Download Energy Saving of Conservation Voltage Reduction Based on Load

sustainability
Article
Energy Saving of Conservation Voltage Reduction
Based on Load-Voltage Dependency
Ahmet Onen
Department of Electrical and Electronic Engineering, Abdullah Gul University, Kayseri 38080, Turkey;
[email protected]; Tel.: +90-352-2248800
Academic Editor: Andrew Kusiak
Received: 15 July 2016; Accepted: 10 August 2016; Published: 15 August 2016
Abstract: Reducing voltage to reduce energy consumption, referred to as conservation voltage
reduction (CVR), can lead to energy savings. Calculating the effects of reducing voltage requires
accurate load models. This paper investigates a load-voltage dependency factor that can be
measured via field experiments using common existing instrumentation. A relationship between
the load-voltage dependency factor and the percentage of the load that is constant impedance and
the percentage that is constant power is presented. A new coordinated control algorithm using the
load-voltage dependency factor measured by the field experiments is proposed. Parametric studies
are presented which compare CVR with coordinated control versus traditional control. Across the
two model comparisons of minimizing energy consumption, the coordinated control for conservation
voltage reduction showed significant energy reduction over local control.
Keywords: voltage dependency; coordinated control; energy reduction; conservation voltage reduction
1. Introduction
Many loads consume less energy when operated at a lower voltage. Not all loads exhibit this
behavior, but often an aggregate load, such as the load on a distribution feeder, will consume less
energy when operated at a lower voltage [1,2]. Lowering voltage to reduce the energy consumed by
the aggregate load of a feeder is referred to as conservation voltage reduction, or CVR. The CVR factor
for this work is defined as the percentage power flow change that occurs when the supply voltage is
lowered by 1%.
The ability to accurately estimate the energy saving of CVR versus its cost, and to evaluate control
strategies for implementing CVR, depend upon having load models that reflect the real-world, voltage
dependency of the aggregate load [3]. Developing such models requires field experiments involving
changing the voltage magnitude and measuring the change in load. When voltage control devices
exist in a substation, they may be used to raise and lower the voltage in a series of experiments to
measure load-voltage dependency.
When measuring the effects that voltage changes have on energy consumption, instrumentation
often exists for just measuring voltage and current magnitudes, and not for accurately measuring
power flows. Such instrumentation can be used to measure the change in current magnitude relative to
a change in voltage magnitude, referred to here as the load-voltage dependency factor. The load-voltage
dependency factor may be related to other load models, such as loads modeled as a combination of
constant power, constant impedance, and constant current.
The load models that are used in power system analysis usually consist of constant impedance
and constant power loads. In reality these types of loads do not behave the same with respect to
voltage on the system [4]. In this paper this load-voltage dependency is taken into account and its
effect on a coordinated control algorithm is investigated.
Sustainability 2016, 8, 803; doi:10.3390/su8080803
www.mdpi.com/journal/sustainability
Sustainability 2016, 8, 803
2 of 11
There are two main types of control in literature, local and coordinated [5,6]. Local controllers
include time-controlled capacitors and voltage-controlled capacitors. With technological development,
decisions can be made by coordinated control where measurements and model information are
used to decide optimum settings for local controllers. Local controllers for switched capacitors,
voltage regulators and load tap changers use voltage set points which specify a range for control; the
coordinated controller would update the voltage set-points [7,8].
Controllable devices include switched capacitor banks, voltage regulators, load tab changers
(LTC) and distributed energy resources (DER) [9,10]. The set points determined by the coordinated
controller will, for instance, coordinate a capacitor bank’s operation with other controllable devices on
the feeder. The set points to the switched capacitor may change due to a number of conditions such as
growth in loading level, reconfiguration of the feeder, and based on different type of loads and their
voltage sensitivity.
In the literature a number of papers have been written to explain the effects of coordinated control
on power systems. Abdel Salam, T.S. et al. [7] describes an algorithm for optimal capacitor placement
based on determining nodes whose voltages affect losses in the system the most, but load-voltage
dependency factors are not considered. Authors in [11] propose the coordinated control algorithm with
two modes; efficiency and CVR, but the authors assumed a load-voltage dependency factor [12] uses
coordinated control for reduction of losses and to reduce controller motion, again without considering
load-voltage dependency. In [13] the author proposes a coordinated control algorithm to achieve better
voltage and operation times of LTC and switched capacitors, but again did not consider load-voltage
dependency [14] explores a coordinated control method for load tab changer and switching capacitor
to reduce the operation of both devices without considering different types of loads.
The work here focuses on a new coordinated control algorithm and uses field measurements of
load-voltage dependency factors. Approximately 1200 feeders are evaluated for CVR performance,
where 11 feeders were selected for being top CVR performers and used in a pilot study. The feeders
considered are divided into two types, short feeders, such as those commonly occurring in urban areas,
and longer feeders, such as those commonly occurring in rural areas.
This paper is organized as follows: the load-voltage dependency factor is discussed in Section 2;
the coordinated control algorithm is introduced in Section 3; a case study and a discussion of the
results are provided in Section 4; and, finally, conclusions are given in Section 5.
2. Load-Voltage Dependency Factor
Load-voltage dependency here relates percent changes in voltages to percent changes in currents.
This representation of load-voltage dependency can be calculated using existing measurements
of current magnitude and substation voltage magnitude, which are measurements that exist for
many distribution feeders that are instrumented. Table 1 shows some representative load-voltage
dependency factors.
Table 1. Load-voltage dependency factor for different loads.
Load Component
(% Change in I)/(% Change in V)
Battery Charge
Fluorescent Lamps
Constant Impedance
Air Conditioner
Constant Current
Pumps, Funs other Motors
Small Industrial Motors
Constant Power
1.59
1.07
1
−0.5
0
−0.92
−0.9
−1
From Table 1 it may be noted that, for common loads, there can be a significant difference between
the load-voltage dependencies. However, many distribution feeders have instrumentation that may
Sustainability 2016, 8, 803
3 of 11
Sustainability 2016, 8, 803
3 of 11
be used to determine the load-voltage dependency factor, but are not instrumented so that changes
changes in power may be accurately determined. For many existing feeders, measurement of loadin power may be accurately determined. For many existing feeders, measurement of load-voltage
voltage dependency at the start of a feeder may be made using existing instrumentation by running
dependency at the start of a feeder may be made using existing instrumentation by running many
many experiments, where the voltage magnitude at the start of the feeder is stepped up and down,
experiments, where the voltage magnitude at the start of the feeder is stepped up and down, and the
and the change in current magnitude is measured. Since load-voltage dependency factors vary with
change in current magnitude is measured. Since load-voltage dependency factors vary with the load
the load mixture, these experiments can be run at different times, such as during the summer and
mixture, these experiments can be run at different times, such as during the summer and during the
during the winter, to determine time varying values to use for the load-voltage dependency.
winter, to determine time varying values to use for the load-voltage dependency.
The relation between the load-voltage dependency factors for constant impedance and constant
The relation between the load-voltage dependency factors for constant impedance and constant
power load models will now be considered and is shown Figure 1.
power load models will now be considered and is shown Figure 1.
Figure
1. (a)
Circuit
for for
load-voltage
dependent
load;
andand
(b) (b)
circuit
for for
constant
impedance
andand
Figure
1. (a)
Circuit
load-voltage
dependent
load;
circuit
constant
impedance
constant
power
load.
constant power load.
Equation
(1) (1)
relates
thethe
load-voltage
dependency
Equation
relates
load-voltage
dependencymodel;
model;the
theright
righthand
handside
sideof
of Equation
Equation (1),
(1), to
to aa load
impedance,
while
thethe
leftleft
hand
side
of of
loadmodel
modelconsisting
consistingofofconstant
constantpower
powerand
andconstant
constant
impedance,
while
hand
side
Equation
(1) (1)
is the
load-voltage
dependency
model:
Equation
is the
load-voltage
dependency
model:
V×
f VN IN +
VI×=I =
V2 V
+
R
(1)(1)
where
f = fraction
of load
that is
constant
power;
V = actual
voltage
magnitude;
I = actual
current
where
of load
that
is constant
power;
V = actual
voltage
magnitude;
I = actual
current
f = fraction
vN
magnitude;
V
=
nominal
voltage
magnitude;
I
=
nominal
current;
R
=
constant
impedance
=
;
N
N
(1− f ) I
magnitude; V = nominal voltage magnitude; I = nominal current; R = constant impedance
N
2
N
N
P = f VN IN = constant power load; VR = constant
impedance load; and V × I = voltage dependent
=
=constant impedance load; and V × I = voltage
; =
= constant
power load; V
2 (1− f ) I
2
(
)
V
V
−
VN
N
load. Substituting VR =
and I =( IN) +
KIN into Equation (1) gives:
VN
VN
dependent load. Substituting
=
and I =
+
into Equation (1) gives:
(V − V )
V 2 (1 2− f ) IN
) N ) = f VN IN +
V (1  f ) I
V ( IN + (V  NV NKI
V ( I N  VN
K I N )  fV N I N  VN
VN
VN
where K = load−voltage dependency where:
where K = load−voltage dependency where:
V − VN
K
∆I =
∆I =VN
N
(2)
(2)
If K = 1, then the load becomes a constant impedance load when K = 0, the load becomes a constant
If K =When
1, then
becomes
a constant
impedance
load when K = 0, the load becomes a
current load.
K =the
−1load
the load
becomes
a constant
power load.
constant
load. When
K = −1tothe
becomes
a constant
power load.
Let the current
actual voltage
be related
theload
nominal
voltage
by:
Let the actual voltage be related to the nominal voltage by:
V = hVN
(3)
(3)
V= ℎ
Substituting
Equation
(3) (3)
intointo
Equation
(2),(2),
we we
obtain
K asKaasfunction
of hofand
f, asf,given
by: by:
Substituting
Equation
Equation
obtain
a function
h and
as given
f − hf  h h2 (h12 (1
− f )f )
K= K
+ 
h(h −h(1h)  1) h(hh−
( h 1)1)
Applying L’Hospital’s rule to Equation (4) for h = 1, we get:
(4)(4)
Sustainability
Sustainability 2016,
2016, 8,
8, 803
803
44 of
of 11
11
Applying L’Hospital’s rule to Equation
lim(4) for
K h 1= 1,2we
f get:
h 1
(5)
limh→1 K = 1 − 2 f
(5)
Given an experimentally-measured value of K and a value of h, Equations (4) and (5) may be
for different
seasons
used Given
to determine
the value of f . f values
an experimentally-measured
value associated
of K and a with
valueKofvalues
h, Equations
(4) and
(5) mayand
be
length
feeders are
plottedassociated
in Figure with
2. K values for different seasons and length of
used toofdetermine
themeasured
value of fand
. f values
feeders are measured and plotted in Figure 2.
Figure 2. Variation of load-voltage dependency factor K and fraction f of the load that is under constant
Figure
dependency
factor
K and
fraction
f of the load that is under
power 2.
as Variation
a functionofofload-voltage
the season and
length of the
feeder
power
load.
constant power as a function of the season and length of the feeder power load.
Figure 2 illustrates how the load-voltage dependency factor can vary with seasons and length
Figure 2 illustrates how the load-voltage dependency factor can vary with seasons and length of
of the feeder. The feeders are divided into two types based on length, where short feeders are less
the feeder. The feeders are divided into two types based on length, where short feeders are less than
than 5 miles long and long feeders are more than 5 miles. Short feeders consist of 60% constant power
5 miles long and long feeders are more than 5 miles. Short feeders consist of 60% constant power
(40% constant impedance) during the summer, while during the winter the same feeders behave with
(40% constant impedance) during the summer, while during the winter the same feeders behave with
45% constant power (55% constant impedance). The long feeders operate with 55% constant power
45% constant power (55% constant impedance). The long feeders operate with 55% constant power
load during the summer and 80% constant power load during the winter.
load during the summer and 80% constant power load during the winter.
3. Coordinated Control Algorithm
3. Coordinated Control Algorithm
The coordinated control algorithm serves here for minimizing energy delivered by reducing
Thewhile
coordinated
control
serves
for minimizing
energy delivered by reducing
voltage
maintaining
the algorithm
voltage within
anhere
acceptable
range.
voltage
while
maintaining
the
voltage
within
an
acceptable
range.
The control algorithm first discovers all the controllable devices—switchable capacitors,
Theregulators,
control algorithm
first
discovers allfeeders.
the controllable
devices—switchable
voltage
voltage
load tap
changers—in
Additionally,
the coordinatedcapacitors,
control algorithm
regulators,
load
tap
changers—in
feeders.
Additionally,
the
coordinated
control
algorithm
automatically discovers control devices when they inserted or removed. In addition to this discovery,
automatically
discovers
devices when
inserted
or removed.
addition
to this discovery,
the coordinated
controlcontrol
also categorizes
thethey
control
devices
as singleInstep
or multiple
step and
the coordinated control also categorizes the control devices as single step or multiple step and
determines optimum set points for the local controllers. Single step devices in this study are switched
determines optimum set points for the local controllers. Single step devices in this study are switched
capacitors since it has just ON or OFF mode. Voltage regulators and load tap changers are multi-step
capacitors since it has just ON or OFF mode. Voltage regulators and load tap changers are multi-step
control devices, since they have a 16 step size in this study and, based on the required voltage algorithm,
control devices, since they have a 16 step size in this study and, based on the required voltage
decides which steps are the best choice for voltage interval.
algorithm, decides which steps are the best choice for voltage interval.
Figure 3 shows how the discovery process of coordinated control works. This discovery needs to
Figure 3 shows how the discovery process of coordinated control works. This discovery needs
run before the coordinated control runs since the coordinated control needs to know all controllable
to run before the coordinated control runs since the coordinated control needs to know all
devices and to categorize the control devices.
controllable devices and to categorize the control devices.
Sustainability 2016, 8, 803
Sustainability 2016, 8, 803
Sustainability 2016, 8, 803
5 of 11
5 of 11
5 of 11
Figure
3.3.Controller
discovery
and categorization.
categorization.
Figure3.
Controllerdiscovery
discoveryand
Figure
Controller
categorization.
To
be
able
reachits
itsobjective,
objective,the
thecoordinated
coordinatedcontrol
controltries
tries
minimize
the
customer
voltage
To
be
able
to
reach
the
coordinated
control
triestoto
tominimize
minimize
the
customer
voltage
To
be
able
toto
reach
its
objective,
the
customer
voltage
while
maintaining
the
customer
voltage
within
the
required
operating
range.
Figure
4
shows
the
while
voltage within
within the
therequired
requiredoperating
operatingrange.
range.Figure
Figure
4 shows
the
whilemaintaining
maintaining the
the customer
customer voltage
4 shows
the
coordinated
control
algorithm
and
illustrates
the
use
of
load-voltage
dependency
factors
in
the
coordinated
control
algorithm
and
illustrates
the
use
of
load-voltage
dependency
factors
in
the
coordinated control algorithm and illustrates the use of load-voltage dependency factors in the energy
energyreduction
reductioncalculation.
calculation.
energy
reduction
calculation.
Figure4.4.Coordinated
Coordinatedcontrol
controlalgorithm
algorithmwith
withload-voltage
load-voltagedependency
dependencyfactor.
factor.
Figure
Figure 4. Coordinated control algorithm with load-voltage dependency factor.
Sustainability 2016, 8, 803
Sustainability 2016, 8, 803
6 of 11
6 of 11
The steps in the algorithm are as follows:
Figure
3 to
know
thethe
status
of
Step 1: Coordinated
Coordinated control
controlruns
runsthe
thediscovery
discoveryalgorithm
algorithmfirst
firstfrom
from
Figure
3 to
know
status
control
devices
and
categorizes
thethe
control
devices.
of
control
devices
and
categorizes
control
devices.
Step 2: The existing system can be classified as a base case, and then initial voltages and energy
delivered are calculated for the base case.
Step 3:3:Check
whether
or not
dependency
exists, and
exit ifand
there
is no
Check
whether
orload-voltage
not load-voltage
dependency
exists,
exit
if load-voltage
there is no
dependency.
load-voltage dependency.
Step 4: If a voltage-dependent load exists, then use the value from field measurements.
measurements.
Step 5: Coordinated
Coordinated control
control applies
applies load-voltage
load-voltage dependency
dependency factors
factors to
to estimate
estimate the energy
reduction achieved.
Step 6: When the iteration has finished, the coordinated control uses the solution as the starting
solution for the next iteration.
Step 7:
7: When
When the
thealgorithm
algorithmhas
hasfinished,
finished,the
the
optimum
local
control
points
minimizing
optimum
local
control
setset
points
for for
minimizing
the
the
energy,
while
maintaining
the voltage
within
the required
limits,
determined.
loadload
energy,
while
maintaining
the voltage
within
the required
limits,
havehave
beenbeen
determined.
4. Case
CaseStudy:
Study:CVR
CVRand
andEnergy
EnergyResults
Results
4.
In this
this section
section results
results from
from aaCVR
CVRstudy
studyon
ona asystem
systemconsisting
consistingofofmore
more
than
1200
feeders
In
than
1200
feeders
is
is
presented.
presented.
4.1. Results
Results for
for the
the Entire
Entire System
System
4.1.
Statistical variations
variations in
in load-voltage
load-voltage dependency
dependency factors
factors and
and CVR
CVR factors
factors for
for different
Statistical
different seasons
seasons
are
shown
in
Figures
5
and
6.
The
reason
the
load-voltage
dependency
and
CVR
factors
varying
for
are shown in Figures 5 and 6. The reason the load-voltage dependency and CVR factors varying for
different seasons
different
seasons is
is the
the change
change in
in types
types of
of loads
loads used
used in
in different
different seasons.
seasons.
Summer statistics
Wintter statistics
600
400
350
500
300
400
250
300
200
150
200
100
100
0
-0.5
50
0
0.5
1
CVR factor(%)
1.5
0
-0.5
0
0.5
1
CVR factor(%)
Figure 5. CVR factors for different seasons.
Figure 5. CVR factors for different seasons.
1.5
Sustainability 2016, 8, 803
Sustainability2016,
2016,8,8,803
803
Sustainability
7 of 11
7 7ofof1111
600
600
Summerstatistics
statistics
Summer
600
600
500
500
500
500
400
400
400
400
300
300
300
300
200
200
200
200
100
100
100
100
0 0-2
-2
00
22
44
voltagedependency
dependencyfactor(%)
factor(%)
voltage
Wintterstatistics
statistics
Wintter
0 0-4
-4
-2-2
00
22
44
voltagedependency
dependencyfactor(%)
factor(%)
voltage
Figure 6. Load-voltage dependency factors for different seasons.
Figure6.6.Load-voltage
Load-voltagedependency
dependencyfactors
factorsfor
fordifferent
differentseasons.
seasons.
Figure
Implementing
CVR
with
coordinated
control
versus
implementing
CVR
with
local
control
ImplementingCVR
CVRwith
withcoordinated
coordinatedcontrol
controlversus
versusimplementing
implementingCVR
CVRwith
withlocal
localcontrol
controlisis
is
Implementing
compared
next.
From
Figures
7–10
it
may
be
seen
that
the
load
energy
is
significantly
lower
with
comparednext.
next.From
FromFigures
Figures7–10
7–10ititmay
maybe
beseen
seenthat
thatthe
theload
loadenergy
energyisissignificantly
significantlylower
lowerwith
with
compared
coordinated
control.
InInFigures
7–10
only
feeders
that
have
an energy
savings
moremore
than than
1% with
CVR
coordinated
control.
Figures
7–10
only
feeders
that
have
an
energy
savings
1%
with
coordinated control. In Figures 7–10 only feeders that have an energy savings more than 1% with
are
considered,
whichwhich
amounted
to 260to
feeders.
CVR
areconsidered,
considered,
amounted
260feeders.
feeders.
CVR
are
which amounted
to 260
savingsduring
duringthe
reduction.
Figure7.7.Urban
Urbanarea
areafeeder’s
feeder’ssavings
thesummer
summerwith
withpercentage
percentagevoltage
voltagereduction.
reduction.
Figure
Figures7–10
7–10
show
the
percent
energy
savings
with
respect
percent
voltage
reduction.
Figures
savings
with
respect
totopercent
voltage
reduction.
InIn
Figures
7–10show
showthe
thepercent
percentenergy
energy
savings
with
respect
to
percent
voltage
reduction.
Figure
7
summer
savings
are
illustrated
for
the
urban
area’s
feeders
while
Figure
8
shows
winter
Figure
7 summer
savings
areare
illustrated
forfor
thethe
urban
area’s
feeders
In Figure
7 summer
savings
illustrated
urban
area’s
feederswhile
whileFigure
Figure8 8shows
showswinter
winter
savingsfor
forthe
theurban
urbanarea’s
area’sfeeders.
feeders.For
Forurban
urbanarea’s
area’sfeeders
feedersduring
duringthe
thesummer
summerthe
theload-voltage
load-voltage
savings
savings
for
the
urban
area’s
feeders.
For
urban
area’s
feeders
during
the
summer
the
load-voltage
dependencyfactor
factor
−0.2%,
which
means
thethe
system
consists
60%
constant
power
loadand
and40%
40%
dependency
which
means
the
system
consists
ofof60%
constant
power
load
dependency
factorisis
is−0.2%,
−0.2%,
which
means
system
consists
of 60%
constant
power
load
and
constant
impedance
load.
Average
customer
energy
savings
for
the
urban
area’s
feeders
during
the
constant
impedance
load. load.
Average
customer
energy
savings
for the
area’sarea’s
feeders
during
the
40% constant
impedance
Average
customer
energy
savings
forurban
the urban
feeders
during
summer
is
around
2.5%.
summer
is around
2.5%.2.5%.
the summer
is around
Sustainability 2016, 8, 803
Sustainability 2016,
2016, 8,
8, 803
803
Sustainability
8 of 11
of 11
11
88 of
Figure 8. Urban area feeder’s savings during the winter with percentage voltage reduction.
Figure 8. Urban area feeder’s savings during the winter with percentage voltage reduction.
Figure 8. Urban area feeder’s savings during the winter with percentage voltage reduction.
Figure 8 looks at the urban area’s feeders during the winter time where the load-voltage
Figurefactor
looks
at the
thewhich
urbanmeans
area’sthe
feeders
during
theofwinter
winter
time where
where
the
load-voltage
dependency
is 0.1%,
systemduring
consists
45% constant
powerthe
load
and 55%
Figure
88 looks
at
urban
area’s
feeders
the
time
load-voltage
dependency
factor
is
0.1%,
which
means
the
system
consists
of
45%
constant
power
load
andthe
55%
constant
impedance
Average
customer
urban
area’s power
feedersload
during
dependency
factor isload.
0.1%,
which means
the energy
system savings
consists for
of 45%
constant
and
55%
constant
impedance
customer
energy
savings
for urban
area’sarea’s
feeders
duringduring
the winter
winter
time
is around load.
3.3%.
constant
impedance
load.Average
Average
customer
energy
savings
for urban
feeders
the
time
is
around
3.3%.
winter time is around 3.3%.
Figure 9. Rural areas feeder’s savings during the summer with percentage voltage reduction.
Figure 9. Rural areas feeder’s savings during the summer with percentage voltage reduction.
Figure 9. Rural areas feeder’s savings during the summer with percentage voltage reduction.
Figure9 looks
9 looks
the
rural
area’s
feeders
during
summer,
where
the
load-voltage
dependency
Figure
atat
the
rural
area’s
feeders
during
thethe
summer,
where
the
load-voltage
dependency
factorisisaround
around0.1%,
0.1%,which
whichmeans
meansthat
thatthe
thesystem
systemconsists
consistsofof55%
55%constant
constantpower
powerload
loadand
and45%
45%
factor
Figure 9 looks
at the rural
area’s feeders
during the
summer, where
the load-voltage
dependency
constant
impedance
load.
The
average
customer
energy
savings
for
rural
area’s
feeders
during
the
constant
load.
The average
customer
energy
savings
rural
area’spower
feedersload
during
factor isimpedance
around 0.1%,
which
means that
the system
consists
of for
55%
constant
and the
45%
summer
time
is
around
2.8%.
summer
is around
2.8%.
constanttime
impedance
load.
The average customer energy savings for rural area’s feeders during the
summer time is around 2.8%.
Sustainability2016,
2016,8,8,803
803
Sustainability
9 of
9 of
1111
Figure 10. Rural area feeder’s savings during the winter with percentage voltage reduction.
Figure 10. Rural area feeder’s savings during the winter with percentage voltage reduction.
Figure1010the
therural
ruralarea’s
area’sfeeders
feedersduring
duringthe
thewinter
winterare
areconsidered,
considered,where
wherethe
theload-voltage
load-voltage
InInFigure
dependencyfactor
factorisis0.57%,
0.57%,which
whichmeans
meansthat
thatthe
thesystem
systemconsists
consistsofof78.5%
78.5%constant
constantpower
powerload
loadand
and
dependency
11.5%constant
constantimpedance
impedanceload.
load.The
Theaverage
averagecustomer
customerenergy
energysavings
savingsfor
forrural
ruralarea’s
area’sfeeder
feederduring
during
11.5%
thewinter
winterisisaround
around3%.
3%.
the
4.2.Results
Resultsforfor1111Feeders
Feeders
4.2.
Ofthe
the260
260feeders
feedersconsidered
consideredininSection
Section4.1,
4.1,the
the1111feeders
feedersthat
thathad
hadthe
thebest
bestCVR
CVRperformance
performance
Of
were
identified.
It
should
be
noted
that
these
top
performing
feeders
all
had
a
relatively
flatvoltage
voltage
were identified. It should be noted that these top performing feeders all had a relatively flat
profile.Table
Table2 2shows
showsthe
theload-voltage
load-voltagedependency
dependencyfactors
factorsfor
forthese
thesefeeders
feeders
for
the
different
seasons.
profile.
for
the
different
seasons.
As
can
see
from
the
table,
during
the
summer
the
load-voltage
dependency
factors
are
higher
than
As can see from the table, during the summer the load-voltage dependency factors are higher than
during
the
winter.
during the winter.
Table2.2.Load-voltage
Load-voltagedependency
dependencyfactor
factorfor
fortop
topperforming
performingfeeders.
feeders.
Table
Feeder
Feeder No.
No.
1
1
22
33
4
5
66
7
7
8
89
9
10
10
11
11
Load-VoltageDependency
DependencyFactor
Factor
Load-Voltage
Summer
Winter
Summer
Winter
−0.051
−0.051
−0.049
−0.049
0.000
0.000
−0.131
−0.131
−0.055
−0.055
−0.064
−0.064
−0.036
−0.036
−0.058
−0.058
−0.025
−0.025
−0.057
−0.057
−0.106
−0.106
0.007
0.007
0.049
0.049
0.000
0.000
−−0.008
0.008
−−0.018
0.018
0.020
0.020
0.069
0.069
0.057
0.057
0.087
−0.087
0.030
−−0.030
0.246
−0.246
Using the load-voltage dependency factors, the coordinated control was used to analyze the
Using the load-voltage dependency factors, the coordinated control was used to analyze the 11
11 feeders every hour for an entire year (8760 time points). Figure 11 compares the power savings of
feeders every hour for an entire year (8760 time points). Figure 11 compares the power savings of
coordinated control with the power savings of local control.
coordinated control with the power savings of local control.
Sustainability
Sustainability2016,
2016,8,8,803
803
10
10of
of11
11
Figure
coordinated control
control power
power savings
savingsfor
for11
11top
topperforming
performingfeeders
feedersasasa
Figure 11.
11. Local
Local control
control versus
versus coordinated
afunction
functionofofhour
hourofofthe
theyear.
year.
Based
11,11,
savings
from
the coordinated
control
is 2–3istimes
(around(around
10 GWhr)
Basedon
onthe
theFigure
Figure
savings
from
the coordinated
control
2–3 greater
times greater
10
than
thethan
localthe
control
savings.savings.
The savings
changechange
throughout
the year
based
on demand.
It is
GWhr)
local control
The savings
throughout
theisyear
is based
on demand.
interesting
to note
that the
the demand,
the greater
the savings.
It is interesting
to note
thatlarger
the larger
the demand,
the greater
the savings.
5.
Conclusions
5. Conclusions
Load-voltage
cancan
be measured
experimentally
on many
existing
Load-voltagedependency
dependencyfactors
factors
be measured
experimentally
on feeders
many with
feeders
with
measurement
devices. devices.
With respect
to load-voltage
dependency
factors,factors,
load combinations
were
existing measurement
With respect
to load-voltage
dependency
load combinations
determined
which which
were equivalent
to the load-voltage
dependency
factors. factors.
A new coordinated
control
were determined
were equivalent
to the load-voltage
dependency
A new coordinated
algorithm
that made
use
of load-voltage
dependency
factors was
described.
control algorithm
that
made
use of load-voltage
dependency
factors
was described.
Two
models
were
compared
based
on
the
reduction
of
energy.
The
Two models were compared based on the reduction of energy. Thefirst
firstmodel
modelinvolved
involved feeders
feeders
with
local
control
and
the
second
model
involved
feeders
with
coordinated
control.
The
results
with local control and the second model involved feeders with coordinated control.
Theshowed
results
that
coordinated
control with
conversation
voltage reduction
helps to reduce
energythe
delivered.
showed
that coordinated
control
with conversation
voltage reduction
helps the
to reduce
energy
Results
presented
information
about energyabout
reduction
forreduction
different load-voltage
dependency
delivered.
Resultsprovide
presented
provide information
energy
for different load-voltage
factors.
Across
the
two
model
comparisons
of
minimizing
energy
consumption,
the
coordinated
dependency factors. Across the two model comparisons of minimizing energy consumption, the
control
for conversation
reduction
showed
significant
energy
reductionenergy
over local
control.
coordinated
control for voltage
conversation
voltage
reduction
showed
significant
reduction
over
local control.
Acknowledgments: The work is supported by the project “Rastgele Yük Eğrisi metodu ile elektrik kayıplarının
hesaplanması”
funded
bywork
national
institute AGU
project
BAP with
project
ID 48.ile elektrik kayıplarının
Acknowledgments:
The
is supported
by thewith
project
“Rastgele
Yükthe
Eğrisi
metodu
hesaplanması”
funded
by
national
institute
AGU
with
project
BAP
with
the
project
ID 48.
Conflicts of Interest: The authors declare no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.
References
References
1.
Ahmet, O. Model-Based Grid Modernization Economic Evaluation Framework. Ph.D. Thesis, Virginia
Institute and State
Blacksburg,
VA, Evaluation
USA, MarchFramework.
2014.
1. Polytechnic
Ahmet, O. Model-Based
GridUniversity,
Modernization
Economic
Ph.D. Thesis, Virginia
2.
Gönen,
T.
Electric
Power
Distribution
System
Engineering,
2nd
ed.;
CRC
Press:
Boca Raton, FL, USA, 2008.
Polytechnic Institute and State University, Blacksburg, VA, USA, March 2014.
3.
Broadwater,
R.; Hambrick,
J. Modeling
and Verification
of Press:
Distributed
Generation
and Voltage
2. Davis,
Gönen,M.W.;
T. Electric
Power Distribution
System
Engineering,
2nd ed.; CRC
Boca Raton,
FL, USA,
2008.
Regulation
Equipment
for
Unbalanced
Distribution
Power
Systems.
Available
online:
http://iucat.iu.edu/
3.
Davis, M.W.; Broadwater, R.; Hambrick, J. Modeling and Verification of Distributed Generation and
iue/7650458
(accessed on
2 August 2016).
Voltage Regulation
Equipment
for Unbalanced Distribution Power Systems. Available online:
Eminoglu,
U.;
Hocaoglu,
M.H.
New Power
Method
for Radial Distribution System including Voltage
4.
http://iucat.iu.edu/iue/7650458A(accessed
on 2Flow
August
2016).
Dependent
Load
Models.
Electr.
Power
Syst.
Res.
2005,
76,
106–114.
4.
Eminoglu, U.; Hocaoglu, M.H. A New Power Flow Method for[CrossRef]
Radial Distribution System including
5.
Hambrick, J.; Broadwater, R.B. Configurable, Hierarchical, Model-based Control of Electrical Distribution
Voltage Dependent Load Models. Electr. Power Syst. Res. 2005, 76, 106–114.
Circuits. IEEE Trans. Power Syst. 2011, 26, 1072–1079. [CrossRef]
5.
Hambrick, J.; Broadwater, R.B. Configurable, Hierarchical, Model-based Control of Electrical Distribution
Circuits. IEEE Trans. Power Syst. 2011, 26, 1072–1079.
Sustainability 2016, 8, 803
6.
7.
8.
9.
10.
11.
12.
13.
14.
11 of 11
Hambrick, J.; Broadwater, R. Advantages of Integrated System Model-Based Control for Electrical
Distribution System Automation. In Proceedings of 18th IFAC World Congress, Milano, Italy,
28 August–2 September 2011.
Abdel-Salam, T.S.; Chikhani, A.Y.; Hackam, R. A New Technique for loss reduction using compensating
capacitors applied to distribution system with varying load condition. IEEE Trans. Power Deliv. 1994, 9,
819–827. [CrossRef]
Joon-Ho, C.; Seon-Ju, A.; Hae-Kon, N.; Jae-Chul, K.; Seung-Il, M.; Won-Wook, J.; Il-Keun, S. The Coordinate
Control Method of LTC Transformer and Capacitor Banks at Distribution Substation. J. Electr. Eng. Technol.
2012, 7, 320–329.
Jung, J.; Cho, Y.; Cheng, D.; Onen, A.; Arghandeh, R.; Dilek, M.; Broadwater, R.P. Monte Carlo analysis of
plug-in hybrid vehicles and distributed energy resource growth with residential energy storage in Michigan.
Appl Energy 2013, 108, 218–235. [CrossRef]
Arghandeh, R.; Onen, A.; Broadwater, R.P. Distributed energy storage system control for optimal adoption
of electric vehicles. In Proceedings of IEEE Power and Energy Society General Meeting, San Diego, CA, USA,
22–26 July 2012; pp. 1–8.
Onen, A.; Cheng, D.; Arghandeh, R.; Jung, J.; Woyak, J.; Dilek, M.; Broadwater, R. P. Smart model based
coordinated control based on feeder losses, energy consumption, and voltage violations. Electr. Power
Compon. Syst. 2013, 41, 1686–1696. [CrossRef]
Jung, J.; Onen, A.; Arghandeh, R.; Broadwater, R.B. Coordinated Control of Automated Devices and
Photovoltaic Generators for Voltage Rise Mitigation in Power Distribution Circuits. Renew. Energy 2014, 66,
532–540. [CrossRef]
Park, J.Y.; Nam, S.R.; Park, J.K. Control of a ULTC considering the dispatch schedule of capacitors in
a distribution system. IEEE Trans. Power Syst. 2007, 22, 755–761. [CrossRef]
Niknam, T. A new HBMO algorithm for multiobjective daily volt/var control in distribution systems
considering distributed generators. Appl. Energy 2011, 88, 778–788. [CrossRef]
© 2016 by the author; 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/).