concentrated and distributed loads.
Energies 2020, 13, x FOR PEER REVIEW 5 of 23
Energies 2020, 13, 2385 5 of 23
One of the studies on WLS-based SE considers the characteristics of a distributed load, and the
load measurement values are considered as pseudo measurements through current measurements
One of the studies on WLS-based SE considers the characteristics of a distributed load, and the
and nominal voltages [13]. The voltage estimation and section load calculation are performed in three
load measurement values are considered as pseudo measurements through current measurements
steps. In step 1, the switches are arranged in a group according to their measurement quality. The
and nominal voltages [13]. The voltage estimation and section load calculation are performed in
total load of sections is calculated by using the nominal voltage, current and phase difference
three steps. In step 1, the switches are arranged in a group according to their measurement quality.
measured at the RTU of DGs and circuit breaker of the main feeder. Thus, the total load calculated is
The total load of sections is calculated by using the nominal voltage, current and phase difference
divided by the number of sections. In step 2, an estimated voltage is obtained by WLS-based SE in
measured at the RTU of DGs and circuit breaker of the main feeder. Thus, the total load calculated is
which the section loads calculated above, and weights based on the measurement quality, are applied.
divided by the number of sections. In step 2, an estimated voltage is obtained by WLS-based SE in
In step 3, the nominal voltage used in step 1 is updated for an estimated voltage. The total load of the
which the section loads calculated above, and weights based on the measurement quality, are applied.
switch group is re-calculated and then divided into individual section loads. This method considers
In step 3, the nominal voltage used in step 1 is updated for an estimated voltage. The total load of the
the distributed loads of sections and the absence of measurement. However, as the calculated load
switch group is re-calculated and then divided into individual section loads. This method considers
based on nominal voltage is used as input of the SE, the voltage estimation depends on the input load.
the distributed loads of sections and the absence of measurement. However, as the calculated load
Moreover, the estimation error is not identified, and the correction is difficult because HPM is not
based on nominal voltage is used as input of the SE, the voltage estimation depends on the input load.
considered. Critique
Moreover, the estimation error is not identified, and the correction is difficult because HPM is not
considered. Critique
2.2.2. Studies on Voltage Estimation Based on Section Load Center
2.2.2.Park
Studies
et al.onproposed
Voltage Estimation Based on Section
a voltage estimation methodLoad Center the distributed load based on the
considering
load Park
centeret [16]. The load acenter
al. proposed voltageis estimation
a virtual point
methodat the load of a the
considering section and is assumed
distributed load basedto on
be
concentrated [25]. As illustrated in Figure 2, the section load center is obtained
the load center [16]. The load center is a virtual point at the load of a section and is assumed to by calculating the
average
be points of[25].
concentrated loads,
As which are distributed
illustrated in Figure 2,in thesections,
section and
load aggregating them in
center is obtained byacalculating
form of a
concentrated
the load. A
average points of section load center
loads, which is calculated
are distributed based onand
in sections, the aggregating
capacity of an equipment
them orof
in a form thea
contract power information of high voltage load. The section load center is obtained
concentrated load. A section load center is calculated based on the capacity of an equipment or the by applying
Equationpower
contract (1) [25].information of high voltage load. The section load center is obtained by applying
Equation (1) [25]. P∑ ( ∗ )
𝐿𝐶 = m (dk ∗ Lk ) , (1)
=∑
LC = kP 1
m , (1)
k=1 Lk
where m denotes the total number of loads between node n and node n + 1, 𝑑 denotes the ratio (0~1)
where m denotes
of the distance the node
from total number
n to kthof loads
load between
among nodedistance
the total n and node n+
from dk denotes
1, node
the thenode
n to the ration(0~1)
+ 1.
of the distance from node n to kth
𝐿 is the magnitude of the kth load. load among the total distance from the node n to the node n + 1. Lk
is the magnitude of the kth load.

Figure 2. Section with distributed load.

Figure 2. Section with distributed load.
This study assumed a fixed load center (FLC) calculated by using only static data like the
This study
equipment assumed
capacity or the acontract
fixed load
powercenter (FLC)
of high calculated
voltage by using
customers. only static
However, as thedata like
actual the
loads
equipment
vary by time, capacity
the loadorcenter
the contract powerasofindicated
also changes, high voltage
by thecustomers. However,
solid line of Figure 3a.asThe
the voltage
actual loads
drop
vary by time, the load center also changes, as indicated by the solid line of
changes when the section load center varies as shown in Figure 3b. Therefore, the estimationFigure 3a. The voltage
error
drop
is changes
increased when
when the
the section
load centerload center is
variation varies as shown inFurthermore,
not considered. Figure 3b. Therefore, the estimation
voltage estimation was
error is increased when the load center variation is not considered. Furthermore,
performed for the main feeder with loads only. However, the distribution network may include DGs voltage estimation
was lateral
and performed for the
feeders. main feeder
Therefore, withof
the effect loads
theseonly. However,
elements needsthe
to distribution
be considered.network may include
DGs and lateral feeders. Therefore, the effect of these elements needs to be considered.
Energies 2020,
Energies 13, x2385
2020, 13, FOR PEER REVIEW 66 of
of 23
23
Energies 2020, 13, x FOR PEER REVIEW 6 of 23

(a) (b)
(a) (b)
Figure 3. Comparison with fixed and corrected load center. (a) Variation of load center by load
Figure 3.3. Comparison
Comparisonwith
with fixed
fixed andand corrected
corrected loadload center.
center. (a) Variation
(a) Variation of load of loadbycenter
center by load
load variation.
variation. (b) Voltage drop.
(b) Voltage(b)
variation. drop.
Voltage drop.

3.
3. Proposed
Proposed Voltage
Voltage Estimation Method
3. Proposed Voltage Estimation Method
As
As illustrated
illustratedin inFigure
Figure4, there
thereare aremultiple measurement devices in ina distribution network. In
As illustrated in Figure 4,4,there are multiple
multiple measurement
measurement devices
devices a distribution
in a distribution network.
network. In
recent
In years,
recent PQPQ
years, meter,
meter, PMUs,
PMUs, and andother
otherHPMs HPMshave havebeenbeenused usedmore
morethan than ever
ever before.
before. The
The zone
recent years, PQ meter, PMUs, and other HPMs have been used more than ever before. The zone
where high-precision voltage measurement at both ends of the main feeder through HPM or CVT of
where high-precision voltage measurement at both ends of the main feeder through HPM or or CVT
CVT of of
the
the substation is defined as the correctable zone. Inside the correctable zone, there are several sections
the substation
substation is defineddefined as the correctable zone. Inside
the correctable Inside the the correctable
correctable zone, zone, there
there are
are several
several sections
sections
divided
divided based on on thethe measurement
measurement units units that
that cannot
cannot measure
measurehigh-precision
high-precisionvoltage, voltage,such suchas asRTUs.
divided based
based on the measurement units that cannot measure high-precision voltage, such as RTUs.
RTUs.
In
In this section, the loads are irregularly distributed. The proposed estimation method consists of three
In this section, the
this section, the loads
loads areare irregularly
irregularly distributed.
distributed. The The proposed
proposed estimation
estimation method method consists
consists of three
of three
steps.
steps. In
In step
step 1,
1, section
section types
types are
are identified
identified according
according to
to whether
whether they
they are
are in
in the
the main
main or
or lateral
lateral feeder
feeder
steps. In step 1, section types are identified according to whether they are in the main or lateral feeder
and
and whether there are only loads or DGs installed. The The section
section types
types areare classified
classified into
into load
load dominant
and whether
whether there are only loads or DGs DGs installed.
installed. The section types are classified into load dominant
dominant
sections
sections and generation dominant sections by using the measurement data, such as current
sections and andgeneration
generation dominant
dominant sections by using
sections by the usingmeasurement
the measurement data, suchdata, as current
such magnitudes
as current
magnitudes
and and voltage–current
voltage–current phase differences phaseatdifferences
both endsat ofboth
eachends of each
section. section.
In step 2, theIn step 2, of the voltage
magnitudes and voltage–current phase differences at both ends of each section. Involtage
step 2, thethe main
voltage
of the main
feeder is feeder isby
estimated estimated
using thebyinitial
usingsection
the initial
load section
center load
and center and measurements
measurements data. If the data. If the
estimated
of the main feeder is estimated by using the initial section load center and measurements data. If the
estimated
values andvalues
voltage and voltage measurements
measurements of the HPMs ofdo
the HPMs do anot match, aparameter
correction parameter
the sectionof the
estimated values and voltage measurements of thenot match,
HPMs do not correction
match, a correctionofparameter ofload
the
section
center load center isThen, calculated. Then,ofthe voltage of iseach sectionagain
is estimated again bythe considering
sectionisloadcalculated.
center is calculated. the voltageThen, the each section
voltage estimated
of each by considering
section is estimated correction
again by considering
the correction
parameter of the parameter
section load of the section
center load center
according according to Ifthe section types. If thereHPMsare
the correction parameter of the section load to the section
center according types.to the there are additional
section types. If there in
are
additional
the HPMs
correctable in the
zone, the correctable
estimation zone, the estimation
accuracy can be accuracy
further can be After
improved. further improved.
that, the Afterofthat,
voltages the
additional HPMs in the correctable zone, the estimation accuracy can be further improved. After that,
the
nodesvoltages of the feeder
in the lateral nodes in the lateral feeder are approximately calculated based on the voltage
the voltages of the nodes are in theapproximately
lateral feeder calculated based on the
are approximately voltage estimation
calculated based on of thethe main
voltage
estimation
feeder. of
In step the main
3, main feeder.
the active andIn step 3, the active and reactive powers at both ends of the section are
estimation of the feeder. In reactive
step 3, the powers
activeatand both ends of
reactive the section
powers at bothare endscalculated by using
of the section are
calculated
voltage by using The
estimation. voltage estimation.
line loss within aThe line isloss
section within abysection
calculated a correctedis calculated
section load by acenter
corrected and
calculated by using voltage estimation. The line loss within a section is calculated by a corrected
section load center
measurement data. and Thus,measurement
the section loads data. are Thus, the section
calculated. loadsare
If there are calculated. If within
there are lateral
section load center and measurement data. Thus, the section loads arelateral feeders
calculated. If there are a section,
lateral
feeders
the within
section loads a section, the
are distributed section loads
to complete are distributed to complete the estimation.
feeders within a section, the section loads arethe estimation.
distributed to complete the estimation.

Figure 4. Division of zone according to installation of high-precision measurements (HPMs).

Figure 4. Division of zone according to installation of high-precision measurements (HPMs).
Figure 4. Division of zone according to installation of high-precision measurements (HPMs).
3.1. Step 1: Identification of Section Types Using Measurement Data
3.1. Step 1: Identification of Section Types Using Measurement Data
3.1. Step 1: Identification
As mentioned of Section
above, Typescorrects
this study Using Measurement Datacenters for voltage estimation. Here,
the section load
As mentioned
the section types above,
need to be this study corrects
distinguished since the
the section
voltage load
drop centers for voltage
characteristics estimation.
are different due Here,
to the
As mentioned above, this study corrects the section load centers for voltage estimation. Here,
the section
change types
of power need to be distinguished since the voltage drop characteristics are different due to
the section types flow
needdirection. In the radialsince
to be distinguished distribution network,
the voltage where no DG exists,
drop characteristics or the DG
are different duehas
to
the change of power flow direction. In the radial distribution network, where no DG exists, or the
the change of power flow direction. In the radial distribution network, where no DG exists, or the
 Energies 2020, 13, x FOR PEER REVIEW 7 of 23
Energies 2020, 13, 2385 7 of 23

DG has a small capacity, the current flow has behavior as illustrated by Figure 5a. On the other hand,
a if the capacity,
small number and capacityflow
the current of DGs are increased,
has behavior the current
as illustrated flow of
by Figure 5a.a Onsection has one
the other hand, of the
if the six
behaviors according to the DG power output and load. If the section
number and capacity of DGs are increased, the current flow of a section has one of the six behaviors load center is corrected without
considering
according to thetheDG type of section,
power output theandcorrection
load. If thewill be wrong.
section This is
load center study classifies
corrected sections
without into two
considering
types: load dominant and generation dominant sections. In this
the type of section, the correction will be wrong. This study classifies sections into two types: study, the division between theload
load
and the generation dominant section is distinguished by
dominant and generation dominant sections. In this study, the division between the load and thethe magnitude of the current flowing
through both
generation ends of
dominant the section
section and the voltage–current
is distinguished by the magnitude phase of difference.
the currentBy takingthrough
flowing cosine on boththe
voltage–current phase difference, the current direction can be obtained.
ends of the section and the voltage–current phase difference. By taking cosine on the voltage–current This direction information
and the
phase current the
difference, magnitude at each end
current direction can ofbe the sectionThis
obtained. can direction
be used to determineand
information if the
thecurrent
currentis
supplied or absorbed in this section. If it is a section that absorbs
magnitude at each end of the section can be used to determine if the current is supplied or absorbed current, it is categorized into inthe
load dominant section and if it is a section that supplies current, it is categorized
this section. If it is a section that absorbs current, it is categorized into the load dominant section and if into the generation
it dominant
is a sectionsection. The load
that supplies dominant
current, section is the
it is categorized intoonethewhere the power
generation dominantconsumption
section. ofTheloads
loadis
dominant section is the one where the power consumption of loads is larger than the power output ofas
larger than the power output of generators. There are three types of load dominant sections,
illustrated There
generators. in Figure 5a–c.types
are three Whenofthe load current
dominant flows in the as
sections, same direction
illustrated in at both5a–c.
Figure ends,Whenand ifthe the
magnitude
current flows of in the current
the same at the sending
direction end is
at both ends, and larger
if thethan the oneofatthe
magnitude thecurrent
receiving end,
at the the section
sending end
is larger than the one at the receiving end, the section becomes a load dominant section (Figuretowards
becomes a load dominant section (Figure 5a,b). If the current flowing at both ends is directed 5a,b).
Ifthe
theinside
current offlowing
a section, at the
bothsection
ends is is directed
a load dominant
towards sectionthe inside (Figure 5c). The the
of a section, generation
section is dominant
a load
dominant section (Figure 5c). The generation dominant section is the one where the power output ofof
section is the one where the power output of generators is larger than the power consumption
loads. As isshown
generators largerin Figure
than 5d–f, there
the power are threeoftypes
consumption loads. of As
generation
shown indominantFigure 5d–f, sections.
there areWhen threethe
current
types flows in the
of generation same direction
dominant sections.atWhenboth ends and if flows
the current the magnitude
in the same ofdirection
current at atthe
bothsending
ends and end
if the magnitude of current at the sending end is less than that at the receiving end, the section is the
is less than that at the receiving end, the section is a generation dominant section (Figure 5d,e). If a
currents of both ends are directed towards the outside of the section,
generation dominant section (Figure 5d,e). If the currents of both ends are directed towards the outside it is a generation dominant
ofsection (Figure
the section, it is5f).
a generation dominant section (Figure 5f).

(a) (b) (c)

(d) (e) (f)

Figure
Figure 5. 5.Currents
Currents behavior
behavior of of load
load dominant
dominant sections
sections (a–c)
(a–c) andand of generation
of generation dominant
dominant sections
sections (d–
(d–f).
f).
3.2. Step 2: Calculation of Correction Parameter of the Section Load Center and Voltage Estimation
3.2.Figure
Step 2:6Calculation
shows the of Correction
flowchart ofParameter of thevoltage
the proposed Sectionestimation.
Load CenterInand
theVoltage Estimation
first step, the initial load
centersFigure
of each sectionthe
6 shows areflowchart
calculatedofby using
the Equation
proposed (1). estimation.
voltage Figure 7 shows anfirst
In the equivalent
step, thecircuit
initialof a
load
distribution line. Vector values are marked with and the other values are scalar values.
centers of each section are calculated by using Equation (1). Figure 7 shows an equivalent circuit of a Therefore,
Esdistribution
is the voltage vector
line. Vectorof the sending
values end, Er with
are marked is theand
voltage vectorvalues
the other of theare
receiving end, and
scalar values. I is the
Therefore,
current
𝐸 is the voltage vector of the sending end, 𝐸 is the voltage vector of the receiving end, and 𝐼 is in
vector flowing in the line. The equation for calculating the voltage magnitude of Er is shown the
Equation (2) [25].flowing in the line. The equation for calculating the voltage magnitude of 𝐸 is shown
current vector
in Equation (2) [25]. Er = Es − I (Rcosθr + Xsinθr ), (2)

where Es is the magnitude of the sending 𝐸 end

= 𝐸voltage, Er is +
− 𝐼(𝑅𝑐𝑜𝑠𝜃 𝑋𝑠𝑖𝑛𝜃
the ),
magnitude of the receiving end voltage,
(2)
I is the magnitude of current, θr is the phase difference between the voltage and current at the receiving
where 𝐸 is the magnitude of the sending end voltage, 𝐸 is the magnitude of the receiving end
end, and R, X are the line impedances. In the network where the load is located in the section as
voltage, 𝐼 is the magnitude of current, 𝜃 is the phase difference between the voltage and current
 Energies 2020, 13, x FOR PEER REVIEW 8 of 23
Energies 2020, 13, 2385 8 of 23

at the receiving end, and 𝑅, 𝑋 are the line impedances. In the network where the load is located in
the section
shown as shown
in Figure in Figure
8, there are two8,types
thereofare two types
currents: theof currents:
current In+1the current
flowing 𝐼 node
from flowing
n to node +1
fromnnode
n tothe
and node n + 1Inand
current − Inthe
+1 current
flowing 𝐼 −
from 𝐼node flowing
n to thefrom
load.node n to
Therefore, the load. Therefore,
considering LC init considering
is the ratio
𝐿𝐶the load
of is the ratioposition
center of the load center0 position
(between and 1) to(between 0 and 1)
the total length to the
of the total
line, the length
voltageofdropthe by
line, the
each
voltagecan
current drop
be by each current
calculated can be calculated as follows.
as follows.

Figure6.6.Flowchart
Figure Flowchartof
ofproposed
proposedvoltage
voltageestimation.
estimation.

𝐸 ==
EDrop1 In+𝐼1 (Rcosθ
(𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃)
+ Xsinθ )
(3)
(3)
EDrop2 = LCinit (In − In+1 )(Rcosθ + Xsinθ),
𝐸 = 𝐿𝐶 (𝐼 − 𝐼 )(𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃),
where EDrop1 is the voltage drop from the current In+1 and EDrop2 is the voltage drop from the current
where 𝐸 is the voltage drop from the current 𝐼 and 𝐸 is the voltage drop from the
In − In+1 . Therefore, the voltage of node n + 1 can be calculated by the following equation [16].
current 𝐼 − 𝐼 . Therefore, the voltage of node n + 1 can be calculated by the following equation
[16]. En+1, est = En − In+1 (Rcosθ + Xsinθ) − LCinit (In − In+1 )(Rcosθ + Xsinθ)
(4)
𝐸 , ==E𝐸n,est −−𝐼 (In(𝑅𝑐𝑜𝑠𝜃
+1 + LC+ init𝑋𝑠𝑖𝑛𝜃)
× (In −−In𝐿𝐶+1 ))((𝐼 )(𝑅𝑐𝑜𝑠𝜃
− 𝐼 + Xsinθ
Rcosθ ), + 𝑋𝑠𝑖𝑛𝜃)
(4)
where En,est is the initial = 𝐸 , − (𝐼voltage
estimated + 𝐿𝐶value × at − 𝐼 nth))(𝑅𝑐𝑜𝑠𝜃
(𝐼 the node, and + 𝑋𝑠𝑖𝑛𝜃),
In is the current measured
the nth
atwhere 𝐸 node. R and
is the X are
initial the linevoltage
estimated resistance
valueand reactance
at the nth node,of aand
section,
𝐼 is the θ is themeasured
andcurrent measuredat
,
voltage–current
the nth node. Rphase are the lineLCresistance
and Xdifference. init is the andinitial load center.
reactance This equation
of a section, and 𝜃 is
is based on the
the measured
assumption that the power flows
voltage–current phase difference. 𝐿𝐶 in one direction
is the initial load center. This equation is based gap
in a load-only distribution network, where the of
on the
the voltage–current phase difference between node n and node n + 1 is small.
assumption that the power flows in one direction in a load-only distribution network, where the gap However, when DGs are
interconnected to the distribution
of the voltage–current system,
phase difference the direction
between node nof thenode
and power n +flow is changed,
1 is small. and when
However, the phase
DGs
difference between node n and node n + 1 may increase by more than 180 ◦ . Therefore, as shown in
are interconnected to the distribution system, the direction of the power flow is changed, and the
Equation (5), the voltage
phase difference between drop
nodeof neach
andline
node should be calculated
n + 1 may increase by based
more onthan
the load
180°.center. In the
Therefore, ascase of
shown
in Equation (5), the voltage drop of each line should be calculated based on the load center. In the
 Energies 2020, 13, 2385 9 of 23
Energies 2020, 13, x FOR PEER REVIEW 9 of 23

Energies
case of 2020,
the 13, xside
FOR PEER REVIEW 9 of 23
the left sideleft
based on based on the load
the section section loadincenter
center Figurein8,Figure 8, thedrop
the voltage voltage drop is calculated
is calculated by using theby
using the impedance
impedance until until the
the section section
load centerload
amongcenter
the among the total impedance
totalinimpedance of the section of the
andsection
the and and
current the
case
currentof the left side
and voltage–currentbased on the
phase section
differenceload center
measured Figure 8, the voltage drop is calculated by
voltage–current
using the impedance phaseuntil
difference
the measured
section load at node
center n. In at
among the node
thecase n.
total
In
thethe
of impedancecase
right of the
side,
of
right side,
the section
the voltage drop
and
theis
the
voltage
calculateddropby issubtracting
calculated the by impedance
subtracting to thethe
impedance
section loadto center
the section
from load
the center
total from the of
impedance total
the
current andofvoltage–current
impedance the section and phase
currentdifference
andphase measured atphase
voltage–current node difference
n. In the case of the at
measured right side,
node n +the
1.
section
voltage and
drop current
is and
calculatedvoltage–current
by subtracting the difference
impedance measured
to the at
sectionnode
loadn + 1.
centerFor these
from reasons,
the total
For
the these
initialreasons,
voltage theestimated
initial voltage
usingisEquation
estimated (5).using Equation (5).
impedance of the is section
𝐸 and =current
𝐸 , − and(𝐼 voltage–current
× 𝐿𝐶 × (𝑅𝑐𝑜𝑠𝜃 phase difference
+ 𝑋𝑠𝑖𝑛𝜃 ) measured at node n + 1.
,
For these reasons, the initial voltage is estimated using Equation (5). (5)
En+1,est =+𝐼 − ((1
En,est × In − LCinit)××(Rcosθ
× 𝐿𝐶 (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 ))
n + Xsinθn )
𝐸 , = 𝐸 , − (𝐼 × 𝐿𝐶 × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 ) (5)
+In+1 × (1 − LCinit ) × (Rcosθn+1 + Xsinθn+1 )) (5)
+𝐼 × (1 − 𝐿𝐶 ) × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 ))

Figure 7. The equivalent circuit of distribution line.

Figure 7. The equivalent circuit of distribution line.
Figure 7. The equivalent circuit of distribution line.

Figure 8. The section with initial load center.

Figure 8. The section with initial load center.
If an HPM exists at the end of a feeder, the error between the estimated voltage using Equation (5)
If anprecise
HPM exists at the end Figure
of the 8. The section
a feeder, withbetween
initial load thecenter.
and the measurement of HPMthe canerrorbe known, and estimated
should be voltage
corrected. using InEquation
addition,
(5)
the section types mentioned above should be considered for this correction. In order to performthe
and the precise measurement of the HPM can be known, and should be corrected. In addition, the
section If an HPM exists at the endshould
of a feeder, the error between the estimated Involtage using Equation
voltagetypes estimation mentioned using above
the correction be considered
of the load center, for this correction.
the load dominant order
section to isperform
aggregated the
(5) and estimation
voltage the precise using measurement of the of HPM can be known, and
loadshould be corrected. isIn addition, the
to a single load as shownthe in correction
Figure 9a and thetheload center,
generation the
dominant dominant
section is section
aggregated aggregated
to a single to
section
agenerator
single loadtypes asmentioned
shown in above
Figure should
9a and be
the considered
generation for this
dominant correction.
section isIn order
aggregated to perform
to a the
single
as shown in Figure 9b. We estimate voltages sequentially from the start point to the end of
voltage estimation
generator as shown using the correction
in necessary
Figure 9b. toWe of the voltages
estimate load center, the load dominant
sequentially from section is aggregated to
the main feeder. It is consider whether the voltage rises or the
drops start atpoint
the (nto+the 1)th endnodeof
a
the single
main load as shown in Figure 9a and the generation dominant section is aggregated to a single
based on feeder.
the nth Itnode is necessary
accordingtotoconsider the change whether
of thethe load voltage
center.rises or drops at the (n + 1)th node
generator
based on as nth
the shown node inaccording
Figure 9b.to We the estimate
change voltages
of the loadsequentially
center. from the start point to the end of
As illustrated in Figure 5, the current flow of a section has one of six behaviors. In the load
the main feeder. It in
As illustrated is necessary to consider whether athe voltage rises or drops at the (n In + 1)th node
dominant section, theFigure current 5, at
the current
both endsflow of the ofsectionsection canhas flow one inof six behaviors.
three behaviors as the load
shown in
based
dominant on the nth
section, node according to the change of the load center.
Figure 5a–c. If the the current
current at both
behaviors at ends of theofsection
both ends the load can flow in section,
dominant three behaviors as shownasinshown Figure in 9a,
Figure As5a–c.illustrated in Figure 5, the current flow of a section has one of six behaviors. shownIn the load
are equal to If the current
Figure 5a, then behaviors
the load at both LC
center endsmoves
L
of the load
to thedominant
right and section,
the largeascurrent Ininflows
Figure to
dominant
9a, are equal section,
to Figure the5a,current
then at both
the load ends ofLCthe moves
center sectiontocan the flow
right in
andthreethe behaviors
large current as shown
I flows in
the longer section. Therefore, the voltage drop increases, so that the voltage of the node n + 1 will be
Figure
to 5a–c.
the longer If the
section. current
Therefore,behaviors at both ends of the load dominant section, as shown in Figure
lowered. In the section that hasthe thevoltage
currentdrop behaviorincreases,
as shown so that in the
Figurevoltage5b, as of the
the load
nodecenter
n + 1 will LCL
9a,
be are
lowered.equal In to Figure
the section5a, then
that the
has load
the center
current LC
behavior moves as to
shown the right
in and
Figure the
5b, large
as the current
load I flows
center LC
moves to the right, the larger current In+1 flows to the shorter section, and the voltage rise decreases.
to the longer
moves to the section.
right, the Therefore,
larger currentthe voltage
I beflows drop toincreases,
theIfshorter sosection,
that theand voltage of the rise
theisvoltage nodedecreases.
n + 1 will
Therefore, the voltage at node n + 1 will lowered. the current behavior like Figure 5c, then as
be lowered.
Therefore,
the sectionthe
In
load
the
voltagesection
centeratLC
that
node has
n + 1to
moves
the
willcurrent
thebe
behavior
lowered.
right,
as shown
If the current
the current
in
In flowsbehavior
Figure
to the longer
5b, as the
is likesection;load 5c, thenLC
center
Figure thereby as
the
moves to the right, the largerL current I flows to the shorter section, and the voltage rise decreases.
the section
voltage of theloadnode center n +LC 1 willmoves to the right,
be lowered. the current
Therefore, in the𝐼 load flows to the longer
dominant section, section;
the voltagethereby of
Therefore,
the voltage the
of voltage
the node at
n +node
1 n +be1 lowered.
will will be lowered.
Therefore, If thein current
the load behavior
dominant issection,
like0 Figure
the 5c, thenof
voltage as
the node n + 1 decreases as the section load center moves to the right ( LCL → LCL ) regardless of the
thenode
the section n +load center LC
1 decreases as the moves
section to load
the right,
centerthe movescurrent to the 𝐼 right
flows( to LCthe → longer section; thereby
LC ) regardless of the
current behaviors.
the voltage
current of the node n + 1 will be lowered. Therefore, in the load dominant section, the voltage of
behaviors.
In the generation dominant sections, the current at both ends of the section can flow in three
the node
In thenas + 1 decreases
generation as the section loadthe center moves to the right ( the LC section
→ LC can ) regardless of the
behaviors shown indominant Figure 5d–f. sections,
If the current current at both
behaviors atends
both of ends of the generator flowdominant
in three
current behaviors.
behaviors
section, asas shown
shown in in Figure
Figure 9b,5d–f.
are equalIf thetocurrent
the onebehaviors
shown in Figure at both5d,e, endsthenof the thegenerator
magnitude dominant
of In and
In astheshowngeneration dominant sections, the current at both ends of thethen section can flow in of three
In+1 are opposite to those of Figure 5a,b. For this reason, as the load center moves right, the voltageIof
section, in Figure 9b, are equal to the one shown in Figure 5d,e, the magnitude
behaviors
and I as shown
are opposite in Figure
to those 5d–f. If
of Figure the current behaviors at both ends of the generator dominant
the node n+1 will be raised. In the current5a,b. For this
behavior, reason,inasFigure
as shown the load 5f, as center movesload
the section right, the
center
section,
voltage as shown
of thetonode in Figure
n+1 will 9b, are equal to the one shown in Figure 5d,e, then the magnitude of I
LCDG moves the right, the be raised.
current In+In1 flows
the current
to the behavior,
longer section; as shownthereby in Figure
the voltage 5f, asofthethesection
node n
and I
load center LC are opposite to those of Figure
moves to the right, the current I 5a,b. For this reason, as the load center
flows to the longer section; thereby the voltage moves right, the
voltage of the node n+1 will be raised. In the current behavior, as shown in Figure 5f, as the section
load center LC moves to the right, the current I flows to the longer section; thereby the voltage
 Energies 2020, 13, 2385 10 of 23
Energies 2020, 13, x FOR PEER REVIEW 10 of 23

of+the
1 isnode n +Therefore,
raised. 1 is raised.inTherefore, in thedominant
the generation generation dominant
section, section,
as shown in as shown
Figure 9b,in Figure
the 9b,of
voltage the
the
voltage of the node n + 1 increases as the section load center moves to the right (0 LC → LC ).
node n + 1 increases as the section load center moves to the right ( LCDG → LCDG ).
This
Thischaracteristic
characteristicofofvoltage
voltagevariation
variationshould
shouldbebe
considered
consideredto to
correct thethe
correct section load
section center.
load center.
Moreover, since the section load center needs to be located between both ends of the
Moreover, since the section load center needs to be located between both ends of the section, the section, the load
load
center should be corrected within the section.
center should be corrected within the section.

(a) (b)
Figure
Figure9. 9.Voltage variation
Voltage characteristics
variation byby
characteristics load center
load variation
center of of
variation (a)(a)
load and
load and(b)(b)generator
generator
dominant
dominant section.
section.

InInthis
thispaper,
paper,thethecorrection
correctionparameter
parameterofofthe thesection
sectionload loadcenter definedasas𝛼αLCwhich
centeris isdefined whichis is
calculated
calculated totoeliminate
eliminate thetheerror
errorbetween
betweenthe thevoltage
voltageestimation
estimation obtained
obtained bybyEquation
Equation (5)(5)and
andthe
the
voltagemeasurement
voltage measurementatatthe thenode
nodewherewherean anHPMHPMisisinstalled.
installed. In Inorder
order to find α𝛼LC , ,Equation
to find Equation(6) (6)can
canbe
beobtained
obtainedby substitutingLC
bysubstituting 𝐿𝐶init →→ LC𝐿𝐶init + + αLC 𝛼 of of
Equation
Equation (5)(5)forforthethe
load dominant
load dominant section.
section. Since
Sincethe
voltage
the voltagedrop according
drop according to theto movement
the movement of loadof center between
load center the loadthe
between andload generation dominant
and generation
sections is
dominant the opposite,
sections the voltage
is the opposite, theestimation of the generation
voltage estimation dominant dominant
of the generation section cansectionbe obtained
can beas
shown in Equation (7).
obtained as shown in Equation (7).
𝐸 , = 𝐸 , − (𝐼 × (𝐿𝐶 + 𝛼 ) × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 )
E0n+1,est = E0n,est − (In × (LCint + αLC ) × (Rcosθn + Xsinθn ) (6)(6)
+𝐼 (1 − 𝛼 ) × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 ))
+In×+1 ×−(1𝐿𝐶 − LCint − αLC ) × (Rcosθn+1 + Xsinθn+1 ))
𝐸 , = 𝐸 , − (𝐼 × (𝐿𝐶 − 𝛼 ) × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 )
E0n+1,est = E0n,est − (In × (LCint − αLC ) × (Rcosθn + Xsinθn ) (7)
+𝐼 × (1 − 𝐿𝐶 + 𝛼 ) × (𝑅𝑐𝑜𝑠𝜃 + 𝑋𝑠𝑖𝑛𝜃 )) , (7)
+In+1 × (1 − LCint + αLC ) × (Rcosθn+1 + Xsinθn+1 )),
where 𝐸 ,0 is the estimated voltage value considering 𝛼 for the node n. The voltage at the node
where E is the estimated voltage value considering αLC for the node n. The voltage at the node m
m with ann,est
HPM is obtained by subtracting all the voltage drops until the end of the line in the section
with an HPM is obtained by subtracting all the voltage drops until the end of the line in the section
from the voltage at the start point of the feeder. Equation (8) is used to calculate 𝛼 , which equates
from the voltage at the start point of the feeder. Equation (8) is used to calculate αLC , which equates the
the estimated voltage value0 𝐸 , to the measurement 𝐸 , in node m.
estimated voltage value Em,est to the measurement E0m,meas in node m.
𝐸 , =𝐸 ,
= 𝐸, −∑ ∈ (𝐼 × 𝐿𝐶 ,
+𝛼 ×𝛽 +𝐼 × 1 − 𝐿𝐶 ,
−𝛼 ×β
E0m,meas = E0m,est
+I × (1 − 𝐿𝐶 P , −𝛼  ) × β ) − ∑ ∈ ( 𝐼 × (𝐿𝐶 , −𝛼 ) 
= E
×𝛽 +𝐼 1,meas − × (1 − LC( In × LC + α
+n,n𝛼+1 ) × 𝛽 )
init LC × β n + In + 1 × 1 − LC init n,n+1
− α LC × βn
ST∈Load ,
=𝐸, − ∑( + 𝐿𝐶In+1 ,× 1×−(𝐼 × 𝛽 −−𝐼 αLC× × 𝛽 βn+)1+ ) −𝐼 × 𝛽(In ×) LCinit
   
− αLC
P
LCinit n,n+1 n,n+1 (8)
ST∈DG
−𝛼× β× (∑In+∈1 × 1(𝐼− × LC𝛽init− 𝐼 +× 𝛽 ×)β− n+∑ 1 ) ∈ (𝐼 × 𝛽 − 𝐼 × 𝛽 ))
 
n+ αLC
n,n+1 (8)
initn,n+1 × (In × βn − In+1 × βn+1 ) + In+1 × βn+1 )
P
=𝐸, − 𝐸= ,E1,meas − (LCP
− αLC × ( (In × βn − In+1 × βn+1 ) − (In × βn − In+1 × βn+1 ))
P
−𝛼 × (∑ ST∈Load ∈ (𝐼 × 𝛽 − 𝐼 × 𝛽 ) − ST∈DG ∑ ∈ (𝐼 × 𝛽 − 𝐼 × 𝛽 )),
= E1,meas − Em,est
where 𝐿𝐶 is the initial section P n and n + 1. 𝛽 is the 𝑅𝑐𝑜𝑠𝜃 +
, − αLC × ( (In ×load center between node
βn − I n + 1 × βn + 1 ) − (In × βn − In+1 × βn+1 )),
P
𝑋𝑠𝑖𝑛𝜃 of the node n, and ST denotes ST∈Load each section type. In Equation ST∈DG (8), the constant term except 𝛼
on the right side is equal to the initial voltage estimate of the node m. Accordingly, if all the values
where LCinitn,n+1 is the initial section load center between node n and n + 1. βn is the Rcosθn + Xsinθn
except 𝛼 are put from the right side to the left side, an estimation error is obtained, as expressed
of the node n, and ST denotes each section type. In Equation (8), the constant term except αLC on the
in Equation (9). If Equation (9) is arranged with respect to α , the section load center can be
right side is equal to the initial voltage estimate of the node m. Accordingly, if all the values except αLC
expressed by using Equation (10).
are put from the right side to the left side, an estimation error is obtained, as expressed in Equation (9).
𝐸 , − 𝐸 , = 𝛼 × (∑ ∈ (𝐼 × 𝛽 − 𝐼 ×𝛽 )−∑ ∈ (𝐼 × 𝛽 − 𝐼 × 𝛽 )) (9)
Energies 2020, 13, 2385 11 of 23

If Equation (9) is arranged with respect to αLC , the section load center can be expressed by using
Equation (10).
X X
E0m,meas − Em,est = αLC × ( ( I n × βn − I n + 1 × βn + 1 ) − (In × βn − In+1 × βn+1 )) (9)
ST∈DG ST∈Load

E0m,meas − Em,est
αLC = P (10)
ST∈DG (In × βn − In+1 × βn+1 ) − ST∈Load (In × βn − In+1 × βn+1 )
P

When αLC obtained by using Equation (10) is substituted in Equations (6) and (7), the voltage
estimates of each node can be calculated with the section load center being corrected. The voltage for
the lateral feeder can be estimated if it is interconnected to the main feeder with RTUs. The voltage
of the lateral feeder is estimated by using the estimated voltage value at the node before the lateral
feeder branches off from the main feeder. In case the HPM is installed at the end of a lateral feeder,
the voltage estimation with the section load center corrected by the αLC , is conducted by using
Equations (5)−(10). In case there is no HPM, the voltage estimation with fixed load center is performed
by using Equation (5).

3.3. Step 3: Section Load and Loss Calculation

After the voltages at both ends of each section are estimated, the section load is calculated by using
the voltage estimation. For the calculation of the section load, the active and reactive powers flowing
in each node are calculated based on the voltage estimation, current magnitude, and voltage–current
phase differences. Then, the losses caused by currents flowing in each line are calculated by using
Equation (11). The conventional method sets up an equation under the assumption that the section
load center is located at the middle of the section including only loads. For this reason, the line loss
was calculated by using a simple formula Ploss = 3 × I2 × R, which includes the average value of
currents measured at both ends of the section [15]. However, the section load center moves in real time,
and the current may flow bidirectionally due to the DGs. Therefore, this study proposes a method of
calculating line loss at both ends by considering the section load center, as expressed in Equation (11).

Plossn,n+1 = 3 × (In )2 × LCnewn,n+1 × Rn,n+1 + (In+1 )2 × (1 − LCnewn,n+1 ) × Rn,n+1

 
(11)
Qlossn,n+1 = 3 × (In )2 × LCnewn,n+1 × Xn,n+1 + (In+1 )2 × (1 − LCnewn,n+1 ) × Xn,n+1 ,
 

where Plossn,n+1 , Qlossn,n+1 are the loss of active and reactive powers between the nth and (n + 1)th nodes.
LCnewn,n+1 is the section load center between the nth and (n + 1)th nodes, which is corrected by αLC .
The section load considering the losses can be calculated as in Equation (12).

Ploadn,n+1 = P(n) − P(n + 1) − Plossn,n+1

(12)
Qloadn,n+1 = Q(n) − Q(n + 1) − Qlossn,n+1

In the case of the generation dominant sections, the active and reactive powers flowing in each
node are calculated with a negative sign due to the voltage–current phase difference. For this reason,
the power output of each section is calculated as negative.
As shown in Figure 10, the loads are distributed in a section with a lateral feeder where the RTU
is installed. As the active and reactive powers flow in each node, the section load center and section
loads are already known through the analysis of a straight line, and only the position and magnitude of
each load need to be calculated. A straight line section is divided into two sections, and the load center
of the straight line section is applied at the load centers of the divided sections. Then, the magnitudes
of the two loads are calculated by using Equations (13) and (14) to ensure that the values are the same
as the load centers of the straight section, which is not divided before.

l1 × LC × L1 + (l1 + l2 × LC) × L2
= LC, (13)
Ltot
by 𝛼 . The section load considering the losses can be calculated as in Equation (12).
𝑃 ,
= 𝑃(𝑛) − 𝑃(𝑛 + 1) − 𝑃 ,
(12)
𝑄 ,
= 𝑄(𝑛) − 𝑄(𝑛 + 1) − 𝑄 ,
In the case of the generation dominant sections, the active and reactive powers flowing in each
node are2020,
Energies calculated
13, 2385 with a negative sign due to the voltage–current phase difference. For this reason, 12 of 23
the power output of each section is calculated as negative.
As shown in Figure 10, the loads are distributed in a section with a lateral feeder where the RTU
is installed. As the active and reactive powers + L2 in
L1 flow =L tot , node, the section load center and section
each (14)
loads arel already known through the analysis of a straight line, and only the position and magnitude
where, 1 and l2 denotes the ratio (0~1) of the distance from node n and node n + 1 to lateral feeder
ofamong
each load need to be calculated.
the total distance A straight
from the node linenode
n to the section
n + is1, divided into two
respectively. sections,
As all and except
the values, the loadL1
center of the straight line section is applied at the load centers of the divided sections.
and L2 , in Equations (13) and (14) are known, L1 and L2 can be obtained by solving the simultaneous Then, the
magnitudes
equations. of the two loads are calculated by using Equations (13) and (14) to ensure that the values
are the same as the load centers of the straight section, which is not divided before.

Figure 10. Load distribution in the section where lateral feeder exists.
Figure 10. Load distribution in the section where lateral feeder exists.
4. Case Study
Case studies were performed by test systems. Three phase radial distribution networks were
implemented. ACSR of 160 mm2 with the impedance of (0.1823 + j.0.3901) Ω/km was used for the line
of 22.9 kV distribution network. The number of RTUs installed in the topology of the test model was
determined according to the Korean switch installation regulations [26].
A total of four case studies were carried out. For each case study, selective comparisons were
performed between the estimation methods that use a fixed load center (FLC method [16]), a method
using the state estimation (IRWLS method [14]), and the proposed method in this study. Since both the
proposed method and the FLC method use the measured values at the switch, the current through
the switch by assumed load and PV were extracted using Simulink, and the voltage and load were
estimated using a program implemented in MATLAB. To conduct a simulation using the IRWLS
method, K-mean clustering was implemented in MATLAB, and the IRWLS used a Pandapower tool
(a Python-based tool) [27]. The estimation was conducted by using the voltage and active/reactive
power of the substation measured using Simulink and the estimated load and PV output values
obtained by using K-mean clustering as input values of the IRWLS. In order to perform state estimation
using the IRWLS of Pandapower, a standard deviation of the measurement error is required. For the
voltage and active/reactive power of the substation, a standard deviation is assumed 10−8 and the
standard deviation of load and PV used are 0.02 and 0.5, respectively.
In case study 1, voltage estimation by three methods was compared for a simple distribution
system. To conduct the comparison, the actual load measurement data presented elsewhere [28] were
used as load values. As the training data for K-mean clustering of the IRWLS method, the data from
29 June to 3 July were used. For the test data, load data from 6 July were used. For the PV power
generation, an actual output data of 250 kW PV (on Building No. 5 of College of Engineering at
Chonnam National University, South Korea) from 29 July to 2 August were used as training data,
and the data from 5 August were used as the test data. The PV output measurement data used for
training and the test data are shown in Figure 11. Case study 2 analyzed how the estimated error of
the previous node affects the voltage estimation of the next node for the linear section network of
case study 1. The simulation of the estimated error accumulation of the conventional method and the
proposed method was performed. In case study 3, the estimated error of the proposed method was
analyzed according to the number of HPM installations for networks with multiple lateral feeders
and PVs similar to the actual distribution network. In case study 4, the network of case study 3 was
simulated for the measurement noises that could occur in actual network operation. The simulation
was performed by applying the load and PV profile used in case study 1. In addition, to analyze the
estimation errors of the proposed method and the existing methods, according to measurement errors,
the estimation results of the proposed method and the FLC method were derived by applying 1–5%
of the normally distributed noise to the current measurements. The estimation result of the IRWLS
Energies 2020, 13, 2385 13 of 23

method was derived by applying 1–2% of the normally distributed noise to the load measurements.
Energies 2020, 13, x FOR PEER REVIEW 13 of 23
Finally, the estimation errors of each method were compared with each other.

Figure 11. Output measurement data of actual PV.

Figure 11. Output measurement data of actual PV.
4.1. Case 1: Simple Radial Distribution System with One DG
4.1. Case 1: Simple Radial Distribution System with One DG
To compare the estimation results of the conventional and proposed method, a simulation was
performedTo compare
for the test thesystem
estimation resultsinofFigure
illustrated the conventional
12. It consists and ofproposed method,
five sections a simulation
all in the main feeder was
performed for the test system illustrated in Figure 12. It consists
without lateral, and it was assumed that only one DG is installed in the feeder. The input data are as of five sections all in the main feeder
without
shown lateral,
in Table 2. and
An HPM it waswas assumed
installedthatatonly one of
the end DG theis line,
installed in the were
and RTUs feeder. The input
installed datanode.
at each are as
The length of each section was 4 km and the impedance of the section was 4 × (0.1823 + j.0.3901 Ω/km).
shown in Table 2. An HPM was installed at the end of the line, and RTUs were installed at each node.
It The
waslength
assumed of each thatsection was 4 km
each section was and the impedance
composed of three of distributed
the section was loads4 ×and/or
(0.1823DGs + j.0.3901
where Ω/km).
the
It wasbetween
length assumed thatand
loads each thesection
magnitudewas composed
of loads and of three
DGs were distributed loads
arbitrarily and/orThe
chosen. DGs where the
simulation
waslength between
carried out byloads applying and thethepreviously
magnitudementioned of loads and DGs
actual were arbitrarily
measured load andchosen. The simulation
the PV output profile
towas
eachcarried
load and outPV. by applying the previously mentioned actual measured load and the PV output
profile
The to each loaderrors
estimation and PV. of each method for 24 h are listed in Table 3. The proposed method was
shown the smallest estimationoferror
The estimation errors eachinmethod
both voltagefor 24andh areloadlisted in Table 3.
estimations. It The proposed
obtained method
an average was
error
ofshown
0.0088% the smallest
and a maximum estimationerrorerror in bothfor
of 0.0457% voltage and load
the voltage estimations.
estimation and anIt average
obtainederror an average error
of 0.1258%
of 0.0088% and a maximum error of 0.0457% for the voltage estimation
and a maximum error of 1.1968% for the load estimation. For the result of voltage estimation, the FLC and an average error of 0.1258%
and a maximum
method had the lowest error of 1.1968%for
accuracy forthethevoltage
load estimation.
estimation For the an
with result of voltage
average error estimation,
of 0.1037% the andFLCa
method had
maximum error theof lowest
0.3228%. accuracy
The true forandtheestimated
voltage estimation
voltage value withofantheaverage error ofand
FLC, IRWLS, 0.1037% and a
proposed
maximum
methods error4ofare
at node 0.3228%.
shown The true and
in Figure 13. estimated
The resultsvoltage value of the
of the proposed FLC, IRWLS,
method and IRWLS and method
proposed
methods
were similar at to
nodethe 4trueare value,
shownbut in Figure
the FLC 13.method
The results
had aofsignificantly
the proposed method
bigger and IRWLS
estimation errormethod
than
thewere similar
other methods.to theThe true value,method
IRWLS but the couldFLC methodestimatehad more a significantly
accurately usingbiggermeasurement
estimation error datathan
of
individual loads, and the proposed method enabled accurate estimation through section load correctionof
the other methods. The IRWLS method could estimate more accurately using measurement data
individual
using the HPM. loads,
However,and the FLC proposed
methodmethod showedenabled accurate error
a large estimation estimation
compared through
to othersection
methods load
because there was no measurement of individual loads or correction of estimation errors. In case of theto
correction using the HPM. However, the FLC method showed a large estimation error compared
other
load methods the
estimation, because
IRWLS there
method was hadno measurement
the lowest accuracy of individual loadsestimation
for the load or correction withof anestimation
average
errors.
error In case and
of 6.1981% of the load estimation,
a maximum error of the 193.8%.IRWLS When method
a loadhad the lowest
estimation erroraccuracy
of 193.8% for the load
occurred
estimation with an average error of 6.1981% and a maximum error
at 17:00, the actual load of Section 3 was 65.6 kW, while the IRWLS method estimated the load to be of 193.8%. When a load estimation
error
192.8 of The
kW. 193.8% occurred
absolute at 17:00,ofthe
magnitude erroractual
was load
smallofatSection 3 was 65.6
approximately kW,
127.2 while
kW, but the
the IRWLS
error rate method
was
estimated the
significantly load
large atto be 192.8
193.8%. InkW. The absolute
the process of themagnitude of error was
K-mean clustering, small
which at approximately
generates load and 127.2
PV
kW, but the error rate was significantly large at 193.8%. In the process
input data of IRWLS, the deviation of the input value is large because of the variation value of the PV of the K-mean clustering, which
generates
output loadthan
is larger and thatPV input
of thedataload.ofResults
IRWLS,ofthe loaddeviation
estimation of the
for input value
Sections 3 andis large because
4 by each of the
method
variation value of the PV output is larger than that of the load.
are shown in Figure 14. As mentioned above, Section 3, where PV is interconnected, is the generation Results of load estimation for Sections
3 and 4 section
dominant by each withmethod
a largeare PV,shown
and theinIRWLS Figure 14. Ashad
method mentioned
the lowestabove, Section
estimation 3, where
accuracy thanPVthe is
interconnected, is the generation dominant section with a large PV, and the IRWLS method had the
lowest estimation accuracy than the other methods owing to the change in PV output. However, in
Section 4, where only the loads existed, all three methods had similar accuracy.
Energies 2020, 13, 2385 14 of 23

other methods owing to the change in PV output. However, in Section 4, where only the loads existed,
allEnergies 2020, 13, x FOR
three methods hadPEER REVIEW
similar accuracy. 14 of 23

Figure 12. Simple distribution system with PV.

Figure 12. Simple distribution system with PV.
Table 2. Input and measurement data of case study 1.
Table 2. Input and measurement data of case study 1.
Line Section Load
From Initial
Section
To Node Line Section Load Reactive
Active
No. Node Length Load
Section From To No. No.ActivePower ReactivePower Initial Load
Center
(km)
Length
No. Node Node No. No. Power (kW) (kVar)
Power Center
(km)
1 1.177 (kW) (kVar)
1 956.3 463.1
1 1 2 1 2 1.1770.459 2 141.8 68.7 0.631
3 1.351 1 956.3 463.1
2 0.459 3 476.8 230.9
1 1 2 4 1.013 2 141.8 68.7 0.631
3 1.351
1 1.154 3 476.8 230.9
4 2 1.0131.410 1 778.0 376.8
2 2 3 1 3 1.1540.356 2 770.1 373.0 0.735
1 3 778.0 729.2 376.8353.1
2 4 1.4101.08
2 2 3 2 770.1 373.0 0.735
3 1 0.3561.381 1 228.3 110.5
3 729.2 353.1
3 3 4 4 2 1.080.772 2 817.7 396.0
0.795
1 3 1.3811.233 3 −29,709.2 140.8
1 228.3 110.5
4 0.614 (PV)
2 0.772 2 817.7 396.0
3 3 4 1 1.684 0.795
3 1.233 3 1−29709.2356.4 140.8172.6
2 0.775
4 4 5 4 3 0.6141.162 2 (PV) 219.1 106.1 0.951
3 628.6 304.4
1 4 1.6840.379
1 356.4 172.6
2 1 0.7750.589
4 4 5 2 1 219.1 701.9 106.1339.9 0.951
3 2 1.1621.053
5 5 6 3 2 628.6 756.9 304.4366.6 0.431
4 3 0.3790.784 3 147.5 71.4
4 1.574
1 0.589
1 701.9 339.9
2 1.053
Table53. Comparison
5 of6 voltage and load estimation2accuracy
756.9
of fixed load366.6 0.431
center (FLC) method,
3 0.784
3 proposed
iteratively re-weighted least square (IRWLS) method, and 147.5method. 71.4
4 1.574
Voltage Estimation Error Load Estimation Error
Method
Table 3. Comparison
Average of voltage
Maximum and load estimation
Standard accuracy
Average fixed load
of center (FLC)
Maximum method,
Standard
iteratively re-weighted
(%) least square
(%) (IRWLS) method, and proposed
Deviation (%) method. (%) Deviation
FLC Voltage Estimation Error
0.1037 0.3228 0.0801 0.1800 Load Estimation
1.6618 Error
0.2303
Method
Method Average Maximum Standard Average Maximum Standard
IRWLS (%) (%) Deviation (%) (%) Deviation
0.0153 0.0464 0.0101 6.1981 193.85 19.781
Method
FLC
Proposed 0.1037 0.3228 0.0801 0.1800 1.6618 0.2303
Method 0.0088 0.0457 0.0096 0.1258 1.1968 0.2308
Method
IRWLS
0.0153 0.0464 0.0101 6.1981 193.85 19.781
Method
Proposed
0.0088 0.0457 0.0096 0.1258 1.1968 0.2308
Method
Energies 2020, 13, x FOR PEER REVIEW 15 of 23

Energies 2020,
Energies 13,13,
2020, x FOR
2385 PEER REVIEW 1515
of of
23 23

Figure 13. Comparison of true value and voltage estimation value of FLC method, IRWLS method,
Figure 13. Comparison of true value and voltage estimation value of FLC method, IRWLS method,
and proposed
Figure method at node
13. Comparison 4. value and voltage estimation value of FLC method, IRWLS method,
of true
and proposed method at node 4.
and proposed method at node 4.

(a) (b)

Figure 14. Comparison

Figure 14. Comparison (a)
of true value
of true andand
value loadload
estimation values
estimation of of
values FLC method,
FLC (b)
IRWLS
method, IRWLSmethod,
method,and
and
proposed
proposed method
method in in
(a) (a) Section
Section 3 3
and and
(b) (b) Section
Section 4. 4.
Figure 14. Comparison of true value and load estimation values of FLC method, IRWLS method, and
proposed
4.2. methodofinEstimation
Case 2: Impact (a) SectionError
3 andAccumulation
(b) Section 4.
4.2. Case 2: Impact of Estimation Error Accumulation
4.2.The
Case The analysisofof
2: Impact which node’s
Estimation estimation error has the cumulative effect of the error in calculating
analysis of which node’sError Accumulation
estimation error has the cumulative effect of the error in calculating
the next node’s estimate is an important factor to verify the practical applicability of the proposed
the nextThe node’s estimate
analysis is an important factorerror
to verify thecumulative
practical applicability of theinproposed
method. To thisof which
end, node’s
we added estimation
an error to node 1has the
of the test network effectin caseof1the
anderror
analyzed calculating
how it
method. To
theaffects this
next node’s end, we
estimateadded
is an an error
important to node
factor 1 of
to the test
verify network
the in
practical case 1 and
applicability
the voltage estimation of other nodes. Figure 15 shows the simulation results of the existing
analyzed
of the how
proposed it
affects the voltage
method.
method Toand estimation
thisthe
end, we added
proposed of other
method nodes.
an error
for Figure
topropagation
the node 15 the
1 of shows the the
of test simulation
network
cumulative caseresults
inerror. 1As
and of theofexisting
analyzed
a result how it
adding
method
affects and the
the voltage
the estimation proposed method
estimation
error for
of other
of 0.1–0.5%, the propagation
nodes. Figure
the estimation of
error15 the cumulative
of shows error.
the simulation
the previous As a
node was resultsresultof of
accumulated theadding
existing
as it is,
themethod
estimation
and theand error
estimation of 0.1%–0.5%,
the proposed
error ofmethod the
the nextfor estimation
nodethetended error
propagation of the
of the
to increase previous
when node
cumulative
the existingwas
error.accumulated
methodAs awasresult
used. as it
of In is,
adding
the
and the
thecase estimation error
of the proposed
estimation of the next
method, when
error of 0.1%–0.5%, node
thethe tended to
estimation
estimation increase
error
error of of when
thethe the
previous
previous existing method
nodewas
node occurs, was used.as In
the estimation
accumulated it is,
theand
case
errorofof the
the proposed
next node method,
eventuates. when
However, the estimation
it can be error
observed of
that the
there
the estimation error of the next node tended to increase when the existing method was used. In previous
was no node occurs,
accumulation of the
the
estimation
theeffect error
of theofproposed
casetowards theend
the next node
of the eventuates.
feeder.
method, Thesethe
when However,
results it error
were obtained
estimation can be observed
because
of thatnode
thereerror
the estimation
the previous was was
occurs, nothe
accumulation
corrected of the
through effect
the towards
HPM. the end of the feeder. These
estimation error of the next node eventuates. However, it can be observed that there was noresults were obtained because the
estimation
accumulationerror was
of thecorrected throughthe
effect towards theend
HPM. of the feeder. These results were obtained because the
estimation error was corrected through the HPM.
Energies 2020, 13, x FOR PEER REVIEW 16 of 23

Energies 2020,
Energies 2020,13,
13,2385
x FOR PEER REVIEW 16 of 23
16 of 23

(a) (b)
Figure 15. Comparison of voltage estimation error accumulation of (a) FLC method and (b) proposed
method. (a) (b)
Figure 15.
Figure 15. Comparison of voltage
Comparison estimation
of voltage errorerror
estimation accumulation of (a) FLC
accumulation method
of (a) and (b) proposed
FLC method and
4.3. Case 3: Radial Distribution System with Multiple DGs and Lateral Feeder
method.
(b) proposed method.
The test system of Figure 16 was used to evaluate the effects of multiple DGs and lateral feeder
4.3. Case 3: Radial Distribution System with Multiple DGs and Lateral Feeder
4.3.
both inCase
the3:proposed
Radial Distribution
method and System
in thewith Multiple DGs
conventional and Lateral
method. The testFeeder
system model is based on an
actual The test system
Korean of Figure
distribution 16 was[26]
network usedand
to evaluate
its the effects
parameters
The test system of Figure 16 was used to evaluate the effects of multiple areofasmultiple
shown DGs
in DGs and lateral
Table 4. At
and
feeder
first,feeder
lateral the
both
voltage in the proposed method and in the conventional method. The test system model is based on an
both inestimation
the proposed errors wereand
method compared between themethod.
in the conventional conventional
The testmethod
system (no model HPM) and on
is based thean
actual Korean
proposed distribution network [26] and its parameters are as shown in Table 4.toAt first, the voltage
actual Korean distribution network [26] and its parameters are as shown in Table 4. At first, of
method when one HPM was installed at node 6 (HPM #1). In order analyze the effect the
estimation
the number errors
of HPMs wereoncompared
the between
estimation the conventional
accuracy, the voltagemethod (no HPM)
estimation errors and
werethe observed
proposed by
voltage estimation errors were compared between the conventional method (no HPM) and the
method when one HPM was installed at node 6 (HPM #1). In order to analyze the effect of the number
increasing
proposed the HPMswhen
method installed
one at nodes
HPM was 12 (HPM #2), 9 (HPM #3), and
#1). 14 (HPM #4). Figurethe 17 effect
showsof
of HPMs on the estimation accuracy, the installed at node
voltage estimation 6 errors
(HPM were In order to
observed analyze
by increasing the
each
the node voltage
number of HPMsmagnitude
on the through the
estimation conventional
accuracy, the method,
voltage proposed
estimation errors method,
were and theby
observed
HPMs installed at nodes 12 (HPM #2), 9 (HPM #3), and 14 (HPM #4). Figure 17 shows each node voltage
measurements.
increasing Compared to theatconventional method, the proposed method estimates the 17voltage
magnitude the HPMs
through theinstalled
conventional nodes 12 (HPM
method, #2), 9method,
proposed (HPM #3),
andand 14 (HPM
the measurements. #4). Figure
Compared shows
more
each accurately. Figure 18 presents the maximum voltage estimation error, average voltage
to the node voltagemethod,
conventional magnitude throughmethod
the proposed the conventional
estimates themethod,
voltage more proposed
accurately.method,
Figureand18 the
estimation
measurements.error, Compared
and standard to deviation
the of errors
conventional obtained
method, theby the proposed
proposed method method withthe
estimates different
voltage
presents the maximum voltage estimation error, average voltage estimation error, and standard
HPMs.
more When
deviation of one
accurately. HPM
errors was 18
Figure
obtained installed
bypresents at node
the proposedthe 6,maximum
the average
method with voltage
voltage
different estimation
estimation
HPMs. When error was
error,one significantly
average
HPM voltage
was
reduced.
estimation When two
error,6,and
installed at node HPMs
the standard were installed
deviation
average voltage at nodes
of errors
estimation 6 and
obtained
error 12, the average
by the proposed
was significantly and
reduced.method maximum
When two with voltage
different
HPMs
estimation
HPMs.
were errors
When
installed atremarkably
one HPM6was
nodes anddecreased.
installed at When
12, the averagenodeand 6,two
the other
average
maximum HPMs wereestimation
voltage
voltage installed additionally
estimation error
errorswas at nodes
significantly
remarkably
9 reduced.
and 14, the
decreased. average
When
When two
two and
HPMsmaximum
other HPMs voltage
were
were installed estimation
installed
at nodes 6errors
additionally
and 12,decreased
at nodes 9 slightly.
the average and 14,andthe average and
maximum voltage
maximum voltage
estimation estimation errors
errors remarkably decreased
decreased. Whenslightly.
two other HPMs were installed additionally at nodes
9 and 14, the average and maximum voltage estimation errors decreased slightly.

Figure 16. Radial distribution system with PVs and lateral feeders.

Figure 16. Radial distribution system with PVs and lateral feeders.

Table 4. Input data of case study 2.

Figure 16. Radial distribution system with PVs and lateral feeders.
Line Section Load
Table 4. Input data of case study 2.Reactive Initial Changed
Section From To Active
Length Load Load
No. Node Node No. Line No. Power Load
Section Power
(km) Center Center
Initial Changed
Section From To (kW)
Active (kVar)
Reactive
1No. 1 2 1No. Length
0.123 1No. 66.7 25.1 Load Load
Node Node Power Power 0.368 0.152
(km) Center Center
(kW) (kVar)
1 1 2 1 0.123 1 66.7 25.1 0.368 0.152
Energies 2020, 13, 2385 17 of 23

Table 4. Input data of case study 2.

Line Section Load

Initial Changed
Section From To Active Reactive
Length Load Load
No. Node Node No. No. Power Power
(km) Center Center
(kW) (kVar)
1 0.123
1 66.7 25.1
2 0.323
1 1 2 2 109.0 51.0 0.368 0.152
3 0.240
3 37.3 13.6
4 0.114
1 0.715
1 344.7 140.1
2 0.790
2 2 3 2 453.0 217.2 0.616 0.401
3 0.944
3 247.2 95.0
4 0.751
1 1.122
1 37.9 15.0
2 1.035
3 3 4 2 858.5 253.6 0.727 0.511
3 0.382
3 590.9 227.7
4 0.660
1 0.84
1 394.7 160.4
2 1.124
4 4 5 2 878.1 383.4 0.956 0.740
3 0.598
3 4.7 2.1
4 0.638
1 0.280 1 117.4 49.8
2 0.774 2 372.0 147.4 - -
5 5 6
3 0.244 3 24.2 11.60
−1000.0
4 0.301 4 0
(PV)
1 0.237 1 71.1 23.4
6 7 - 2 0.110 2 196.6 94.6 0.450 0.235
3 0.608 3 620.6 290.0
1 0.255 1 173.0 59.1
2 0.810 2 416.8 142.2
7 8 9 0.342 0.127
3 0.611 3 315.9 135.9
−1000.0
4 0.725 4 0
(PV)
1 0.979
1 247.9 86.7
2 0.918
8 10 11 2 411.6 178.3 0.549 0.333
3 0.139
3 82.5 27.1
4 0.363
1 1.333
1 220.5 103.1
2 0.821
9 11 12 2 115.4 54.0 0.314 0.098
3 0.527
3 578.0 242.9
4 0.519
1 0.087 1 60.3 23.8
10 12 - 2 0.248 2 63.7 23.5 - -
3 0.235 3 197.5 83.3
1 0.131 1 164.3 56.6
2 0.905 2 95.0 40.5
11 13 14 0.516 0.301
3 0.727 3 818.2 277.1
−1000.0
4 0.637 4 0
(PV)
Energies 2020, 13, x FOR PEER REVIEW 18 of 23

Energies 2020,
Energies 13, x2385
2020, 13, FOR PEER REVIEW 18 of
18 of 23
23

Figure 17. Comparison of voltage magnitude of FLC method, proposed method, and true value.
Figure 17. Comparison of voltage magnitude of FLC method, proposed method, and true value.
Figure 17. Comparison of voltage magnitude of FLC method, proposed method, and true value.

Figure 18. Comparison of estimation result with proposed method according to additional high-precision
Figure 18. Comparison
measurements (HPMs). of estimation result with proposed method according to additional high-
precision measurements (HPMs).
4.4. Case
Figure4: 18.
Presence of Current
Comparison Measurement
of estimation Noise
result with proposed method according to additional high-
precision
4.4. Case measurements
4: Presence (HPMs).
of Current Measurement Noise
The causes of measurement uncertainty occurring in the actual distribution network can be
largelyThedivided into the measurement error ofoccurring
the sensorinand the the asynchronous
distributionthat occurscan in the
4.4. Case 4:causes
Presence of ofmeasurement
Current Measurement uncertainty Noise actual network be
process of collecting
largely divided into the measurement
measurement information
error of the of thesensorRTUand to the theDMS. Therefore,that
asynchronous a noise
occurs injection
in the
test is
processThe ofcauses
important
collecting of
to measurement
verify uncertainty
the applicability
measurement information to an occurring
ofactual
the RTU in to
thethe
network. actual
DMS.distribution
Actual network data
Therefore, anetwork can
mayinjection
noise be be
used
largely
for the
test divided
test, but itto
is important into the
is verify measurement
difficultthetoapplicability error
verify the accuracy of the
to an actual sensor
of thenetwork.and the
methodology asynchronous
Actualusingnetwork that
the data occurs
data withmay random in
be usedthe
process
noise.
for of collecting
theTherefore,
test, but it in measurement
is this study,
difficult totheverifyinformation
load theand of the
PV profiles
accuracy of theRTU
used tointhecase
methodology DMS.1 areTherefore,
considered
using the dataa noise the injection
aswith accurate
random
test
value,
noise.is important
and the true
Therefore, to verify
in values the applicability
of switch
this study, the load andto
measurement PVanprofiles
actual
and bus network.
used voltage
in case Actual
were1 are network
extracted
considered dataasmay
through be used
theMATLAB
accurate
for the
simulation,
value, test,
and the but
and it
true is difficult
thevalues
noise of to
was verify
injected
switch the accuracy
based on the
measurement of the methodology
andinternational
bus voltage were using
standards.
extractedthe data
Forthrough with
the testMATLAB random
network,
noise.
the Therefore,
network of in
case this
studystudy,
3 was the load
used, and
and PV
the profiles
estimated used in
results
simulation, and the noise was injected based on the international standards. For the test network, case
of 1
the are
FLC considered
method, as
IRWLS the accurate
method,the
value,
and and
the the
proposed true values
method of switch
were measurement
compared. However, and bus
it
network of case study 3 was used, and the estimated results of the FLC method, IRWLS method, andisvoltage
difficult were
to extracted
consider through
the asynchronousMATLAB of
simulation,
the measurement
proposed and the
method atnoise
thewere wascompared.
injected
laboratory based
level, on theit
therefore
However, international
the isweight
difficult ofto standards.
the measurement
consider For asynchronous
the theerror
test network, of the
of the sensor the
network
was highlyof case study
reflected. 3 was
Regardingused, and
the the
range estimated
of the currentresults of
magnitude
measurement at the laboratory level, therefore the weight of the measurement error of the sensor was the FLC method,
measurement IRWLS
noise method,
to be and
added,
the proposed
as the
highly maximum
reflected. method noisewere
Regarding compared.
of class 5 is setofHowever,
the range ±5%
asthe [29],
current it isthedifficult
currenttomeasurement
magnitude consider the noise
measurement asynchronous
noise to wasbe simulated
added,of theas
measurement
with an incremental
the maximum atnoise
the laboratory
variation
of class 5 of islevel,
1% as
set therefore
from
± 5% ±1%~±5%.thethe
[29], weight For the
current of thecasemeasurement error
of load measurements,
measurement noise wasof simulated
the simulations
sensor with was
highly
were reflected.variation
conducted
an incremental Regarding
by injecting the
of 1%±1% range
fromand±1%~±of theof
±2% current
5%. load magnitude
Formeasurement
the case of load measurement
noise noisesimulations
as the maximum
measurements, to be
noiseadded,
waswereas
set
the maximum
to ±2% by class
conducted noise of
2 [30]. The
by injecting class 5
± 1%noise is set
and ±model as ±
2% of was5% [29],
loadsimulated the
measurement current
by the measurement
noisenormal noise
as thedistribution. was
maximum noise simulated
The distribution with
was set to ±
an
was incremental
simulated variation
for 1000 of 1%
cases, andfrom the ±1%~±
results 5%. wereFor the
analyzed case
2% by class 2 [30]. The noise model was simulated by the normal distribution. The distribution of
withload themeasurements,
mean, maximum, simulations
and were
standardwas
conducted
deviation. by
The injecting
random ± 1%
noisesandwere ± 2% of load
generated measurement
as many as
simulated for 1000 cases, and the results were analyzed with the mean, maximum, and standard noise
“the numberas theofmaximum
nodes wherenoise the was set
measuring to ±
2% by class
instrumentThe
deviation. 2 [30].
is installedThe
random noises noise
× 1000”model
were was
according
generatedsimulated
to asthemany by the
noiseasmodel normal distribution.
type. The
“the number noises
of nodes The distribution
werethe
where added was
to the
measuring
simulated
measurement for 1000
values cases,
by and
using thethe results
1000 sets were
of noisesanalyzed
generated
instrument is installed × 1000” according to the noise model type. The noises were added to the with as the
above mean,
and maximum,
multiplying and
(1 + standard
noise) PU
deviation.
measurement The
to the measurement random
valuesvalues noises
by using were
of current
the 1000generated
magnitude as
sets of noises many
for each as “the number
simulation
generated as above of nodes
case.and where
In order the measuring
for the noise
multiplying to be
(1 + noise)
instrument
within
PU to the is installed
the measurement
range ×values
1000” of
with a confidence according
of 99%, in
current tothe the
magnitude casenoiseoffor1%model
noise
each type.
of theThe
simulation normalnoises
case. Inwere
orderadded
distribution thetonoise
forsimulation, the
measurement
the noise is values
generated by sousing
that the
the 1000
mean sets
is of
0 noises
and 3σ generated
is 0.01
to be within the range with a confidence of 99%, in the case of 1% noise of the normal distribution PU. as above
Furthermore, and multiplying
as mentioned (1 + noise)
above,
PU
the to the is
noise
simulation, measurement
multiplied
the noise isby values of current
the measurement
generated so that the magnitude
values
meanof 0for
is the and each3σ simulation
current magnitude
is 0.01 case.
for In orderas
simulation.
PU. Furthermore, formentioned
the
Figure noise19
to be within
shows
above, the
the noise the distribution.
noise range with a The
is multiplied confidence
by thevalue ofofthe
measurement 99%, in values
X-axis the
is thecase of
the1%
added
of noisewhich
noise,
current of the isnormal
magnitude varied fordistribution
by 1% from
simulation.
simulation, the noise is generated so that the mean is 0 and 3σ is 0.01 PU. Furthermore, as mentioned
above, the noise is multiplied by the measurement values of the current magnitude for simulation.
of ± 5% current measurement noise and ± 2% load measurement noise are shown in Figure 20. The
results obtained by the IRWLS method and the proposed method were approximately closer to the
true value. However, the result obtained by the FLC method had a significantly large estimation error.
The IRWLS method did not show a large increase in estimation error when injecting the normal
distribution
Energies noise
2020, 13, 2385 of 1%~2%. The WLS-based state estimation is a method that assumes random19noise of 23
in the form of normal distribution, such as the form injected in this study, so the IRWLS method was
not sensitive to normal distribution noise. In addition, the proposed method did not show a large
±1% to ±5%
increase in theand simulated
estimation for abecause
error, total of the
fiveestimation
cases. To show as the
error was graphical
corrected form, the
through the HPM
addedeven
noise if
for each case was treated in increments of 0.1%. For the case of ±2% (black solid
there was a noise in the measured values. However, the estimation error of the FLC method was line), the generated
random
increasednoise
greatlyis shown
becauseasthe 0.1% for +0.09%
measured valueandwith−1.1% for −1.05%.
the noise was usedThe value
directly foron the Y-axis
estimation is the
without
number of noises in the range. Therefore, Figure 19 is a graph of the number of
any correction. From Table 5 and Figure 20, the proposed method was verified to provide higher noises corresponding
to each range.
estimation accuracy than the other methods even after the presence of noise in the measurements.

Figure 19. Measurement error distribution of normal distribution.

Figure 19. Measurement error distribution of normal distribution.
The voltage estimation errors of each method based on the measurement noise are shown in
TableTable
5. Comparison of the
5. Comparison current
of voltage noise value
estimation of ±5%
accuracy andmethod,
of FLC the loadIRWLS
noisemethod, ±2%,
value ofand which are
proposed
the maximum
method. noise values assumed in the case study, indicates that the proposed method showed
the highest estimation accuracy, whereas the FLC method showed the lowest estimation accuracy.
Voltage Estimation Error
When the results were compared with the noise-free case, the FLC method again showed the highest
Method Noise Standard
increase in the estimation error. When
Averagethe (%)
proposedMaximum
method was compared with the IRWLS method,
(%)
Deviation
both methods exhibited similar results in the case of average and maximum estimation errors.
± 0% 0.0236 0.1266 0.0231
IRWLS
Table 5. Comparison± 1%of voltage 0.0237
estimation accuracy0.1263 0.0231 method, and
of FLC method, IRWLS
Method
proposed method. ± 2% 0.0235 0.1268 0.0230
± 0% 0.0857 0.5377 0.1219
Voltage Estimation Error
FLC
Method ± 1% Noise 0.0882 0.5194 0.1250
Method ± 2% Average
0.0903 (%) Maximum
0.5406 (%) Standard
0.1251 Deviation
± 3% ±0% 0.0889 0.0236 0.52850.1266 0.12360.0231
IRWLS Method ±1% 0.0237 0.1263 0.0231
±2% 0.0235 0.1268 0.0230
±0% 0.0857 0.5377 0.1219
±1% 0.0882 0.5194 0.1250
FLC ±2% 0.0903 0.5406 0.1251
Method
±3% 0.0889 0.5285 0.1236
±4% 0.0905 0.5824 0.1266
±5% 0.0957 0.6210 0.1304
±0% 0.0222 0.0853 0.0089
±1% 0.0229 0.0857 0.0089
Proposed ±2% 0.0228 0.0830 0.0090
Method
±3% 0.0226 0.0942 0.0085
±4% 0.0231 0.0976 0.0093
±5% 0.0234 0.0982 0.0094

The true voltage values at node 5 and the estimated values obtained by each method for the
case of ±5% current measurement noise and ±2% load measurement noise are shown in Figure 20.
Energies 2020, 13, 2385 20 of 23

The results
Energies obtained
2020, 13, by the
x FOR PEER IRWLS
REVIEW method and the proposed method were approximately closer to
20 of 23
the true value. However, the result obtained by the FLC method had a significantly large estimation
error. The IRWLS method ± 4% 0.0905
did not show 0.5824
a large increase in estimation error when0.1266
injecting the normal
± 5% The WLS-based
distribution noise of 1%~2%. 0.0957 state estimation
0.6210is a method that0.1304
assumes random noise
± 0%
in the form of normal distribution, 0.0222
such 0.0853
as the form injected in this study, so 0.0089
the IRWLS method was
not sensitive to normal± distribution
1% 0.0229
noise. In addition,0.0857
the proposed method 0.0089
did not show a large
increase inProposed ± 2%error, because
the estimation 0.0228the estimation0.0830
error was corrected 0.0090
through the HPM even
if there wasMethod
a noise in ±the
3%measured0.0226
values. However, 0.0942
the estimation error 0.0085
of the FLC method was
increased greatly because
± 4%the measured value with the noise
0.0231 was used directly
0.0976 for estimation without
0.0093
any correction. From ±Table5% 5 and Figure
0.0234 20, the proposed method
0.0982 was verified
0.0094to provide higher
estimation accuracy than the other methods even after the presence of noise in the measurements.

Figure 20. Comparison of true value and voltage estimation value of FLC method, IRWLS method,
Figure 20. Comparison of true value and voltage estimation value of FLC method, IRWLS method,
and proposed method at node 5.
and proposed method at node 5.
4.5. Summary of Case Studies
4.5. Summary of Case Studies
The results of case studies are summarized as follows.
The results of case studies are summarized as follows.
(1) Comparison of the voltage estimation accuracy of the FLC, IRWLS, and the proposed method
(1) Comparison of the voltage estimation accuracy of the FLC, IRWLS, and the proposed method
through case study 1 indicates that a significant improvement in the estimation accuracy of the
through case study 1 indicates that a significant improvement in the estimation accuracy of the
proposed method was observed when compared with the FLC method and a slight improvement
proposed method was observed when compared with the FLC method and a slight improvement
accuracy was observed when compared with the IRWLS method. Comparison of the load
accuracy was observed when compared with the IRWLS method. Comparison of the load estimation
estimation accuracy shows that the proposed method had a significant improvement in the
accuracy shows that the proposed method had a significant improvement in the estimation error
estimation error when compared with the FLC and IRWLS methods. The FLC method and IRWLS
when compared with the FLC and IRWLS methods. The FLC method and IRWLS method were
method were significantly affected by PV, so the accuracy of load estimation was low, but the
significantly affected by PV, so the accuracy of load estimation was low, but the proposed method
proposed method was verified so that it was less affected by PV and showed more accurate
was verified so that it was less affected by PV and showed more accurate estimation results than
estimation results than other methods.
other methods.
(2) (2)The proposed
The proposed method
method performs
performssequential
sequentialvoltage estimation
voltage estimationfrom the starting
from nodenode
the starting to the
toend;
the
therefore, the
end; therefore, the cumulative
cumulative effect
effect of
of the
the estimation
estimation errorerror was
was analyzed
analyzed in in case
case study
study 2.
2. In
In the
the FLC
FLC
method,method, the estimation
the estimation error oferror of the previous
the previous node isnode is accumulated
accumulated as itthe
as it is, and is, and the estimation
estimation error of
error of the next node increases. In the case of the proposed method, when
the next node increases. In the case of the proposed method, when the estimation error of the previous the estimation error
of the previous node occurs, the estimation error of the next node eventuates.
node occurs, the estimation error of the next node eventuates. However, it was confirmed that there However, it was
confirmed that there was no accumulation of the effect toward the feeder
was no accumulation of the effect toward the feeder end. These results indicate the robustness of the end. These results
indicate
proposed the robustness
method of the proposed
for the cumulative method for
error propagation andthe
itscumulative
effectiveness error propagation
for its applicationand its
to the
effectiveness
actual network. for its application to the actual network.
(3) (3)When
Whenthe the
HPM is installed
HPM only at
is installed the at
only endtheof end
the main
of thefeeder,
mainitfeeder,
was confirmed that the estimation
it was confirmed that the
error increased as the length of the lateral feeder increased. For some
estimation error increased as the length of the lateral feeder increased. For some cases, cases, an additional HPM
an additional
may be required for the long-distance lateral feeder. In case study
HPM may be required for the long-distance lateral feeder. In case study 3, it was confirmed that,3, it was confirmed that,
whenwhen an additional
an additional HPMHPM waswas installed
installed at the
at the lateral
lateral feeder,
feeder, thetheestimated
estimatederrorerrorwas
wassignificantly
significantly
improved compared to the case where an HPM was installed only at the end of the main feeder. This
can be viewed as a significant reduction in the error for the network of this case study. If the actual
network for application is determined, the number of installed HPMs can be examined through
Energies 2020, 13, 2385 21 of 23

improved compared to the case where an HPM was installed only at the end of the main feeder.
This can be viewed as a significant reduction in the error for the network of this case study. If the
actual network for application is determined, the number of installed HPMs can be examined
through simulation as in the case study in this paper. The final number of installations will be
determined by combining requirements such as the estimation accuracy and installation costs
desired by the utility company.
(4) When the measurement noises were injected into each estimation method in case study 4,
the proposed method showed robustness to noise. The result of voltage estimation simulation
with measurement noise was that the proposed method had higher accuracy than the FLC and
IRWLS methods. Therefore, the proposed method can provide an improved estimation accuracy
compared to the existing methods even with the presence of measurement noises. This is a
significant advantage of applying the proposed method to the actual distribution system.

In conclusion, verification of the proposed method was performed in a test model similar to an
actual distribution network in this study, and through comparison with existing methods under the
same conditions, the superiority of the proposed method can be summarized as follows. First, under the
condition that PV is interconnected and there is no measurement noise, the existing methods were
greatly affected by PV, and the estimation accuracy of voltage and load was low. Although the proposed
method was less affected by PV, the estimation accuracy was more accurate. Second, the proposed
method showed that the estimated error of the previous node had little effect on the next node, so the
cumulative effect of the error was small. Third, under the presence of noise in the measurement
values of the load and current, even in the case of maximum measurement noise, it was proven
that the estimation accuracy of the proposed method was higher than other methods. These results
show that the proposed method has a great advantage in the application of the actual network,
considering the uncertainty of measurement and the recent trend that many PVs are interconnected to
the distribution network.

5. Conclusions
In this paper, the voltage estimations in distribution networks were proposed. To overcome the
limited applicability to actual distribution networks in the methods used in existing studies, this study
considered the distributed loads in a distribution network and proposed a method of correcting the
section load center through the measurements of HPMs. Even if only one HPM is installed, the accuracy
of the voltage estimation of the main feeder can be greatly improved by the correction of the section
load center. Through more accurate main feeder voltage, the estimation accuracy of the lateral feeder
can also be greatly improved. By improving the voltage estimation accuracy, the section load can be
calculated more accurately.
For the studies estimating voltage and load, verification of the proposed method through testing
is very important. We believe that there are two types or stages of verification. The first one is for
the proposed methodology itself; it verifies the validity of the formulas and the accuracy of the logic.
At this stage, it is necessary to examine the problems of the methodology itself for the application in an
actual network, and it is important to consider various factors such as measurement uncertainty and
DGs. The second is to verify the applicability of the proposed method by installing in the operation
system of an actual distribution network. This stage is not only to verify the methodology but also
to check the various variables related to a distribution network and communication to increase its
applicability when applying to the actual network.
In this paper, the first stage of verification of the accuracy of the methodology itself was performed.
In this context, in order to consider two factors of measurement uncertainty such as the measurement
error of the sensor and the asynchronous of the RTU, a noise test based on international standards
was performed. Since the actual network measurements have random noises, it is difficult to know
the true value, and it is difficult to use the actual data for the test input/output data at this stage. It is
possible to obtain accurate data by installing metering out fit (MOF), etc. in the entire actual network,
Energies 2020, 13, 2385 22 of 23

but this method is very difficult to implement at the laboratory level. Therefore, in this study, a noise
injection test was performed using simulation-based true value as input/output data. Through the
test, the proposed method was compared with previous studies and demonstrated methodological
superiority. However, since it is difficult to consider the asynchronous of measurement at the laboratory
level, much weight has been given on the measurement error of the sensor, and it is difficult that the
study has included all factors of an actual network. Therefore, it is thought that the second stage of
verification considering the uncertainty of measurement and environmental variables in the actual
network needs to be performed as a further study in the future.
Despite the advantages of the proposed approach, the following aspects still need to be considered.
First, as mentioned above, the voltage estimation method proposed in this paper is based on the
correction of the estimation error by the HPM at the feeder end. Therefore, in the case of a long-distance
lateral feeder without the HPM installed, the error of voltage estimation can only be increased. This may
require additional installation of HPMs on long-distance lateral feeders and utility companies need
to increase their installation and operation costs. However, the recent trend of grid modernization
from increased distribution network interconnection of distributed generator is accompanied by the
need to increase the installation of the new operating systems (ADMS, etc.) and new instruments
for monitoring; such an additional installation of HPMs will not act as a major obstacle. Secondly,
the purpose of the voltage estimation proposed in this paper is to estimate the voltage on the MV side
for real-time operation of the network (voltage control, real-time network reconfiguration, etc.) at the
DMS level. In general, the LV network, which is a consumer network and not the utility’s operating
area, is not considered in this study. The estimation was performed based on the representative phase
in this paper, but in the case of a three-phase unbalanced load, it can be considered if the estimation is
performed with each phase as the representative phase. Thirdly, the proposed method aims to increase
the precision of voltage and load estimation by adding only a minimum number of high-precision
measurements to the network where the RTU is installed and operated. The proposed method is
considered to be applicable to networks that apply automation of distribution networks (such as
metropolitan networks in Korea, Japan, Hong Kong, Singapore, Asia, Europe, and North America)
without significant additional cost. In addition, as mentioned above, grid modernization with the
recent increase in the interconnection of distributed generator in distribution networks will lead to a
gradual increase in the networks to which the proposed method can be applied.

Author Contributions: C.-H.O. prepared the manuscript and implemented the theory and simulations. S.-I.G.
analyzed the data. S.-Y.Y. and S.-J.A. supervised the study and discussed the results. J.-H.C. analyzed the
simulation results and commented on the manuscript. All authors have read and agreed to the published version
of the manuscript.
Funding: This research was supported by Korea Electric Power Corporation (Grant number: R18XA04).
This research was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and
the Ministry of Trade, Industry and Energy (MOTIE) of the Republic of Korea (No. 2019381010001B).
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. McDonald, J. Adaptive intelligent power systems: Active distribution networks. Energy Policy 2008,
36, 4346–4351. [CrossRef]
2. Meliopoulos, A.P.S.; Polymeneas, E.; Tan, Z.; Huang, R.; Zhao, D. Advanced distribution management
system. IEEE Trans. Smart Grid 2013, 4, 2109–2117. [CrossRef]
3. American National Standard for Electric Power Systems and Equipment Voltage Ratings (60 Hertz); NEMA ANSI
C84.1-2011; ANSI: Washington, DC, USA, 2011.
4. Jin, T.H.; Chung, M.; Shin, K.Y.; Park, H.; Lim, G.P. Real-time dynamic simulation of korean power grid
for frequency regulation control by MW battery energy storage system. J. Sustain. Dev. Energy Water
Environ. Syst. 2016, 4, 392–407. [CrossRef]
5. Virk, A.K. Maximum Loading of Radial Distribution Networks. Master’s Thesis, Electrical and
Instrumentation Engineering Department, Thapar University, Patiala, India, 2009.
Energies 2020, 13, 2385 23 of 23

6. IEC 61000-4-30: Testing and Measurement Techniques—Power Quality Measurement Methods; IEC: Geneva,
Switzerland, 2012.
7. Characteristic of Metering Out Fit. Available online: http://www.transformadores.cl/en/project/3-elements-
metering-outfit/ (accessed on 9 September 2019).
8. Micro-Synchrophasors for Distribution Grids. Available online: https://www.naspi.org/sites/
default/files/2016-10/psl_mceachern_micro_synchrophasor_distribution_grid_20160323.pdf (accessed on
9 September 2019).
9. Ju, Y.; Wu, W.; Ge, F.; Ma, K.; Lin, Y.; Ye, L. Fast decoupled state estimation for distribution networks
considering branch ampere measurements. IEEE Trans. Smart Grid 2018, 9, 6338–6347. [CrossRef]
10. Kong, X.; Chen, Y.; Xu, T.; Wang, C.; Yong, C.; Li, P.; Yu, A. hybrid state estimator based on SCADA and PMU
measurements for medium voltage distribution system. Appl. Sci. 2018, 8, 1527. [CrossRef]
11. Muscas, C.; Pau, M.; Pegoraro, P.A.; Sulis, S. Uncertainty of voltage profile in PMU-based distribution system
state estimation. IEEE Trans. Instrum. Meas. 2016, 65, 988–998. [CrossRef]
12. Muscas, C.; Sulis, S.; Angioni, A.; Ponci, F.; Monti, A. Impact of different uncertainty sources on a three-phase
state estimator for distribution networks. IEEE Trans. Instrum. Meas. 2014, 63, 2200–2209. [CrossRef]
13. Yun, S.-Y.; Chu, C.-M.; Kwon, S.-C.; Song, I.-K.; Choi, J.-H. The development and empirical evaluation of the
korean smart distribution management system. Energies 2014, 7, 1332–1362. [CrossRef]
14. Ali, A.-W.; Jianzhong, W.; Nick, J. State estimation of medium voltage distribution networks using smart
meter measurements. Appl. Energy 2016, 184, 207–218.
15. Park, J.-Y.; Jeon, C.-W.; Lim, S.-I. Accurate section loading estimation method based on voltage measurement
error compensation in distribution systems. J. Korean Inst. Illum. Electr. Install. Eng. 2016, 30, 43–48.
16. Park, S.-H.; Lim, S.-I. Voltage estimation method for distribution line with irregularly dispersed load.
Trans. Korean Inst. Electr. Eng. 2018, 67, 491–497.
17. Jin, S.L.; Gi, H.K. Comparison analysis of the voltage variation ranges for distribution networks.
In Proceedings of the 2017 IEEE International Conference on Environment and Electrical Engineering
and 2017 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I & CPS Europe), Milan, Italy,
6–9 June 2017; pp. 1–3.
18. Korea Consumer’s Electrical Installation Guide; Korea Electric Association: Seoul, Korea, 2013.
19. CSA. C22.3 NO.1-15; CSA: Toronto, ON, Canada, 2015; pp. 29–51.
20. NFPA 70, National Electrical Code, 2005th ed.; NFPA: Quincy, MA, USA, 2005.
21. Jackson, J. Future Energy, 2nd ed.; Elsevier: Amsterdam, The Netherlands, 2014; pp. 633–651.
22. AMDS of ABB. Available online: https://library.e.abb.com/public/002f1838275e45caabba43ba03d07a2e/
ADMS_9AKK106930A8216-A4-web.pdf (accessed on 20 October 2019).
23. ADMS of, GE. Available online:. Available online: https://www.ge.com/digital/sites/default/files/download_
assets/brochure_reliability%20response_PowerOnAdvantage-web.V1.pdf (accessed on 20 October 2019).
24. Active Network Management of Siemens. Available online: https://pdfs.semanticscholar.org/a2e3/
44fedf33d1009a11c643ff84b537b289ebee.pdf (accessed on 20 October 2019).
25. Gonen, T. Electric Power Distribution System Engineering, 2nd ed.; CRC Press: Boca Raton, FL, USA, 2008.
26. Choi, J.-H.; Park, D.-H. Mid-to-Long Term Operation Plan of Distribution Control Center According to Expansion of
Distribution System Intelligent Equipment (Final Report); KEPCO: Naju, Korea, 2017; pp. 14–38.
27. Panda Power. Available online: http://www.pandapower.org/ (accessed on 6 April 2020).
28. Commercial and Residential Hourly Load Profiles for all TMY3 Locations in the United States. Open EI.
Available online: https://openei.org/doe-opendata/dataset/commercial-and-residential-hourly-load-profiles-
for-all-tmy3-locations-in-the-united-states (accessed on 5 April 2020).
29. IEC 61869-2, Instrument Transformers—Part 2: Additional Requirements for Current Transformers; IEC: Geneva,
Switzerland, September 2012.
30. IEC 62053-11, Electricity Metering Equipment (a.c.)—Particular Requirements—Part 11: Electromechanical Meters
for Active Energy (Classes 0, 5, 1 and 2); IEC: Geneva, Switzerland, 2003.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).