Electrical Power and Energy Systems 27 (2005) 398–408

www.elsevier.com/locate/ijepes

Placement of distributed generators and reclosers for distribution

network security and reliability
D.H. Popovića,*,1, J.A. Greatbanksb,1, M. Begovićc, A. Pregeljd,2
a
Energy Networks Association (ENA), 1 Stanhope Place, London W2 2HH, UK
b
AEA Technology, London WC1H 0JN, UK
c
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332 0250, USA
d
Georgia Transmission Company, Atlanta, GA, USA

Received 19 August 2003; revised 25 November 2004; accepted 2 February 2005

Abstract
Electricity market restructuring, advances in energy generation technology and agreements on the reduction of global greenhouse gas
emissions have paved the way for an increase in the use of distributed generation. This paper formulates and discusses a methodology for the
optimal siting of distributed generators and reclosers a security and reliability constrained distribution network can accept. Optimal siting is
determined by sensitivity analysis of the power flow equations. The sizing method for a set of loading conditions, generation penetration level
and power factor is formulated as a security constrained optimization problem. The information on optimal generation sites is used further to
optimize system reliability assessed via reliability indices calculation. A genetic algorithm is designed to solve for optimal recloser positions
when distributed generators are deployed in a securely optimal manner.
q 2005 Elsevier Ltd. All rights reserved.

Keywords: Distributed generation; Genetic algorithm; Optimization; Protection; Reliability; Static security

1. Introduction The presence of additional generation on a feeder may also

allow for stand-alone, island mode operation where DGs are
Recent years have seen a trend towards the development supplying portions of the feeder load after a fault has been
and deployment of distributed generation (DG) due to isolated. Islanded operation, however, requires significant
government policy changes and increased availability of coordination of distributed generators with feeder protection
small capacity generation technologies. The nature of devices in order to create possible self-supporting islands.
distributed generation is smaller plant (less than 100 MW) Extensive operations planning system analysis is needed for
with little or limited central control, connected to the distributed generation integration to be successful from both
distribution system. Distribution systems have been tra- system security and reliability point of view. This paper will
ditionally designed to operate with unidirectional power address the issue of coordinated and optimal placement of
flow, from the source (transmission system) to the loads. distributed generators and reclosers into a security con-
Adding DG to a distribution system imposes a different set strained distribution system.
of operating conditions on the network, such as reverse Connecting a DG source to the distribution system must
power flow, voltage rise, increased fault levels, reduced be done so as operating conditions are kept within given
power losses, harmonic distortion and stability problems. limits. Clearly, the effect of adding DG on network security
and reliability will vary depending on its type and position
* Corresponding author.
and (forecast) load at the connection point. Consequently,
E-mail address: dragana.popovic@energynetworks.org one or more sites on a given network may be optimal.
(D.H. Popović). Despite relatively few generators being connected to the
1
Work done while at Imperial College London. network at present, development of distributed generation is
2
Work done while at Georgia Institute of Technology. not envisiged to be centrally planned or operated. With this
0142-0615/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. in mind, DG should be introduced for overall system benefit,
doi:10.1016/j.ijepes.2005.02.002 especially if the grid exists to provide backup capacity. If an
 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408 399

optimal siting strategy is sought, incentive programs may be manner. A 114-bus mixed urban and rural 11 kV feeder in
considered in order to facilitate implementation of DG in the the UK is used to verify and demonstrate the methodology.
desirable location(s). Alternatively, additional charges may
be imposed for DG placed so to cause problems and/or
lower system performance. 2. Operation and feeder design optimization for
Preliminary studies have already shown that unless distributed generation
backup capacity is provided, stand-alone distributed gener-
ation may lower system reliability [1,2]. Similarly, it could The introduction of generating sources into the distri-
harm system reliability if it is not properly coordinated, bution system can significantly impact the operating state and
located and designed to work with existing network dynamics of both the transmission and distribution systems.
protection. In a radial feeder, protection devices are only While at low/modest levels of DG penetration, the impacts on
expected to detect unidirectional flow of current. In a the high voltage transmission system may not be significant,
majority of cases, only one device operates per fault. impacts at the lower voltage distribution level could be much
The control logic for protection devices therefore is larger especially with respect to fault current levels, the
simple—the nearest recloser upstream from the fault magnitude and direction of real and reactive power flow,
location detects the fault current, trips, and goes into a the system voltage (both steady-state and transient) and the
predefined reclosing sequence in order to restore service, in system stability under various small and large signal transient
case the fault was of a temporary nature. If more reclosers conditions. The impacts and interactions can be both positive
are present on the radial feeder, they are coordinated, and negative depending on the distribution network operat-
usually via time lags, such that the recloser closest to the ing characteristics and the distributed generation character-
fault operates. In a DG-enhanced feeder, power flow is not istic, placement and size. A proper placement plays a very
unidirectional and conventional protection logic must be important role since power flows at the interface substations
altered in order for the fault-detecting devices to success- and throughout the networks depend on geographic distri-
fully perform their function [3]. A faulted branch may be bution of all generation sources with respect to demand
energized from both sides and several protection devices irrespective of the voltage at the connection point. For
may need to operate in order to completely interrupt the distributed generation to have a positive effect, it must be at
fault current. Several control control strategies, using only least suitably integrated and coordinated with the distribution
local or SCADA measurements, may be utilized. Distrib- system operating practices and feeder design [4]. In order to
uted generation and storage units, located on the feeder, may further the positive effect and enhance network capacity
be power and/or energy limited, and may include renewable limits while contributing to system security and quality of
DG, whose output is dependent on the meteorological supply, local optimization would be required accompanied
conditions. Those sources may reduce the number of faults with taking advantage of any inherent regulation capability
and/or fault durations for customers within their protection of dispersed generation.
zones, thus increasing the reliability of service. In short, the addition of DG will usually cause changes in
This paper develops a methodology for systematic and voltage magnitudes and power flows. These changes will
rational placement of distributed resources and reclosers in affect system losses. There are obvious implications for the
distribution networks. The following related optimization current rating of lines resulting from modified power flows,
tasks may be investigated: (a) optimize recloser placement and voltage changes could see voltages rise to undesirable
for a given DG allocation, (b) optimize DG placement for a levels. Generators operating with a leading power factor
given recloser allocation, and (c) optimize both recloser and may compound the latter. In addition, DG injected power
DG placement. In this paper, the optimal recloser placement may result in voltage that is within limits at the DG site but
problem is solved for a previously determined optimal could be out of limits further downstream. The addition of
position of the DGs. Both voltage sensitivity analysis and extra power sources to a network also impacts on system
loss sensitivity analysis of the power flow equations are used fault levels and may fault currents increase beyond the
to determine the optimal sites for placement of distributed rating of circuit breakers. The essence is that adding
generators. It is followed by a security constrained generators to a passive distribution system makes it an
optimization method which calculates the quantity of DG active distribution system, akin to a mini transmission
that can be connected to specified points with the system system, and extra thought must be given to its operation and
remaining secure. The assessment takes into account the control. More specifically, in voltage profile and regulation
distributed resource power factor characteristics and load studies, available transmission capacity studies, as well as
profiles for various operating conditions. The information cost studies, the connection point, type, size and location of
on optimal generation sites is used further to optimize DG, the voltage regulator settings and independence
system reliability assessed via calculation of reliability characteristics of the line must all be considered for various
indices which include the DG units. A genetic algorithm is load and load density levels.
designed to solve for optimal recloser positions when Similar considerations must be given to islanding
distributed generators are deployed in a securely optimal response during upstream operation of protection and faults
400 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408

on the system. In a conventional (radial) feeder, protection after a fault between reclosers 1 and 2, recloser 1 operates,
device placement is designed to maximize network leaving all customers downstream without service. If DG is
reliability, and therefore minimize the reliability indices present, recloser 2 would also operate, allowing the portion
assuming energy source(s) located only at substation(s). As of the feeder downstream from it to operate as an island.
a brief reminder, the standard reliability performance In order to operate in island mode, DG(s) have to be able
indices, such as SAIFI, SAIDI and MAIFIe, and the to satisfy the islanded load, and therefore keep both the
composite index obtained as a combination of all three voltage and frequency within acceptable ranges. Islanded
will be considered. The system average interruption operation requires significant coordination of distributed
duration index (SAIDI) and the system average interruption generators with feeder protection devices. The sequence of
frequency index (SAIFI) are typically used to measure events after the fault should be as follows:
average accumulated duration and frequency of sustained
interruptions per customer. The momentary average inter- † DG is tripped, and fault detected and isolated by one or
ruption event frequency index (MAIFIe) measures the more protection devices.
number of momentary interruptions per customer. The † DG reconnects if not within the faulted zone.
recloser placement can be optimized with respect to any of † After the fault is cleared, recloser synchronizes its
these three, or some other, indices. In order to include the reclosing operation with DG.
effects of both sustained and momentary interruptions, a
composite index may be used, as defined below The positions of protection devices and distributed
generators are therefore strongly dependent. Incorrect
ðSAIFI K SAIFIT Þ recloser placement may lead to islands with not enough
CRI Z WSAIFI
SAIFIT generation and would not yield additional reliability
benefits. On the other side, by strategically placing
SAIDI K SAIDIT reclosers, one may be able to significantly increase the
C WSAIDI
SAIDIT reliability of service to customers in such islands.
MAIFIe K MAIFIeT Typically, there will be a momentary interruption to the
C WMAIFIe (1) customers in the island, due to the need for the DG to
MAIFIeT
disconnect after the fault in order not to interfere with
where Wx indicates appropriate reliability index weights, protection devices’ operation. If, however, reclosers are able
and the subscript T indicates the target value. (Note that to disconnect immediately, there may not be even a
these target values indicate satisfactory level of reliability momentary interruption, and thus MAIFIe index may also
for a conventional feeder, and may be exceeded in a DG- be reduced.
enhanced feeder yielding negative values for the composite
index CRI.) The weights are determined by the character-
istics of the feeder, customer mix and utility’s perception of 3. Optimal placement and size of distributed generators
the importance of momentary and sustained outages for the
feeder under study. Optimal placement of distributed generators for
The conventional logic suggests placing a recloser at the enhanced reliability, reduced transmission and distribution
half-way point of a radial feeder with uniformly distributed costs and reduced emissions can be realized only by
load, which, in theory, would yield a 25% feeder-wide considering all factors, including the loss reduction
reliability improvement. Similarly, locations at 1/3 and 2/3 achieved system wide and on the feeders, security limits
of feeder length should be considered for placement of two and cost/benefit analysis. It is a very complex problem
reclosers. In reality, in the presence of critical loads and considering a high number of options in terms of sites and
non-uniform load distributions, utilities often resort to units available and a need to account for a 8760 h load
engineering judgment to place reclosers. As an example, profile and generation profile and associated uncertainties
Fig. 1 shows a typical rural feeder, with substation breaker [5–7]. In [8], the OPF-based optimal placement is proposed
and two reclosers. Assuming there is no DG at the end of the addressing the effect of DGs on the spot prices and stability
feeder, a fault anywhere on the line will lead to the opening limits. Reference [9] investigates locational aspects of DG
of the first recloser upstream from the fault. For example, with respect to transmission and distribution losses. Other
studies have tackled the optimal placement problem using
genetic algorithm techniques [10,11] or tabu/parallel tabu
searches [12]. These techniques vary in complexity and
computation time, to implement and require some degree of
cost data. The tabu search in particular is computationally
intensive. In this study, costs related to adding the DG and
Fig. 1. Strategically placed reclosers increase reliability of the system by transmission/distribution upgrades and/or savings are not
reducing the number of customers affected by the fault. taken into account and the network capacity limits
 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408 401

are evaluated based on the impacts of distributed generation 3.1.2. Loss sensitivity
on the system losses, security and adequacy of supply. The majority of power losses are ohmic in nature caused
To assess network capability to absorb available by power flow through lines and transformers, i.e.
distributed resources safely, a steady-state system represen-
Ploss Z Pðd; VÞ; Qloss Z Qðd; VÞ (5)
tation in the form of power flow equations will be used. The
inverse power flow Jacobian relates changes in power Combining Eqs. (2) and (5) gives
injections to changes in angles and voltages, i.e. 2 3 2 3
vPloss vPloss
" # " vP vP #K1 " # 6 vP 7 h iK1 6 7
6 7 4 vd 5
Dd DP T
Z
vd vV
(2) 4 vPloss 5 Z J vPloss
(6)
DV vQ vQ DQ
vd vV vQ vV
The Loss Sensitivity Index (LSI) is defined as
3.1. Optimal DG siting    
vPloss vPloss
LSI Z w C ð1 K wÞ (7)
In order to determine the most suitable sites for DGs, two vP vQ
sensitivity based approaches related to voltage control and
power loss are proposed. Both a voltage sensitivity index
(VSI) and loss sensitivity index (LSI) are defined and used
3.2. Sizing of DG
to identify and rank the nodes within the network with
respect to receiving new generation. It is assumed that
Determination of the optimal sites for DG placement in
generators can connect to any point in the network subject to
Section 3.1 is followed by determination of the amount of
security constraints and are not restricted in their location by
DG that can be added at these sites without loss increase and
generator controllers or existing protection devices.
operational constraints violation. The sizing method is
formulated as a constrained optimization problem adapted
3.1.1. Voltage sensitivity
from a reactive power compensation sizing algorithm [14]
Assuming that angle-related problems are not a concern,
and capacitor bank sizing algorithm [15,16]. Given
the voltage sensitivity can be defined as
information on the available distributed generation and
   
vV vV assuming no expected load growth in the region of interest,
½DV Z ½DQ C ½DP (3) the objective is to maximize the quantity of distributed
vQ vP
generation connected to a system, i.e.
From (2), for each system node, there is an associated real
power sensitivity (vV/vP) and reactive power sensitivity X
n
max ðPGi C jQGi Þ (8)
(vV/vQ). These values can be used to rank the overall iZ1
voltage sensitivity of each node to real or reactive power
injection. A Voltage Sensitivity Index (VSI) used in ranking where PGi and QGi are the real and reactive power
is defined as [13] injections at each node i, respectively. The equality
    constraints are the power flow equations. The inequality
vV vV constraints are
VSI Z w C ð1 K wÞ (4)
vP vQ
† voltage operational tolerance limits at all buses
The diagonal elements of the Jacobian matrix represent
the sensitivity of one bus voltage magnitude to the injection Vimin % Vi % Vimax (9)
of power at the same bus, whereas the off diagonal elements
represent the sensitivity to power injected at other buses.
† limit on losses
Since the purpose of adding dispersed generation is to bring X X
about an improvement in network performance, the effect of PlossG % Ploss (10)
power injection at a single bus on the voltage sensitivities of ij ij
the whole network must be considered. This is achieved by
where Ploss is the power loss in the line from node i to
expressing the VSI for each node as an Euclidean norm
node j without distributed generation, and PlossG with
normalised across all load buses. The value of the weighting
distributed generation.
factor w will depend on the X/R ratio of the network under
† limit on total power generated by DG subject to a
consideration. In a typical 11 kV UK distribution system, X
penetration level of 20% (e.g. it must not exceed 20%
is approximately equal to R and so a weighting factor of 0.5
feeder load).
is used.
The nodes are ranked according to the VSI value and the X
n X
n X
n X
n
ranked set is used to define the optimum sites to accept PGi % 0:2 PLi ; QGi % 0:2 Q Li (11)
injection of P and/or Q. iZ1 iZ1 iZ1 iZ1
402 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408

where PLi and QLi are the real and reactive loads at (The UK has approximately 100 GW installed capacity and
bus i. 60 GW annual peak load). As such, it is unlikely that many
† branch flows limits (e.g. they must remain below distribution networks will experience more than 10–20%
thermal limits) penetration. Current protection and voltage regulation
practice does not account for high DG penetration. As
Sij % Sijmax (12)
such, automatic tap changing, automatic voltage regulation
on long feeders and directional protection devices may not
† fault current limits (e.g. they must be less than the function correctly with high penetration levels, and a
maximum fault current rating of the switchgear on each penetration level greater than 100% where power is exported
line) from the system (back through grid transformer) is infeasible.
The optimal placement and sizing methods are combined
IFij % IFijmax (13) to add DG penetration with generators connected at optimal
points. The proposed solution algorithm is shown in Fig. 2.
The figure of 20% was chosen for the DG penetration level
The sizing element is an iterative process, and involves
in line with UK government targets for renewable and
repeatedly solving load flow equations. Each iteration has a
distributed generation. The government has pledged to
larger value of DG source connected at predetermined
connect 10 GW of cogeneration and supply 10% of power
(optimal) points. The DG size increment is a fixed size unit,
by renewables by 2010 to meet Kyoto Protocol targets [19]
in this case 10 kW, and is applied to all optimal sites in

Start

Select Data File, Max Generator

Increment, Weighting Factors

Initial System Load Level

Initial Generator Type

Solve Initial Load Flow and Fault Levels

Calculate VSI & LSI

Increment Generators at Viable Nodes

Normalize & Rank

Solve Load Flow and Fault Levels

No

Are Constraints
Breached?

Yes

Remove Last Generator Increment

Repeat for All Remaining Generator Types

Repeat for All Remaining Load Levels

Finish

Fig. 2. Solution algorithm for optimal placement and sizing.

 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408 403

3.3. Numerical study

In order to evaluate the effects of adding DG to a system,

the operating characteristics of the distributed generator
should be taken into account, especially with respect to the
network interface, whether it is synchronous or induction
machine based, directly coupled or connected via an inverter.
The vast majority of distributed generation in the UK serves
the single purpose of exporting power to the power system
(subject to not causing malfunction of regulation or
protection devices) for consumption [17]. Little if any
distributed generation is either controllable or dispatchable
Fig. 3. Load profile for test feeder.
and is not used for any ancillary support. This, coupled with
the regulatory and commercial arrangements for DGs means
a single iteration. The solution is reached when the next that there is no benefit or incentive for the generator to
iteration fails to satisfy one or more constraints. The loading operate at a power factor other than unity (or as close to unity
of a system plays an important role in determining the type as possible). This analysis considers three generator power
of constraint to be violated. Consequently, the methodology factors of unity, 0.95 leading and 0.95 lagging to represent
and solution algorithm are applied over a range of loading likely operating generator characteristics.
conditions. Fig. 3 shows typical load profiles for three The methodology and solution algorithm are tested on
representative days for a test feeder, winter evening peak several distribution systems representing urban, rural and
demand, summer night least demand and spring daytime mixed use 11 kV networks. All networks are assessed for
average demand. These three levels are approximated by the ‘intact’ conditions (e.g. no contingencies). The results
ratio 4:2:1, exploited by the algorithm. shown here are for a 114 bus mixed urban and rural

Fig. 4. Test feeder schematic.

404 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408

Fig. 5. Normalised voltage sensitivity index (wZ0.5).

11 kV system from the UK. Fig. 4 shows a slightly modified are more sensitive to voltages than losses, such as cluster
version of the feeder obtained by aggregating many of 16–22 and 69–72. Within cluster 64–67 the ranking order
smaller (rural) loads located at lateral ends. Consequently, reverses with bus 64 at the lateral end being more sensitive
the number of buses is reduced to 75. Currently, there is no to voltage yet all four buses show similar loss sensitivities.
DG connected anywhere within the feeder. The total The results shown in Figs. 7 and 8 are obtained using LSI
maximum feeder load is 17.4 MW, with 13.8 MW concen- based solution algorithm in Fig. 2 for the three load
trated down the main feeder as housing estates and small conditions (winter, summer, spring) and three generator
industrial units supplied via 300–1000 kVA substations. power factors of 1, 0.95 leading and 0.95 lagging. In all
The two right hand laterals supply small housing clusters cases, a solution was reached when 20% penetration was
and farms via 10–100 kVA transformers and their maxi- achieved, i.e. a breach of constraint (11).
mum loads are 2.5 and 1.1 MW. Fig. 7 shows a noticeable improvement in voltage profile
Fig. 5 shows the VSI results of the optimal siting method along the feeder. Note that in Fig. 7, all voltage profiles have
with the highest ranked sites, as expected, being located been normalised so that the source voltage (node 1) is equal
towards the feeder extremities where nodal voltages are to the winter level of 1.04 pu. Despite only a 20%
lowest. Similarly, Fig. 6 shows the nodes’ LSI values. The penetration, there is roughly a 2% rise in voltage, even in
highest ranked nodes are again located towards the feeder buses on the main feeder geographically distant from the
extremities. Relating Figs. 5 and 6 to Fig. 4, a clustering generators. A similar pattern was observed for the spring
pattern can be clearly observed whereby neighbouring and summer load levels, although not shown. However,
nodes have very similar voltage and loss sensitivities. voltage rises are still moderate enough so as not to cause
Hence, simply placing the DG units at the highest ranked overvoltage problems in any load state. In each case, the
sites would lead to an undesirable cluster of neighbouring leading generator power factor gives rise to the greatest
generators. To avoid this situation and to distinguish voltage improvement and the lagging to the least. Fig. 9
between feasible and non-feasible sites, a cutoff value of compares the relative improvement in voltage profile for
0.015 was chosen for both the VSI and LSI. Of all the 20% penetration with generators placed by both LSI and
feasible sites identified by VSI, 10 were chosen as viable VSI. Over the bulk of the feeder, i.e. the large main section,
generator sites with respect to their spatial distribution there is no difference in voltage improvement. However, on
around the feeder as indicated in Fig. 5 and marked in Fig. 4. the laterals, there is generally a greater voltage lift from
The loss sensitivities in Fig. 6 indicate an even stronger placement by LSI. This is due to the LSI optimal generators
clustering pattern and the eight most sensitive sites being placed at fewer sites and sized larger. The losses
identified as optimal are chosen as to coincide with the results of Fig. 8 all follow the same pattern for each load
VSI-based sites. It is worth noting that within clusters buses state of approximately a 40% reduction in losses with a 20%

Fig. 6. Normalised loss sensitivity index (wZ0.5).

 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408 405

Fig. 7. Voltage profile for feeder with LSI placed DG and winter load. Fig. 9. Voltage profile for unity power factor and winter load.

penetration and winter load. (Note that the efficiency is 1— less sensitive nodes, the voltage rise or loss reduction will be
the ratio of real power losses to real power input. Real power less significant, although line flows and fault levels may
input is the sum of real power losses and real power loads.) become limiting factors. For system planners and operators,
A 40% reduction in losses gives rise to a 2.3% improvement this conservative estimate is likely to be of much greater
in efficiency. With a summer load, a 39% reduction in losses value than a ‘best case scenario’ value showing that more
translates to only a 0.5% rise in efficiency. Although results DG could be accepted, but with greater restrictions on its
are only shown for a generator power factor of 1, similar placement.
results were seen for the other cases, with a leading power
factor leading to most significant loss reduction.
The network performance is also analyzed for VSI-based
4. Genetic algorithm for recloser placement
optimal placement. As expected, due to similarity in optimal
sites, the reduction in losses and hence improvement in
In a DG-enhanced feeder, the optimization of reclosers is
system efficiency were virtually identical to those shown in not as straightforward as in the case of a conventional
Fig. 8. The voltage profile comparison illustrated in Fig. 9 feeder, due to the presence of additional generators, which
shows significant improvement in voltages along the feeder may be able to satisfy portions of the feeder load after fault
but no major differences arising from siting methods. has been isolated, as shown in the simple example presented
Although not included, a set of results for fault levels was above in Fig. 1. In a large, meshed network, the task of
obtained. These indicated an average rise of 7.9% with no locating optimal recloser positions that would create
line experiencing more than a 20% increase. possible self-supporting islands is not trivial. The optimal
The case study yields a set of numerical results for recloser position(s) depend on the types, locations and sizes
optimally placed DG that the system can accept under of distributed generators deployed at the feeder. Conversely,
particular load and generator power factor conditions. These if the reclosers are already placed on the feeder, optimal DG
represent a conservative estimate with DG added to the most positions and sizes can also be determined. Finally, both the
sensitive nodes. If the same quantity of DG were added to placement of reclosers and DG can be optimized concur-
rently during the planning stage of the feeder design. In this
study, a simple genetic algorithm (GA) is proposed, based
on the algorithm presented in [18], to solve for optimal
recloser positions for a given DG allocation. The objective
is to minimize the composite reliability index (CRI),
described in (1) with default weights W (0.2, 0.4, 0.4) and
targets WT (1, 2.2, 7) for SAIFI, SAIDI and MAIFIe,
respectively. The target values for SAIFI, SAIDI and MAIFI
of (1, 2.2, 7) indicate satisfactory level of reliability for a
conventional feeder used as a numerical example, and may
be exceeded in a DG-enhanced feeder yielding negative
values for the composite index CRI. The weights for SAIFI,
SAIDI and MAIFI of (0.2, 0.4, 0.4) reflect the importance
given to the momentary outages, as well as to the durations
of sustained outages. It is worth mentioning that the
Fig. 8. Real power losses for unity power factor DG. algorithm can be readily extended for the remaining two
406 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408

problems by adjusting the optimization function and xi Z ð1 K f ðgÞÞxi C bf ðgÞ (15)

algorithm properties.
Genetic algorithms are suitable candidates for such where b is either the lower or upper bound of the parameter
optimization problems, due to the nature of the optimization space, with probability of 0.5 of choosing each value, and the
function. The search space is spanned by mimicking the non-uniform operator f(g) is defined as
natural principles of reproductive evolution. Starting from an   g t
initial population of individuals, GAs effectively implement f ðgÞ Z r 1 K (16)
G
the ‘survival of the fittest’ strategy—fitter individuals (those
with higher values of the optimization function) are more where r is U(0,1), g current generation number, G maximum
likely to reproduce and/or survive to the next generation, thus number of iterations, and t is the tuning parameter. Multi-
improving the overall population. The population evolves non-uniform mutation applies the single non-uniform
using the genetic operators, selection, mutation and cross- mutation to all parameters of the individual. All three
over. For selecting the individuals chosen to continue into the mutation operators were applied to 5% of the population per
next generation, a normalised geometric ranking selection generation. The algorithm is initialized with randomly
procedure is used. The parameter vector X contains the generated started starting population.
positions of individual reclosers to be placed on the feeder. The test feeder was used to demonstrate the algorithm,
The composite reliability index is chosen as the optimization with DG placed at the optimal nodes from the voltage
function and the probability of selecting each individual is sensitivity analysis—see Fig. 4. It is assumed that the fault
given by (14) incidence rate and the duration of faults (damage restoration
time, DRT) are uniform over all feeder branches. Also, in
qð1 K qÞrK1 the case of the fault, only the minimum number of reclosers
Pðxi Þ Z (14)
1 K ð1 K qÞp closest to the fault should operate, and isolate the fault. The
actual implementation of the control algorithm required for
where q is the probability of selecting the best individual such operation is not the subject of this paper. For each
(given in advance), r is the rank of the individual after sorting recloser configuration, a composite index is calculated by
(1 being the best),and P is the population size. Additionally, determining the reliability zones (zones bounded by the
the elitist model is implemented, with the best individual reclosers), simulating the faults in those zones, determining
always included in the next generation. In the crossover step, the online and offline loads, and finally calculating the
a combination of arithmetic and heuristic operators is composite index. For each reliability zone that has a DG,
employed. Arithmetic crossover provides random mixture after a fault in other zones, the maximum output of all zone
of individuals, while the heuristic operator incorporates generators is compared with the load duration curve for
fitness information, resulting in individuals more likely to zone loads, and the number of faults is reduced by the
have higher values of the objective function. Both crossover percentage of time that the zone generation exceeds zone
operators are applied to 10% of individuals randomly chosen load. For example, if a zone’s generation corresponds to the
from the population in each generation. Three mutation 70% on the zone’s load duration curve (i.e. for 70% of the
operators are used: single uniform, single non-uniform and time generation is higher than load), it is assumed that 70%
multi-non-uniform mutation. Single uniform mutation of the faults outside the zone will not lead to outages within
changes only one of the parameters resulting in a recloser the zone.
being placed randomly on the feeder. Single non-uniform The results of the algorithm are presented in Table 1,
mutation replaces one of the parameters, xi, according to (15) which shows the top three values for the composite index,
Table 1
The composite reliability index for DG enhanced feeder and various recloser placement strategies

Number of reclosers Without DG With DG

Index value Recloser positions Index value Recloser positions
1 0.1862 15–22 0.1351 15–22
0.5855 14–15 0.4717 14–15
0.6063 14–69 0.4840 14–69
2 0.0672 15–22, 42–55 0.0037 12–68, 15–22
0.0677 15–22, 4–55 0.0066 14–15, 15–16
0.0794 14–15, 15–16 0.0212 13–69, 15–16
3 K0.0397 42–55, 14–15, 15–16 K0.1648 2–53, 15–22 46–75
K0.0391 4–55, 14–15, 15–16 K0.1610 2–22, 15–22, 46–75
K0.0276 42–55, 15–16, 15–22 K0.1534 2–53, 15–22, 27–58
4 K0.2519 2–53, 14–15, 15–16, 46–75 K0.2962 2–53, 12–68, 15–22, 46–75
K0.2494 2–22, 14–15, 15–16, 46–75 K0.2923 2–22, 12–68, 15–22, 46–75
K0.2429 2–53, 14–15, 15–16, 27–58 K0.2847 2–53, 12–68, 15–22, 27–58
 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408 407

and corresponding branches at which reclosers are placed, Optimal sites for placing distributed generation are
when up to four reclosers are strategically placed on the identified using both voltage sensitivity and loss sensitivity
feeder. The branch numbering corresponds with numbering analysis of power flow equations. Due to potential
shown in Fig. 4 Two cases are considered: the feeder limitations to the choice of sites, the optimal placement is
without DG, and with 20% DG penetration with the 10 not likely to be applied in practice. Instead, the paper
generators sited by the VSI. In the one recloser case, the illustrates how the information on the sensitivity pattern
recloser placement is dominated by the ‘conventional’ along the feeder expressed via clearly observed clusters of
benefits obtained by placing the recloser towards the middle buses with similar sensitivity values can be used to define a
of the feeder. The additional benefits, obtained by reducing set of viable yet practically suitable sites. With power-
the number and duration of outages during islanded constrained distributed generators placed at viable (near-
operation, do not justify placing a recloser at a different optimal) sites, the improvement in overall reliability is
location. demonstrated by finding optimal recloser positions using a
In the case with two reclosers, note that the optimal specially tailored genetic algorithm.
recloser positions differ significantly. Without DG, Currently, the analysis has been carried out for a variety
reclosers are optimally placed at branches 15–22 and of load conditions and generator power factors only. Further
42–55, isolating two portions of the feeder downstream work is planned to expand this scope to determine and
from buses 15 and 55, and allowing the remaining quantify benefits of scenarios with energy constrained
customers to continue receiving service even after the sources, variable (e.g. renewable) types of distributed
fault in the isolated areas. In a DG-enhanced feeder, generation and storage systems that will require represen-
reclosers are concentrated closer to the DGs, creating tation of time varying load and generator characteristics
islands of supply for customers downstream from bus 15. along with their uncertainties, control capability and
Similar to the one-recloser case, for a fault upstream from associated storage needs.
bus 15, the whole portion of the feeder downstream from
bus 15 may operate as an island. The placement of the
second recloser at the branch 12–68 creates additional
possible island for customers downstream from bus 12; References
they may now remain on-line even after a fault between
reclosers. As a result, the reliability index drops to 0.0037, [1] Alvarado F. Locational aspects of distributed generation. In:
Proceedings of the IEEE power engineering society summer meeting;
as compared to the 0.0672 in the case without DG. Note 2001.
that if reclosers are placed at buses 15–22 and 42–55 [2] McDermott TE, Dugan RC. Distributed generation impact on
(optimal placement for a feeder without DG), the index in reliability and power quality indices. In: Proceedings of the rural
the case with DG would be 0.0247. electric power conference, vol. D3; 2002. p. 1–7.
A similar trend continues for the cases with more than [3] Kojovic LA, Willoughby RD. Integration of distributed generation in
a typical USA distribution system. In: Proceedings of the 16th
two reclosers. In some cases, the composite index becomes CIRED, vol. 4; 2001. p. 5.
negative, indicating the target values for reliability indices [4] Brown R. Modeling the reliability impact of distributed generation.
have been exceeded. This is because target values for In: Proceedings of the IEEE power engineering society summer
reliability indices used in the definition of the composite meeting, vol. 1; 2002. p. 442–6.
index in (1) represent sufficient level of reliability for a [5] Begovic M, Pregelj A, Rohatgi A, Novosel D. Impact of renewable
distributed generation on power systems. In: Proceedings of the 34th
conventional distribution network. Hawaii international conference on system science; 2001.
[6] Carpinelli G, Celli G, Pilo F, Russo A. Distributed generation siting
and sizing under uncertainty. In: Proceedings of the IEEE Porto power
5. Conclusions tech conference; 2001.
[7] Celli G, et al. Probabilistic optimization of MV distribution network
in presence of distributed generation. In: Proceedings of the 14th
This paper has presented a methodology for optimiz- power system computation conferences, Sydney; 2002.
ing and coordinating the placement of distributed [8] Rosehart W, Nowicki E. Optimal placement of distributed generation.
generators and reclosers in a security constrained In: Proceedings of the14th power system computation conference;
distribution network. A systematic and rational placement 2002.
of distributed generation and reclosers is shown to be [9] Griffin T, Tomsovic K, Secrest D, Law A. Placement of dispersed
generations systems for reduced losses. In: Proceedings of the 35th
able to improve both system security and reliability, by Hawaii international conference on system science; 2002.
improving feeder voltage profile, reducing losses and [10] Ippolito M, Morana G, Riva Sanseverino E, Vuinovich F. Risk based
increasing efficiency, and providing energy to some of optimization for strategical planning of electrical distribution systems
the customers, even after the fault in the distribution with dispersed generation. In: Proceedings of the IEEE bologna
system. The level of improvement depends on the type, powertech conference; 2003.
[11] Celli G, Ghiani E, Mocci S, Pilo F. A multi-objective function for the
number and size of the distributed generators, number of optimal sizing and siting of embedded generation in distribution
reclosers, and positions of both generators and reclosers networks. In: Proceedings of the IEEE bologna powertech conference;
on the feeder. 2003.
408 D.H. Popović et al. / Electrical Power and Energy Systems 27 (2005) 398–408

[12] Mori H, Iimura Y. Application for the parallel tabu search to [15] Refaey WM, Ghandalky AA, Azzoz M, Khalifa I, Abdalla O. A
distribution network expansion planning with distributed gener- systematic sensitivity approach for optimal reactive power planning.
ation. In: Proceedings of the IEEE bologna powertech conference; In: Proceedings of the 22nd annual north American power
2003. symposium; 1990. p. 283–92.
[13] Venkataramana A, Carr J, Ramshaw RS. Optimal reactive power [16] Baran ME, Wu FF. Optimal capacitor placement on radial distribution
allocation. IEEE Trans Power Syst 1984;138–44. systems. IEEE Trans Power Deliv 1989;4(1):725–32.
[14] Masoum MAS, Ladjevardi M, Fuchs EF, Grady WM. Optimal [17] Dale L. Distributed generation and transmission. In: Proceedings of the
placement and sizing of fixed and switched capacitor banks IEEE power engineering society winter meeting, vol. 1; 2002. p. 132–34.
under nonsinusoidal operating conditions. In: Proceedings of the [18] Houck CR, Joines JA, Kay MG. A genetic algorithm for function
IEEE power engineering society summer meeting, vol. 2; 2002. p. optimization: a matlab implementation, Technical Report, NCSU-IE; 1995.
807–13. [19] Ofgem/DTI (2003) www.distributed-generation.org.uk