Você está na página 1de 11


5, SEPTEMBER 2014 2339

Optimal ESS Allocation and Load Shedding for

Improving Distribution System Reliability
Ahmed S. A. Awad, Student Member, IEEE, Tarek H. M. EL-Fouly, Member, IEEE, and
Magdy M. A. Salama, Fellow, IEEE

Abstract—The recent deployment of distributed generation has network outages. Consequently, successful islanding operation
led to a revolution in the use of distribution systems and the emer- improves system reliability by preventing loss of load or by min-
gence of smart-grid concepts. Smart grids are intended primarily imizing the loss of energy supplied to non-affected customers
as a means of facilitating the integration of renewable energy
sources and of achieving greater system reliability and efficiency.
during network disturbances. Thus, when a disturbance occurs,
Energy storage systems offer a number of benefits that can help the formation of islands may help improve system reliability if
utilities move toward those goals. One of those benefits is the distributed energy resources are available and are able to op-
capacity to improve system reliability through successful islanding erate in islanded mode. However, because of the stochastic na-
operation. This paper proposes a methodology for the cost-ef- ture of the power generated from renewable-based customer-
fective improvement of system reliability through the allocation owned distributed generation (DG), e.g., wind turbines and pho-
of distributed storage units in distribution systems. The costs of
energy storage installation are optimized with respect to reliability tovoltaic (PV) arrays, distribution utilities cannot rely solely on
value expressed as customers’ willingness to pay in order to avoid such sources as a means of improving system reliability. They
power interruptions. The primary goal of this research was thus to might utilize distributed storage (DS) 1 units as a backup source
determine the optimal combination of storage units to be installed for addressing network disturbances. The primary challenge in
and the loads to be shed so that all possible contingencies can the introduction of this concept of non-volatile distribution sys-
be effectively addressed. A probabilistic approach is therefore
tems is the high installation cost associated with DS, which
adopted that includes consideration of the stochastic nature of
system components. means that, to minimize installation costs and maximize the as-
sociated improvement in reliability, distribution utilities must
Index Terms—Distributed generation, energy storage systems,
calculate the optimal size of the DS units to be installed.
islanding, load shedding, smart grid, willingness to pay.
Several research works have addressed the problem of op-
timal allocation of DERs to enhance the reliability of distribu-
I. INTRODUCTION tion systems as in [3], [4] . As well, DG units and circuit re-
closers are allocated in order to minimize a composite relia-

P OWER SYSTEMS are now evolving from the conven-

tional regulated system, with centralized generation con-
nected to the transmission networks, to a deregulated structure
bility index in [5]. The results proved that the more reclosers
and the higher DER sizes used, the less reliability index can
be obtained. However, the authors did not consider the cost ef-
that allows small generators to be connected directly to the dis- fectiveness of the solution since adding more reclosers and/or
tribution networks. Such networks thus become active and are DG units may be much expensive as compared to the higher
usually referred to as active distribution networks, in which new reliability achieved. The feasibility of battery storage plants to
technologies should facilitate adaptation to such active environ- enhance the power system reliability has been discussed in [6].
ments and enable the use of smart-grid concepts. Energy storage In [7], a technique was further proposed for evaluating the im-
systems (ESSs) are one promising technology that can support provement of power system reliability using ESS without in-
the incorporation of smart grids because of their capacity to en- cluding cost benefit analysis that could justify the installation
able successful islanding and to facilitate the integration of high of ESS for such application.
penetration levels of renewable energy sources (RESs). ESSs The evaluation of the reliability of ESSs in distribution sys-
can also provide additional benefits for distribution utilities, tems has been previously addressed in the literature [9], [8],
such as an efficient expansion alternative, demand side manage- [10]. In [9], a mathematical formulation was derived for the
ment, and methods of mitigating power quality issues [1]. sizing of backup energy storage in order to meet specific reli-
According to [2], distributed energy resources (DERs) can ability targets for critical customers, but since the formulas de-
provide power to the system loads during planned or un-planned veloped did not correspond to the network model, the impact of
the location of the storage was not considered, nor did the au-
Manuscript received April 11, 2013; revised July 02, 2013; accepted April thors optimize the economic benefit of achieving the reliability
05, 2014. This work was supported by the Government of Canada through the target relative to the storage cost.
Program on Energy Research and Development. Date of publication May 09,
2014; date of current version September 05, 2014. Paper no. TSG-00285-2013. The authors in [8] compared the reliability and economic
A. S. A. Awad and M. M. A. Salama are with Electrical and Computer En- benefits of two different control strategies for energy storage in
gineering Department, University of Waterloo, ON N2L 3G1, Canada (e-mail: distribution systems: standby and model predictive controllers
asamir@uwaterloo.ca; msalama@uwaterloo.ca).
T. H. M. EL-Fouly is with CanmetENERGY, Natural Resources Canada, (MPCs). With a standby strategy, during islanding, an ESS is
Varennes, QC J3X 1S6, Canada (e-mail: telfouly@nrcan.gc.ca). 1The terms ESS and DS are used interchangeably in this paper.
Digital Object Identifier 10.1109/TSG.2014.2316197

1949-3053 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

assumed to release its stored energy in order to supply iso- Based on the above survey, it can be concluded that sufficient
lated nodes, thus eliminating loss of load or minimizing loss work has been conducted with respect to evaluating the relia-
of energy for the isolated customers. Once the network is re- bility of ESSs in distribution systems and to assessing the ad-
stored following a disturbance, the ESS is immediately charged equacy of power systems with large penetration levels of wind
and put on standby in preparation for the next disturbance. The energy. However, the literature reveals that the problem of im-
drawback of that proposal is that the charging of the ESS was proving system reliability by ascertaining the most cost-effec-
not included in the analysis, which was also based on the as- tive siting and sizing of DS units in distribution networks has not
sumption that it can be fully charged between any two suc- yet been addressed. The objective of the work presented in this
cessive failures. However, the charging cycle of the ESS must paper was therefore to improve the reliability of a distribution
be considered in order to determine whether it can be fully system through the development of a reliability evaluation tech-
charged during acceptable operating states that do not violate nique combined with an optimization approach for allocating
system constraints. On the other hand, an MPC-based strategy DS units.
implies the forecasting of load demands and electricity prices, In this research, a value-based reliability approach was
followed by a determination of the optimal charging and dis- adopted as a means of improving distribution system relia-
charging power levels that enable the utility to benefit from bility from an economic perspective. In practice, distribution
the electricity price difference between off-peak and peak pe- utilities set arbitrary targets rather than obligatory standards as
riods. This latter strategy is implemented in normal operation; objectives of their reliability indices [17]. The targets usually
however, during islanding, the proposed strategy behaves in a depend on the utilities’ perception of customer tolerance levels
manner similar to that of the standby approach. The authors with respect to interruptions. However, the expensive invest-
therefore concluded that standby control provides greater re- ments that characterize the planning stage mean that planning
liability than the MPC-based approach, but at the expense of decisions should not rely on such rule of thumb criteria for
more costly energy drawn from the substation. reducing the costs associated with customer interruption. In
As reported in the literature, the reliability of ESSs with other words, achieving those arbitrary targets may cost the
wind energy systems has been extensively evaluated [11]–[13]. distribution utilities much more than the customers would
In [11], the authors presented an energy storage model for actually pay as interruption costs. In such a case, the reliability
assessing the adequacy of wind farm ESSs. A number of target would be overestimated and would result in unnecessary
operating strategies were discussed and compared based on an extra costs. As well, in most cases, no reward/penalty system
evaluation of the reliability benefit derived from the energy forces distribution utilities to target specific reliability levels
storage obtained with each strategy. despite the willingness of some customers to pay more for
From another perspective, the authors in [14] presented a sys- greater reliability. Willingness to pay (WTP) therefore repre-
tematic approach for clustering distribution systems into virtual sents the reliability value that utilities might lose if they fail
microgrids based on minimizing energy flows between them. to achieve the desired reliability level for those customers. It
The impact of allocating pre-specified amount of DS units and is consequently crucial to apply optimization approaches that
distributed reactive sources on maximizing the self-adequacy include economic considerations in order to determine the
of formed microgrids was further studied in that paper. More- optimal investment plan as well as the optimal reliability level.
over, the authors in [15] considered optimizing the operation The main contributions of this paper can be summarized as
stage through some control variables, including load shedding follows:
at each bus, but without representing the objective function as • The paper presents a methodology that considers the WTP
monetary value. of customers in determining the most cost-effective siting
The literature includes numerous accounts of the use of time and sizing of DS units in distribution networks.
series models for modeling the stochastic nature of a variety • The approach proposed includes determining the load
of components, such as RES-based DG and load demand, e.g., points to be shed, during contingencies, which minimizes
[11], [16] . Such work usually adopts one year-ahead forecasting the total interruption cost via increasing the probability of
for all loads and generation units. The forecasted profiles are successful islanding operation.
then used for determining the charging/discharging cycle of the • Unlike previous work that applied time series patterns for
ESS at each hour, based on which the size of the ESS is op- optimizing the size of DS units, a probabilistic approach is
timized or the adequacy of the power systems with the ESS proposed in this paper in order to consider the uncertainty
is assessed. Despite the difficulties associated with forecasting of system components.
highly stochastic components, such as wind speed and solar irra- The remainder of this paper is organized in five sections.
diance, the application of time series models in planning studies Section II describes the problem under study. The methodology
provides an optimal solution that is valid only for the time series and the mathematical formulation are explained in Section III.
pattern that is applied. Consequently, the solution obtained is not Section IV introduces a sample case study. The results and con-
guaranteed to be the global optimal for other possible patterns. clusions are then presented in Sections V and VI, respectively.
A preferable solution would therefore be to derive probabilistic
models that take into account all possible system states, as pro- II. PROBLEM DESCRIPTION
posed in this paper. Instead of the usual time series models, any Based on the above discussions, the rationale behind the work
available historical data is used in order to generate probability presented in this paper is the optimization of the investment
distribution functions (PDFs), which are then applied in a prob- costs associated with DS installation relative to the reliability
abilistic approach, as discussed in Section III. value expressed as the customers’ WTP. In reliability based

planning studies as in [18], all load points in a given island are and time-intensive to be solved through deterministic methods.
usually shed during unsuccessful islanding operation, i.e., when The main step in the GA is chromosome encoding; each solu-
the total generation is insufficient to supply the isolated loads. tion (chromosome) consists of integer variables that represent
This practice has been proposed due to the difficulty of deter- the discrete size of DS units and discrete portion of load-shed-
mining the optimal load points to be shed, for each contingency ding decision variables at each bus.
and every state, in order to guarantee successful islanding oper- For every population generated by the GA, the first stage im-
ation. plies a contingency analysis of the distribution system in order
In this work, the goal was thus to determine the optimum to determine how much power is required from each allocated
combination of DS units to be installed and the loads to be shed DS unit in order to supply the demand power required for all
for the most cost-effective improvement in distribution system possible island formations. The next stage utilizes a sequential
reliability. The proposed problem then includes some planning Monte Carlo simulation (MCS) for an evaluation of the relia-
decisions, such as determining the sizes and locations of DS bility of the distribution system through the estimation of the
units to be installed, as well as other contingency planning de- expected energy not supplied (EENS) and the interruption cost
cisions, such as the load points to be shed during contingen- (ECOST) indices [25]. It is worth mentioning that the ECOST
cies. The contingency planning decisions basically aim to in- index is used primarily so that, as discussed earlier, considera-
crease the overall probability of successful islanding operation, tion of the reliability value can be included in the evaluation of
and thus to minimize the total interruption cost. It is worth men- the objective function; the EENS is utilized as a measure of the
tioning that the exact (optimal) amount of load shedding and the improvement in the reliability. All solutions (chromosomes) of
corresponding load points to be shed would be determined in the each population are then evaluated by means of the fitness (ob-
operational stage, and thus they are out of scope of this paper. jective) function. Finally, a new population is generated, and the
Regarding the customers’ WTP, because of the difficulty of entire process is repeated until the stopping criterion is met. The
determining the price that customers would pay for reliability, mathematical formulation is explained in the following subsec-
in reliability assessment studies, the interruption costs usually tions.
adopted are a reflection of customers’ WTP in order to receive
the reliability level required [19]. Interruption costs generally A. First Stage
depend on the customer classification, e.g., residential, indus- In the first stage, a probabilistic approach takes into account
trial, or commercial, and on the characteristics of the interrup- the stochastic nature of all DG sources and load demands. Avail-
tion, e.g., duration and frequency. Since the mid-1980s, nu- able historical data are utilized so that each component is repre-
merous surveys have been conducted by US and Canadian util- sented by a specific PDF. Continuous PDFs are further divided
ities as a means of estimating the price at which the load would into several states with associated probabilities, thus creating a
be curtailed in an effort to address a contingency state (i.e., in- probabilistic model for every component. The number of states
terruption cost). In the case of commercial/industrial customers, for each component should be carefully selected so that the sim-
this cost basically depends on the revenue/production lost; for plicity and accuracy of the analysis are not compromised: a large
residential customers, it is their WTP to avoid service interrup- number of states increases accuracy but at the expense of also
tion. One of those surveys is the Canadian survey conducted in adding to the complexity, and a small number of states has the
order to estimate customer damage functions (CDFNs), which opposite effect. A combined load-DG model can then be gener-
represent the interruption cost as a function of the duration of ated by convolving all of the individual probabilistic models as-
the interruption [20]. In 2008, the US department of energy suming that these individual models of the load and DG sources
funded several studies targeted at estimating interruption costs, are independent (uncorrelated) as in [14], [26] . Such a model
or CDFNs, for a variety of customer categories, as shown in combines all possible operating states for the available DG units
Fig. 1 [21]. In this figure, the CDFN is represented as interrup- and the different load levels. The total number of states is there-
tion cost per unserved kWh. fore equal to the product of the number of states for each com-
As mentioned, the first stage performs an contingency
analysis that considers the failure of every single line in the dis-
This section presents the general methodology adopted for tribution system. When such a disturbance occurs, the protec-
the work presented in this paper. A two-stage model is pro- tion system isolates the faulty section so as to ensure the healthy
posed, which is then combined with a genetic algorithm (GA) operation of the rest of the system. This practice results in the
for minimizing the objective function under study. Evolutionary formation of islands if DG and/or DS units are available and ca-
optimization algorithms, e.g., GA, particle swarm optimization, pable of supplying the load demand in those islands. The contin-
Tabu search, etc., are emerging as efficient optimization tech- gency analysis implies the solving of the load flow (1) to (7) for
niques to solve complicated problems such as DG planning [22] all possible islands in order to determine the power requirements
and unit commitment [23]. GA has been extensively used in from every DS unit for all combined load-DG states, taking into
the literature as in [5], [22] , and it has showed superior per- account the load-shedding decision variables at each bus. It is
formance compared to other meta-heuristic techniques in terms worth mentioning that the power values calculated in (6)–(7)
of the solution error and the execution time [24]. Therefore, GA may be less or more the DS power size, thus leading to suc-
has been utilized to solve the proposed problem that is classi- cessful or unsuccessful islanding operation, respectively, as will
fied as mixed integer non-linear programming, which is hard be explained later on in the second stage.

given in (8) to (16). Constraints (8) to (10) are added to the ob-
jective function (24) using penalty functions (terms), according
to which every term equals zero if the corresponding constraint
is satisfied or equals a large positive value otherwise.
This study is based on a number of islanded operating strate-
• Only dispatchable DGs are allowed to supply reactive
power so that their terminal voltages can thus be regu-
lated. Other types of DG should be provided with VAR
compensators, e.g., capacitor banks, in order to support
their operation in islanded mode. Dispatchable DGs are
therefore modeled as P-V nodes, while other DGs are
treated as P-Q buses.
• Charging DS units in islanded mode is not permitted, and
all DGs are hence assumed to be controllable in order to
avoid any excess generation.
1) Load Flow Equations:
Fig. 1. Customer damage function (CDFN) for a variety of customers. — For no DS installed at bus during contingency and state


— For DS installed at bus during contingency and state :



Fig. 2. Flowchart of the first stage.

All system buses are modeled as either P-Q or P-V nodes (6)
except the nodes that are connected to DS units. In every is-
land, only one DS unit should act as a slack (reference) bus for
all other nodes in that island, and any other available DS units
should be treated as voltage-frequency controlled buses, as de-
scribed in [27]. The latter bus type implies that two additional
equations that represent the droop characteristics be added to (7)
the basic load flow equations. For the sake of simplicity, it is
2) Voltage Limits:
assumed that all droop controller parameters are identical and
that the terminal voltage of each DS unit is set at one per unit. (8)
Applying these assumptions results in equal net active power
generated from every DS node, as calculated in (5). Fig. 2 sum-
marizes the flowchart of the first stage. 3) Line Flow Limits:
System constraints, i.e., nodal voltages and the power flow
limits of the lines, along with DS size constraints, are further (10)

4) DS Size Constraints: determine whether it operates in normal (grid connected) mode

or in islanded mode. During the latter mode, every DS unit in
(11) the island supplies the demand power only if the energy stored
(12) is sufficient and the power required is less than the rated DS
power. The power required from every DS unit at each hour is
recalled from the contingency analysis that corresponds to the
(14) combined load-DG state at that hour. If the energy stored is in-
sufficient or the power required is larger than the rated DS power
5) Load Shedding Constraints:
at a specific hour, islanding operation fails, and all loads must
be disconnected during that hour. In addition, if no DS units are
allocated in a given island, the load-shedding variables are ad-
(16) justed at each hour, based on the total generation, total demand,
and power losses ( ) in that island. The assumption is there-
where and are the system bus indices; is the faulty line (con-
fore that the islanding operation may fail if the total generation is
tingency) index; is the combined load-DG state index; is the
insufficient to satisfy the load demand plus losses for a particular
total number of system buses; and are the DS output
scenario, and hence all loads must be disconnected [18]. When
active and reactive power variables, respectively; and
the system is restored, DS units are permitted to be charged up
are the demand active and reactive powers, respectively; and
to their maximum power rates as long as system constraints are
are the generated active and reactive powers, respectively;
not violated. Each study period starts with all DS units fully
is the set of system buses equipped with DS units for con-
charged; however, the characteristic equations that govern the
tingency ; is the portion of load to be shed that is equal
energy stored in each DS unit at a given hour ( ) are as follows:
to 1 if the load is totally shed; and are the voltage mag-
nitude and the phase angle, respectively; and are the bus
admittance matrix element magnitude and angle, respectively; (19)
is the line flow current; is the DS power size; is the (20)
DS energy size; , , and are the integer decision variables (21)
controlling the size of DS to be installed and portion of load to
be shed at bus , respectively; CB is the set of candidate buses where the superscripts refer to charging and dis-
for DS installation. charging, respectively; and is the round-trip efficiency of
the DS.
B. Second Stage The process is then repeated for several periods (samples)
until the ratio of the standard deviation of the sample mean of the
After the power requirements of each DS unit are determined,
reliability index of interest to the sample mean of the same index
a sequential MCS is performed for the system. The basic idea
becomes less than a predetermined tolerance. This process is
of the MCS is to generate an artificial operating history for the
summarized in Fig. 3. The energy not supplied (ENS) per year is
system under study. To that end, the probabilistic model for
calculated by summing the energy interrupted at all nodes over
every component (i.e., DG and load demand) is utilized to de-
the entire study period, as calculated in (22). The interruption
rive the corresponding cumulative distribution function (CDF).
costs (COST) per year are also computed as in (23) by utilizing
The components’ states can be then generated through the gen-
the CDFN shown in Fig. 1, which is basically a function of the
eration of a uniformly distributed random number between 0
duration of the interruption ( ). The ENS and COST are then
and 1, and rounding it to the nearest cumulative probability in
averaged over the total number of periods in order to estimate
the corresponding CDF.
the expected values for each index: the EENS and the ECOST.
As well, system feeders (lines) are represented using a
In the final step, the ECOST is utilized in order to evaluate the
two-state model (up state and down state), in which up and
objective function in (24).
repair times are calculated based on the generation of a random
number that follows an exponential distribution, as in (17) and
(18) [28].

where and are the up and repair times, in hours, re- (23)
spectively; MTTF: mean time to failure, in hours; MTTR: mean
time to repair, in hours; and : two uniformly distributed
random numbers between 0 and 1.
After a synthetic study period ( years) is generated for all (24)
system components, the study period is divided into segments
(hours). The system is then simulated hour by hour for the en- where is the interruption event index; is the time in hours of
tire study period. For each hour, the system state is checked to the interruption event ; , , and are, respectively, the

Fig. 3. Flowchart of the second stage.

annualized capital power and energy costs, and annual mainte- and system losses during those scenarios. It is also obvious that
nance cost of the DS. the DS installation costs are paid independently based on how
The above objective function minimizes the total cost com- many times faults occur per year. On the other hand, interrup-
prised of the annual installation and maintenance costs of the tion costs are paid annually according to fault incidence rates.
DS units in addition to the annual interruption costs. In this for- Therefore, for systems with low reliability that are characterized
mula, the fixed capital costs are annualized by dividing them by by numerous fault occurrences per year, DS integration is eco-
the present value function (PVF), which is expressed in terms nomically desirable as a means of minimizing the total annual
of the interest rate ( ), inflation rate ( ), and lifetime of the costs.
equipment ( ), as calculated in (25) and (26) [29].
(25) The system used for the case study is a 33-bus radial distri-
bution system, as shown in Fig. 4. The rated active and reactive
(26) power levels of the load points as well as the feeder data are
taken from [30]. The reliability parameters of the substation and
It should also be mentioned that the above objective func- the system feeders are further summarized in Table I [31].
tion inherently minimizes energy losses for all contingency sce- Two different DG types are assumed in the system under
narios because the DS units are sized to supply load demands study: dispatchable DG ( ) based on natural gas and inter-


Fig. 4. System under study.



mittent DG ( ) based on wind. and are placed at discrete sizes in steps of 100 kVA/kWh. Table III lists the an-
buses 33 and 18, respectively. is a 500 kVA synchronous nual capital and maintenance costs for the four candidate tech-
generator that operates at 500 kW (unity power factor) during nologies, based on an assumed 30-year life cycle. Depending on
grid connected mode and supplies the active and reactive land availability and/or utility regulations, the candidate buses
powers required during islanded mode. is a 1 MW wind for DS installation are assumed to be included in set CB: (16,
turbine with power curve parameters as shown in Table II. The 17, 21, 22, 25, and 32). The setting parameters of GA are fur-
sum of the rated DG power levels is confirmed as meeting ther given in Table IV.
the Hydro One capacity requirement that limits installed DG
ratings to 60% of the substation capacity plus the minimum
station load [32]. The wind speed data for the site under study This section summarizes the findings of this research, in
are assumed to reveal a mean wind speed of 6 m/s. Both wind which DS units are optimally allocated, with consideration
speed and wind turbine data are utilized in the development of all possible island formations, in order to improve system
of the probabilistic wind-based DG model, which is based on reliability. Two cases are compared in this section regarding the
the adoption of a Rayleigh PDF for modeling wind speeds, as load shedding scheme applied, namely binary and discrete load
presented in [26]. Load demands are also assumed to follow the shedding. It is worthwhile to mention that GA performs well
IEEE-reliability test system (RTS) discrete load probabilistic when the decisions variables are binary. Therefore, the binary
model explained in [26]. case has been firstly solved, and the results obtained have
For this case study, the following financial parameters are as- been used as initial population for the discrete case in order to
sumed: 5% interest rate and 1% inflation rate. The distribution accelerate the convergence of GA. Moreover, the impact of DG
system is assumed to contain a mix of 70% residential and 30% locations on the results obtained has been further investigated
small commercial and industrial customers. These percentages in the third case. Finally, the last case provides a sensitivity
are used for the estimation of the customers’ WTP through the analysis for the impact of the value of interruption cost on the
weighting of the corresponding CDFNs shown in Fig. 1. As optimal solutions.
well, the DS installation costs are subdivided into three main
parts: the capital power cost of the rotating machine/converter A. Binary Load Shedding
interface (in $/kVA), the capital energy cost of the storage ca- In this case, the load shedding decision variables ( ) are bi-
pacity (in $/kWh), and the annual fixed operation and mainte- nary (i.e., either 0 or 1). For each storage technology, the optimal
nance (O&M) costs (in $/kVA). A lead-acid (LA) battery, com- DS locations and sizes as well as the load points to be shed are
pressed air in vessels (CAS), a sodium-sulfur (Na/S) battery, presented in Table V. As can be seen, all technologies result in
and a vanadium redox (VR) battery were selected as candidate the same allocation of DS units and loads to be shed. Conse-
storage technologies because their power and discharge time quently, interruption costs are found to be the same, while the
capacities are suitable for the application under study. It is as- total cost varies based on the cost of each technology. How-
sumed that the candidate storage technologies are available in ever, these results are system dependent, which means that dif-


Fig. 5. Total annual costs for the base case and different storage technologies Fig. 7. Graphical representation of DS allocation and load shedding during a
(binary load shedding). contingency at the line between buses 1 and 2.

Fig. 8. Graphical representation of DS allocation and load shedding during a

contingency at the line between buses 7 and 8.
Fig. 6. EENS for the base case and the different storage technologies (binary
load shedding). buses 7 and 8. As can be seen, the allocated DS unit at bus 16
and only supply the formed island, while no load points
ferent results might be obtained if system parameters and costs have to be shed in this island.
are changed. Fig. 5 provides a comparison of the total annual After studying the impact of each technology on the total an-
costs for the different technologies relative to the base case (i.e., nual costs, another case study is conducted that investigates the
without DS integration). In the base case, the annual cost is com- allocation of combination of the four candidate DS technolo-
prised of only the costs associated with the interruptions during gies. This case study is implemented via adding an integer deci-
the contingencies. Fig. 5 reveals that LA batteries provide the sion variable that represents the type of DS unit to be installed.
least expensive solution: interruption costs were decreased from The results obtained are confirmed with those of LA batteries
$ 0.78 million (base case) to $ 0.275 million. Fig. 6 depicts the allocation given in Table V. Again, this conclusion depends on
reliability level measured in EENS for each storage technology the system parameters and different costs of DS units.
compared to the base case. Again, since the loads to be shed are
the same for each technology, the same EENS values are de- B. Discrete Load Shedding
picted in each case. In this case, the load shedding decision variables ( ) are dis-
As an example, Fig. 7 shows a graphical representation of LA crete values between 0 and 1 with step size of 0.1. Table VI
battery allocation during a failure at the line between buses 1 and shows the optimal type, location, and size of DS units, and por-
2. The two DS units allocated at buses 16 and 32 are utilized in tions of load points to be shed. The total annual cost is further
order to supply the island formed. Further, load points 22 and given in Table VI. Through comparing these results to the bi-
29 should be disconnected by the utility operator in a timely nary case, it can be concluded that discrete load shedding only
manner in order to avoid overloading the allocated DS units and reduces the interruption costs by almost $ 30 000. This reduction
to guarantee the continuity of the supply in the isolated system. is due to the advantage of discrete load shedding, i.e., flexible
Another example is shown in Fig. 8 for a disturbance between portion of load to be shed at each bus. This flexibility can be



noticed in the amount of load to be shed at bus 29, i.e., 90%,

compared to total load shedding in the binary case. However,
these results depend on the system data and rated power levels
of the load points.

C. Impact of DG Locations

and are re-allocated, in this case, to buses 25 and

22, respectively, in order to investigate the impact of DG loca-
tions. The optimal DS locations and sizes as well as the load
points to be shed are given in Table VII. As can be seen, no load
points have to be shed compared to the previous cases, which
implies that the probability of successful islanding operation is
not affected by disconnecting some load points. Moreover, nei-
ther the sizes nor the locations of the allocated DS units are af-
fected with the new DG locations. However, these results are Fig. 9. Graphical representation of DS allocation and load shedding during a
system dependent, thus they might be changed using different contingency at the line between buses 1 and 2 (new DG locations).
rated power levels of DGs and/or load points.
Moreover, Table VII compares the total annual cost of the op- OPTIMAL DS ALLOCATION AND LOAD SHEDDING (SENSITIVITY ANALYSIS)
timal solution with respect to the new base case, which corre-
sponds to the new DG locations. It is revealed that interruption
costs are increased in this case since the DG units are now closer
to the substation, and thus more load points are deprived from
their output power during contingencies beyond bus 3. Simi-
larly, EENS values are increased with the new DG locations, as
shown in Table VII.
A graphical representation of the new DG locations and LA
battery allocation is further illustrated in Fig. 9 during a failure
at the line between buses 1 and 2, as an example. As can be seen,
no load points have to be shed in this case compared to Fig. 7.

D. Impact of the Value of Interruption Cost

This case investigates the impact of the value of interruption

cost on the optimal size and location of DS units to be installed,
as well as the load points to be shed. Table VIII presents several
scenarios with different percentages of residential customers,
and small commercial and industrial customers. The optimal
solutions for each case are summarized in Table VIII. As ob-
served, the optimal DS allocation and load points to be shed are
duly dependent on the value of the interruption cost. The higher
percentage of commercial and industrial customers, which im-
plies higher interruption cost, the larger sizes of DS units are
Moreover, Figs. 10 and 11 compare the total annual costs and
the reliability level measured in EENS for the various cases. Fig. 10. Total annual costs for the various scenarios of different customers.

[6] W. R. Lachs and D. Sutanto, “Battery storage plant within large load
centres,” IEEE Trans. Power Syst., vol. 7, no. 2, pp. 762–767, May
[7] D. Aming, A. Rajapakse, T. Molinski, and E. Innes, “A technique for
evaluating the reliability improvement due to energy storage systems,”
in Proc. Can. Conf. Electr. Comput. Eng. (CCECE), Apr. 22–26, 2007,
pp. 413–416.
[8] X. Yixing and C. Singh, “Distribution systems reliability and economic
improvement with different electric energy storage control strategies,”
in Proc. IEEE Power Energy Soc. Gen. Meet., Detroit, MI, USA, Jul.
26–28, 2011, pp. 1–8.
[9] J. Mitra, “Reliability-based sizing of backup storage,” IEEE Trans.
Power Syst., vol. 25, no. 2, pp. 1198–1199, May 2010.
[10] X. Yixing and C. Singh, “Adequacy and economy analysis of distri-
bution systems integrated with electric energy storage and renewable
Fig. 11. EENS for the various scenarios of different customers. energy resources,” IEEE Trans. Power Syst., vol. 27, no. 4, pp.
2332–2341, Nov. 2012.
[11] P. Hu, R. Karki, and R. Billinton, “Reliability evaluation of gener-
VI. CONCLUSIONS ating systems containing wind power and energy storage,” IET Gener.,
Transm., Distrib., vol. 3, no. 8, pp. 783–791, Aug. 2009.
In this paper, a two-stage model is proposed for the allocation [12] F. A. Bhuiyan and A. Yazdani, “Reliability assessment of a wind-power
of DS units in distribution systems as a cost-effective means of system with integrated energy storage,” IET Renewable Power Gener.,
vol. 4, no. 3, pp. 211–220, May 2010.
improving system reliability. A value-based reliability approach [13] Z. Yi, Z. Songzhe, and A. A. Chowdhury, “Reliability modeling and
is adopted that includes consideration of customers’ WTP as control schemes of composite energy storage and wind generation
the reliability value benefit of improving system reliability. The system with adequate transmission upgrades,” IEEE Trans. Sustain.
Energy, vol. 2, no. 4, pp. 520–526, Oct. 2011.
total annual costs comprised of DS installation and maintenance [14] S. A. Arefifar, Y. A. I. Mohamed, and T. H. M. El-Fouly, “Supply-ad-
costs as well as the interruption costs are then minimized in equacy-based optimal construction of microgrids in smart distribution
order to determine the optimal combination of DS units to be systems,” IEEE Trans. Smart Grid, vol. 3, no. 3, pp. 1491–1502, Sep.
installed and the loads to be shed during all possible contin- [15] S. A. Arefifar, Y. A. R. I. Mohamed, and T. H. M. El-Fouly, “Compre-
gencies. For calculating the power requirements from the al- hensive operational planning framework for self-healing control ac-
located DS units, a probabilistic approach takes into account tions in smart distribution grids,” IEEE Trans. Power Syst., vol. 28, no.
4, pp. 4192–4200, Nov. 2013.
the stochastic nature of all of the system components: loads [16] Y. M. Atwa and E. F. El-Saadany, “Optimal allocation of ESS in dis-
and existing DGs. A sample case study is presented, and four tribution systems with a high penetration of wind energy,” IEEE Trans.
storage technologies are compared and measured against a base Power Syst., vol. 25, no. 4, pp. 1815–1822, Nov. 2010.
[17] A. A. Chowdhury and D. O. Koval, “Current practices and customer
case with no DS installation. The results show that integrating value-based distribution system reliability planning,” IEEE Trans. Ind.
DS units with distribution systems reduces the utilities’ annual Appl., vol. 40, no. 5, pp. 1174–1182, Sep.-Oct. 2004.
costs because of their capacity to enable successful islanding [18] Y. M. Atwa, E. F. El-Saadany, M. M. A. Salama, R. Seethapathy, M.
Assam, and S. Conti, “Adequacy evaluation of distribution system in-
and to minimize interruption costs, thus providing a cost-effec- cluding wind/solar DG during different modes of operation,” IEEE
tive means of improving system reliability. A sensitivity anal- Trans. Power Syst., vol. 26, no. 4, pp. 1945–1952, Nov. 2011.
ysis is further provided in order to investigate the impact of the [19] P. Wang, Y. Ding, and L. Goel, “Reliability assessment of restructured
power systems using optimal load shedding technique,” IET Gener.,
value of interruption cost on the results obtained. Transm., Distrib., vol. 3, no. 7, pp. 628–640, Jul. 2009.
[20] G. Wacker and R. Billinton, “Customer cost of electric service inter-
ruptions,” Proc. IEEE, vol. 77, no. 6, pp. 919–930, 1989.
ACKNOWLEDGMENT [21] M. J. Sullivan, M. G. Mercurio, and J. A. Schellenberg, “Estimated
value of service reliability for electric utility customers in the United
The authors would like to thank the Government of Canada States,” LBNL Research Project Final Report, Jun. 2009.
[22] M. F. Shaaban, Y. M. Atwa, and E. El-Saadany, “DG allocation for
for financially supporting this research through the Program on benefit maximization in distribution networks,” IEEE Trans. Power
Energy Research and Development. Syst., vol. 28, no. 2, pp. 639–649, May 2013.
[23] T. Niknam and H. Doagou-Mojarrad, “Multiobjective economic/emis-
sion dispatch by multiobjective -particle swarm optimisation,” IET
REFERENCES Gener., Transm., Distrib., vol. 6, no. 5, pp. 363–377, May 2012.
[24] S. Kannan, S. M. R. Slochanal, and N. P. Padhy, “Application and com-
[1] “Enabling tomorrow’s electricity system: Report of the Ontario Smart parison of metaheuristic techniques to generation expansion planning
Grid Forum” [Online]. Available: http://www.ieso.ca/imoweb/pubs/ problem,” IEEE Trans. Power Syst., vol. 20, no. 1, pp. 466–475, Feb.
smart_grid/ Smart_Grid_Forum-Report.pdf 2005.
[2] IEEE Guide for Design, Operation, and Integration of Distributed Re- [25] L. Goel and R. Billinton, “Determination of reliability worth for dis-
source Island Systems with Electric Power Systems, IEEE Std. 1547.4- tribution system planning,” IEEE Trans. Power Del., vol. 9, no. 3, pp.
2011, 2011. 1577–1583, Jul. 1994.
[3] B. Banerjee and S. M. Islam, “Reliability based optimum location of [26] Y. M. Atwa and E. F. El-Saadany, “Probabilistic approach for op-
distributed generation,” Int. J. Electr. Power Energy Syst., vol. 33, no. timal allocation of wind-based distributed generation in distribution
8, pp. 1470–1478, Oct. 2011. systems,” IET Renewable Power Gener., vol. 5, no. 1, pp. 79–88, Jan.
[4] A. S. A. Awad, T. H. M. El-Fouly, and M. M. A. Salama, “Optimal 2011.
distributed generation allocation and load shedding for improving dis- [27] S. Yan, H. Nai-Shan, and S. Zhi-Yuan, “Power flow calculation method
tribution system reliability,” Elect. Power Compon. Syst., vol. 42, no. for islanded power network,” in Proc. Asia-Pacific Power Energy Eng.
6, pp. 576–584, Mar. 2014. Conf. (APPEEC), Wuhan, China, Mar. 28–31, 2009, pp. 1–5.
[5] A. Pregelj, M. Begovic, and A. Rohatgi, “Recloser allocation for im- [28] Y. G. Hegazy, M. M. A. Salama, and A. Y. Chikhani, “Adequacy as-
proved reliability of DG-enhanced distribution networks,” IEEE Trans. sessment of distributed generation systems using Monte Carlo Simula-
Power Syst., vol. 21, no. 3, pp. 1442–1449, Aug. 2006. tion,” IEEE Trans. Power Syst., vol. 18, no. 1, pp. 48–52, Feb. 2003.

[29] G. M. Masters, Renewable and Efficient Electric Power Systems. Tarek H. M. EL-Fouly (M’02) received the B.Sc.
New York: IEEE/Wiley Interscience, 2004. and M.Sc. degrees in electrical engineering from Ain
[30] B. Venkatesh, R. Ranjan, and H. B. Gooi, “Optimal reconfiguration Shams University, Cairo, Egypt, in 1996 and 2002,
of radial distribution systems to maximize loadability,” IEEE Trans. respectively, and the Ph.D. degree in electrical engi-
Power Syst., vol. 19, no. 1, pp. 260–266, Feb. 2004. neering from the University Of Waterloo, Waterloo,
[31] Y. M. Atwa and E. F. El-Saadany, “Reliability evaluation for distri- ON, Canada, in 2008. He joined CanmetENERGY,
bution system with renewable distributed generation during islanded Natural Resources Canada, in 2008, as a Transmis-
mode of operation,” IEEE Trans. Power Syst., vol. 24, no. 2, pp. sion and Distribution Research Engineer, where he is
572–581, May 2009. conducting and managing research activities related
[32] HydroOne. Distributed Generation Technical Interconnection Re- to active distribution networks, microgrids and re-
quirements [Online]. Available: http://www.hydroone.com/Genera- mote communities. In 2010, he was appointed as Ad-
tors/Pages/TechnicalRequirements.aspx junct Assistant Professor at the Electrical and Computer Engineering Depart-
[33] S. Schoenung and C. Burns, “Utility energy storage applications ment, University of Waterloo.
studies,” IEEE Trans. Energy Convers., vol. 11, no. 3, pp. 658–665, His research interests include protection and coordination studies, integration
Sep. 1996. of renewable energy resources, smart microgrids, smart remote community ap-
[34] P. Poonpun and W. T. Jewell, “Analysis of the cost per kilowatt hour plications, demand side management, and forecasting.
to store electricity,” IEEE Trans. Energy Convers., vol. 23, no. 2, pp.
529–534, Jun. 2008.

Ahmed S. A. Awad (S’11) received the B.Sc. and Magdy M. A. Salama (F’02) received the B.Sc.
M.Sc. degrees in electrical engineering from Ain and M.Sc. degrees in electrical engineering from
Shams University, Cairo, Egypt, in 2007 and 2010, Cairo University, Cairo, Egypt, in 1971 and 1973,
respectively. He is currently pursuing the Ph.D. respectively, and the Ph.D. degree in electrical engi-
degree in the Department of Electrical and Computer neering from the University of Waterloo, Waterloo,
Engineering, University of Waterloo, Waterloo, ON, ON, Canada, in 1977. Currently, he is a Professor
Canada. In between September 2007 and 2009, in the Department of Electrical and Computer En-
he worked as an electrical design engineer with gineering, University of Waterloo. He has consulted
Dar El-Handasah (Shair and partners) and Allied widely with government agencies and the electrical
Consultants Co. in Egypt. Since 2008, he has been industry. His research interests include the operation
working with the Department of Electrical Power and control of distribution systems, power-quality
and Machines Engineering, Ain Shams University, Cairo, Egypt, where he is monitoring and mitigation, asset management, and electromagnetics.
currently on a study leave until he receives the Ph.D. degree. Dr. Salama is a Registered Professional Engineer in the Province of Ontario.
His research interests include application of energy storage in smart grids,
integration of renewable energy resources, electricity market equilibrium, as
well as operation and control of distribution systems.