See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315075080

Optimal Unit Commitment of Distributed Generation Resources Using Cuckoo

Optimization Algorithm

Article · June 2016

CITATIONS READS

0 444

1 author:

Mohammad Hamdi
KTH Royal Institute of Technology
438 PUBLICATIONS   6,967 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Multi-drug Strains View project

Desalination by solar energy assisted Membrane Distillation View project

All content following this page was uploaded by Mohammad Hamdi on 15 March 2017.

The user has requested enhancement of the downloaded file.

2016 3th International Computer and Electrical Engineering conference (ICEEC)
Optimal Unit Commitment of Distributed Generation
Resources Using Cuckoo Optimization Algorithm
Farhad Samadi Gazijahani*, Hassan Hosseinzadeh**, and Navid Taghizadegan***
*Azarbaijan Shahid Madani University, f.samadi@azaruniv.ac.ir
**Azarbaijan Shahid Madani University, h.hosseinzadeh@azaruniv.ac.ir
***Azarbaijan Shahid Madani University, n.taghizadegan@azaruniv.edu
Abstract: This paper presents a new probabilistic approach for
optimal unit commitment (UC) of distributed generation (DG) in
smart distribution networks (SDNs) considering economic and
reliability aspects. In this paper the economic and reliable UC of
DG is formulated and solved as an optimisation problem. The
network is assumed to be supplied by dispatchable and non-
dispatchable generators and energy storage systems (ESSs). The
optimization aim is to minimize the overall operating cost, power
losses cost, the pollutants emission cost and cost of energy not
supply (ENS). The economic and reliability evaluation is done
under uncertain behaviour of load and renewable power
generation. The system uncertainties are considered using a set
of scenarios and a scenario reduction method is applied to
enhance a trade-off between the accuracy of the solution and the
computational burden. Cuckoo optimization algorithm (COA) is
implemented to minimize the objective function as an
optimization algorithm. The effectiveness and efficiency of the
proposed method are validated through extensive numerical tests
on PG&E 69-bus test distribution system.
Keywords: Unit commitment, Probabilistic, COA, Smart
distribution networks.
1. Introduction
Smart Grids have been recently proposed in the
literature as an initiative to operate the distribution systems
in a more economic and efficient way. Energy storage
systems are one promising technology that can support the
incorporation of smart grids because of their capacity to
improve the system reliability and to facilitate the
integration of high penetration levels of renewable energy
sources [1]. The MG scheduling in grid-connected and
islanded modes is performed by the MG master controller
based on security and economic considerations. The
master controller determines the MG interaction with the
main grid, the decision to switch between grid-connected
and islanded modes, and optimal operation of local
resources. The MG optimal scheduling performed by the
MG master controller is considerably different from the
unit commitment (UC) problem solved by the ISO for the
main grid [2]. Control of voltage and frequency is
performed by local controllers. Such controllers control
inverters to generate the active power required by demand
side. This is to establish balance between production and
consumption [4]. The set point for each local controller,
that is the active power that should be generated, is
determined by the MGs central controller (MCC). While,
MCC is controlled by the MGs OS [5]. There is important
difference between the optimization of a MG and
conventional economic dispatch problem [6]. At present,
there are some researches on MG optimal energy dispatch.
In the grid-connected mode, the optimal energy dispatch is
analysed in MG [7] and the effect of time-of-use electricity
price and electric energy transaction are considered. To
consider the decentralized optimal power dispatch
strategies in [8], the sharing of marginal cost of each power
source is done through iterative and communication. In
[9], the real time energy optimization scheduling method
is proposed in independent operation mode of MG. The
economic dispatch optimization problem is solved with
different methods in literature [10]. For example, there are
different approaches to tackle this kind of optimization
problem, such as finding multiple Pareto-optimal solutions
[11] or optimizing a single multi objective function
obtained by a weighted sum of the targets [12]. The most
used optimal power flow (OPF) solution techniques are the
Lagrange–Newton method [13] and sequential quadratic
programming (SQP) [14]. In the last few years, stochastic
optimization methods have been applied to the OPF
problem. Reference [15] presents an approach to solve a
single objective OPF problem (to minimize the total
operating cost in a power system) by means of particle
swarm optimization algorithm. In [16] and [17], the multi
objective economic-emission OPF problem is solved using
a differential evolution algorithm. It should be mentioned
that the duty of each MG is to create balance between
supply and demand. In other word, in simple economic
dispatch problem power generation sources such as small-
scale energy resources (SSERs) produce power in order to
supply load demand and with increasing or decreasing fuel
in diesel generators, the balance between supply and
demand is provided. But considering MGs concept in
smart grids, the balance
1
2016 3th International Computer and Electrical Engineering conference (ICEEC)
Fig. 1 .Test distribution system with networked MGs
between generation and load is done through power
exchanging between MGs as well as main grid so that the
total cost of power generation in each MG as well as the
total cost of power exchanging between MGs and main
grid be optimized.
This paper proposes a new approach for probabilistic
optimal unit commitment considering uncertainties in load
and probabilistic modelling of generated power by
renewable SSERs. In order to deal with these uncertainties,
PDF is considered for power generation parameters. The
problem is solved with cuckoo optimization algorithm.
The main contributions of this paper to the research field
are related to probabilistic optimum energy management
with considering reliable and economical aspects for smart
distribution network that will benefit utilities, DG owners
and electricity consumers.
The rest of this paper is organized as follows: Section 2
discusses the probabilistic analysis method and concepts.
In Section 3, the generation and load models are explained.
The problem is formulated in Section 4 and Section 5
explains the solution algorithms and Section VI discusses
implementation and lastly, the conclusions are drawn in
Section VIII.
2. Probabilistic Analysis Method
At present, high penetration of SSERs into
distribution grids affect operation and planning of the
power systems. In WTs and PV power generation, wind
speed, and solar radiation are prime energy sources,
respectively. Because of stochastic behaviour of wind
speed and sun irradiance, power generation of the above
energy resources undergoing significant uncertainties.
Uncertainties analysis of the impact of SSERs such as
wind turbine (WT) and photovoltaic (PV) units on current
distribution systems based on deterministic methods is
complicated scenario. The probabilistic analysis of
distribution systems at presence of uncertainties is a
powerful tool for optimal operation and of distribution
systems. In probabilistic analysis, input data have
probability distribution function (PDF) and these data can
be described by cumulative distribution function (CDF).
Consequently, the obtained results from probabilistic
analysis are also presented in PDF and CDF forms.
Probabilistic analysis is applied in calculation of
probabilistic load flow (PLF) [18] in that authors are
performed a good review on PLF problem. One of the
problems which has stochastic and probability behaviour
at presence of SSERs is economic dispatch (ED).
Considering uncertainties in load consumption and
generated power by generation units, ED problem is
analysed using probabilistic method. In this paper, PDF is
considered for input data such as load, generated power by
SSERs, cost of transaction, and operation costs. Hence, the
output results are represented in framework of PDF.
3. System Model
3.1 Modelling of DG Units
The model describing each DG is described as follows.
a) WT: The power generated by WT as a function of
wind speed can be calculated by:
0 0 <V<PSTG
2
PWT = (A.V +B.V+C)*P rate Vci ≤V≤Vr (1)
Prate Vr ≤V≤Vco
{ 0 Vco ≤V≤∞ }
The relationship between the generated power and wind
speed of the WT can be demonstrated as Fig.2.
b) PV: The power generated by PV depends on the
irradiance and the ambient temperature obtained by:
Г(α+β)
Г(α)+Г(β)
×s(α-1) .(1-s)β-1 ,0≤s≤1,α≥0,β≥0
𝑓(s)= { }
0 , otherwise
(2)
2
2016 3th International Computer and Electrical Engineering conference (ICEEC)

each DG, as well as power sold to or purchased from the

grid on an hourly basis.
n
[fx (Px )+OMx (Px )+ex (Px )]+
∑
F1= { , } (9)
x=1
Closs *PPloss +Closs *QQloss

a) Fuel Consumption Function: Fuel expenses can be

Fig. 2. Output power of wind generation unit related to wind speed. calculated as a function of power generated by the xth DG
unit. The second term in [19], is an income coming from
μ.(1+μ) selling the thermal energy of CHPs. CHPs thermal energy
β=(1-μ).( -1) (3)
σ2 is used to supply the thermal load. In the case of shortage
μ.β of the thermal energy, the boiler supplies the rest of the
α= (4)
1-μ required thermal energy.
GING
Ppv =PSTG * *(1+k(TC -Tref )) (5) PCHP
GSTG fCHP (PCHP )=(CgasCHP * )-(Cth (ƟH⁄ )CHP *PCHP ) (10)
μCHP E

c) CHP: The output power generated by CHP is b) O&M Cost Function: The operation and
controlled by an installed governor. The rate of fuel maintenance (O&M) cost of the xth unit as a function of
consumed by CHP can be expressed by a second-order the power generated can be obtained by:
polynomial in terms of the power generated.
OMx (px )=kOMx * px (11)
LCHP (PCHP )=Θ.PCHP 2 +ϱ.PCHP +γ (6)
d) ESS: Energy storage system (ESS) enhances c) Pollution Function: The main source of air pollution
flexibility in power generation, reliability, and and greenhouse gas emissions are seen to be fossil fuel
consumption. The state of charge (SOC) value at time t is consumption producing gases such as CO2, SO2, NO, and
related to SOC value at time t-1 by the following equation. CO. The emission of each gas is converted to a
The SOC of ESS can be calculated by: corresponding expense through multiplication by a
coefficient as indicated by:
1
SOCt = SOCt-1 + ηch Pch ∆t ˗ Pdis ∆t (7) n
ηdis
3.2 Load Demand Model E(PG )= ∑ 10-2 (αi +βi PGi +γi PGi 2 )+ ζi exp(λi PGi ) (12)
i=1
Because of variable nature of load, it is suitable to model
the stochastic behaviour of the load. This modelling can be d) Power losses cost: Calculation of power losses implies
achieved either by measured data analysis or by solving the load flow problem. In this paper backward-
application of mathematical model. The SDN load forward approach is applied for calculation of power
behaviour is modelled as a normal distribution function. In losses.
this case, the SDN load is defined by a mean value (μ) and (Pb 2 +Qb 2 ) × Rb
standard deviation (σ) as: PLC (t)=(( ∑Nbb=1 [ ⁄ 2 ] )*CPloss ) (13)
Vb

1 ( pl -μ)2 4.2 Reliability evaluation

f(pl )= exp(- ) (8)
√2π×σ 2×σ2
4. Proposed Model
4.1 Cost evaluation
The economic-reliable UC between resources is
obtained by minimizing F1. This function, simultaneously,
minimizes operation and losses cost. Renewable energy
resources, such as WT and PV, do not incur expenses
associated with pollution and fuel. Accordingly, their
maintenance cost is considered a fixed value. Such value,
therefore, is independent of the power generated. In the
grid connected mode, WT and PV operate at their
maximum power point and the rest of the power demand
is provided by other distributed generation (DG) can be
calculated by F1. The optimal values, obtained by
minimizing F1, provide the amount of power generated by
Reliability evaluation of smart distribution networks
with renewable energy resources (RERs) has attracted the
interest of researchers worldwide. The reliability
assessment process is complicated by different output
characteristics of DGs in SDNs. In this paper, the
reliability evaluation is done for SDN under probabilistic
and uncertain behaviours of SDN components. One of the
reliability indices is energy not supply (ENS), in which it
describes the capacity of loss of load (kWh). ENS is
calculated based on power availability of each bus. Let
ENSi be the loss of load obtained for the ith contingency,
with a probability of probi. Then the average energy not
supply or loss of load expectation (AENS or LOLE) is
given by the following [20]:
Nc
AENS= ∑ ENS ×probi (14)
1

3
2016 3th International Computer and Electrical Engineering conference (ICEEC)

which has equality constraints on real and reactive power

at each bus as follows:
(P 2 +Qi 2 )
Pi+1 =Pi -ri i ⁄ 2 -Pi+1 (23)
Vi

(Pi 2 +Qi 2 )
Qi+1 =Qi -xi ⁄ 2 -qi+1 (24)
Fig. 3. Diagram of a radial electrical network Vi
In the above equations, we assume Pi is generated by
The reliability of the network or energy index of reliability
both RES-based DG units which are subject to
(EIR) is then given by the following:
uncertainties and controllable DG units, Qi is generated by
AENS controllable DG units.
EIR=1- (15)
P1
Nres 5. Solution Algorithm
CAENS = [ ∑Nb
b=1 Cint* λb * Lb * ( ∑ Pres Tres+
res=1
The cuckoo optimization algorithm (COA) is inspired
∑Nrep
rep=1 Prep Trep)] (16) by the life of a bird family, called Cuckoo [22]. Special
lifestyle of these birds and their characteristics in egg
In addition to mentioned indices, two other metrics of laying and breeding has been the basic motivation for
reliability which are proposed in [21], are calculated as
development of this new evolutionary optimization
Renewable Energy Penetration (REP) is one of metrics
algorithm. Similar to other evolutionary methods, Cuckoo
that describes the percentage of demand covered by
Optimization Algorithm (COA) starts with an initial
renewable energy (WT and PV units) in a smart grids. All
indices are calculated for SDN. In this paper, we set the population. The cuckoo population, in different societies,
SDNs cost of interruption in 1.23 USD/kWh as it is is in two types: mature cuckoos and eggs. The effort to
described for household consumption. We use the survive among cuckoos constitutes the basis of Cuckoo
following equation to calculate the total interruption cost Optimization Algorithm. During the survival competition
in SDN for the reliability index on a consumer side: some of the cuckoos or their eggs, demise. The survived
cuckoo societies immigrate to a better environment and
ICiday =1.23×CAENS (17) start reproducing and laying eggs. When moving toward
goal point, the cuckoos do not fly all the way to the
destination habitat. They only fly a part of the way and also
F(2)= ∑ (ICdey,MG ) (18) have a deviation. This movement is clearly shown in Fig.4
MG
4. Cuckoos’ survival effort hopefully converges to a state
4.3 Objective Function that there is only one cuckoo society, all with the same
In conclusion, economic and reliability viewpoints profit values.
which have been explained in previous parts are
considered in two unique objective function that 6. Results and Discussions
formulated below:
As shown in Fig. 1, a PG&E 69-bus distribution system
Z=H1 * F(1) + H2 * F(2)] (19) with three DGs and ESSs is used for solving the
probabilistic daily unit commitment (DUC) problem in
4.4 Constraints this paper. Details about the PG&E 69-bus test system can
a) Generation Capacity Constraint: For stable be found in [23]. In smart distribution network consists
operation the real power output of each DG and ESS is three type of SSERs and ESSs to power generation. All
restricted by lower and upper limits as follows: basic data for simulation are mentioned in the rest of this
paper. The aim of this paper is to solve the optimal
probabilistic Unit Commitment problem of DG in SDNs at
PGi min ≤ PGi ≤ PGi max , i=1,…N (20)
the presence of uncertainty in WT, PV uncertainty for a
SOCGi min ≤ SOCGi ≤ SOCGi max , i=1,…M (21) given 24 hours. Because of the intermittent behaviour of
some renewable energy resources and load variation this
b) Power Balance Constraint: The total electric power paper is done based on uncertainty in input data. To model
generation must cover the total electric power demand P D the uncertainty of wind speed and solar irradiation the
and the real power loss in transmission lines Ploss hence: historical measured data with five samples for each hours,
is used as practical data to model the distribution function
n
∑ PGi -PD -Ploss =0. (22) of wind speed and solar irradiation. As an energy
i=1 management this paper is solved the SDN energy
management problem based on measured data of 24 h.
Consider an electrical network as shown in Fig.3,
calculation of implies solving the load flow problem,

4
2016 3th International Computer and Electrical Engineering conference (ICEEC)

Fig. 5. Mean generated power of SSERs in DEM.

Fig. 4. Immigration of a sample cuckoo toward goal habitat.

In this paper, SDN is modelled as a small disco with

capability of power interchange between SDN and the
main grid. To consider the effect of power generation cost
in smart grid, it is assumed that the characteristic of SSERs
such as size, and technology of WT, PV, and CHP is
different. In Table 1, the values of parameters of WT and
PV units are given that are used in simulation.
Based on the above discussion, the cost of per kilowatt
generated power by each SSERs at each MG is different.
Indeed, this paper is modelled as a daily energy
management problem in market operation of smart
distribution network. In this case, the probabilistic energy
Fig. 6. Voltage profile of network based on mean value.
management problem is solved using cuckoo optimization
algorithm as a heuristic optimization algorithm.
In this paper, DUC problem is solved considering PDF
for some parameters of power generation. Proposed
example solves DUC problem at presence of load demand
and renewable SSERs uncertainties. The results includes
the total cost at any states of generated and transaction
powers.
The probabilistic analysis about DUC has a good
prospect in operation, planning, unit commitment problem
studies, etc. PDF of the purchased and sold powers of
SDNs gives an insight vision to the dispatcher to evaluate
the risk of change in system total cost with respect the
variation in load and SSER power. This value at risk for
purchase and sold power can be calculated and can be
useful for dispatchers. The mean value of the PDF can use Fig. 7. Power losses of network based on mean value.
as the power with high probability of the appearance.
Indeed, in this case, it is possible to evaluate the
forecasting errors in load, wind, and solar short-term
scheduling.
The average value of generated power of PV, WT, CHP
and ESS can be shown in Fig. 5 for 24 hours. In
probabilistic analysis PDF form is more accurate than
mean value. PDF of ENS describes each possible state of
it considering related probability.

Table.1.Values of parameters in WT and PV.

WT Pr=250KW VOUT-IN=2m/s Vr=14m/s VOUT-
IN=25m/s
n=4 α=3 β=12 -

PV PSTC=250KW GSTC=1000W/m2 K=0.001 Tc=25 C Fig.8.PDF modeling of PV and WT primary input and generated power
in SDN.

5
2016 3th International Computer and Electrical Engineering conference (ICEEC)

Table.2. Statistical analysis for UC of DG units.

DER PV WT CHP ESS
Period (3h) 1 2 1 2 1 2 3 1 2 3
1 0 0 1 1 1 1 0 -1 -1 -1
2 0 0 0 0 1 1 0 -1 -1 -1
3 1 0 0 0 1 0 1 0 0 0
4 1 1 0 1 0 0 1 1 0 1
5 1 1 1 1 1 1 1 1 1 0
6 1 1 1 0 1 1 1 0 1 1
7 0 1 0 0 1 1 0 1 0 1
8 0 0 0 1 0 0 1 -1 -1 -1

Based on this figure and Table 2 shows the UC of
distributed generation including PV, WT and CHP with
energy storage system at daily unit commitment in the
smart distribution network. In this table each period is
three hours. The voltage profile based on mean values has
been shown in Fig. 6. According to this figure, with using
the proposed method for daily energy management of
SDN, voltage profile of system is improved. Fig. 7 shows
the power losses for proposed method and traditional
energy management for smart distribution network.
According to this figure, it is observed that by applying the
proposed method for energy management the power losses
of system are decreased.
Based on correlation between network input and output
variables, the output of the network parameters show
probabilistic behavior. To explain this, for example, Fig. 8
is illustrated the probabilistic generated power of both WT
and PV units in SDN.
Fig. 9. PDF of ENS and AENS in SDN.
The ENS and AENS of SDN can be shown in PDF form
in Fig. 9. In probabilistic analysis PDF form is more
accurate than mean value. PDF of ENS describes each
possible state of it considering related probability. After
obtaining the PDF of powers and reliability indices, the
operation, emission and reliability costs can be
represented. In this paper one thousand scenarios are
generated using Monte Carlo simulation (MCS) to
represent the prediction errors in the prediction horizon.
Scenario reduction is applied to reduce the computation
efforts while maintaining the solution accuracy. The 1000
generated scenarios are reduced to 30 scenarios.
7. Conclusion
In this paper a framework for optimal probabilistic unit
commitment of DG units in smart distribution networks
considering economic, reliability and environment
operation of distributed energy resources has been
proposed. The uncertainty in SDNs components such as
DGs and load is modelled with PDF. The economic and
reliable UC of DGs in SDNs is formulated as an
optimization problem. A stochastically and probabilistic
modelling of both DGs and load is performed to determine
the optimal management of SDNs with minimum cost
based on economic analysis of the power transaction
between the SDN and main grid. Based on the results, the
mean, standard deviation, and PDF of each generated
power with SSERs is determined considering optimization
constraints. Statistical analysis for generated power and
costs are given. The cuckoo optimization algorithm is
applied to minimize the cost function as an optimization
algorithm. Results show that it is possible to regulate the
power demand and transaction between SDN and the main
grid. Moreover, it is indicated that the power sharing
between DGs with main grid can reduce the total operation
cost of the future distribution network. One of the main
results of this paper by probabilistic modelling of the input
variables, the output variables can be represented as
random variables. This leads to a better and comprehensive
vision for network experts to manage the marginal
operation of the network under uncertainties. This can
guarantee the robust energy management of smart
distribution grids in the presence of network uncertainties.

6
 2016 3th International Computer and Electrical Engineering conference (ICEEC)

Nomenclature

s Set of scenarios SOC State of charge

b Branch σ Standard deviation
PWT(v) Generated power at wind turbine π Mean value of distribution function
Pr,WT Rated power of WT ζ,𝛿 Generator emission characteristics
PD Load demand ENS Energy not supply
v Wind speed η Efficiency
PPV Output power of the module at irradiance 𝜇 Membership function
GING Incident irradiance Cint Price of energy not supply
GSTC Irradiance at standard test condition CPloss Price of losses
PSTC Rated power of photovoltaic Nres Number of nodes isolated during fault
Z Objective function Nrep Number of nodes isolated during fault
Tc, Tr Cell reference temperatures respectively Pres Loads are restored during fault
k Maximum power correction temperature Prep Loads are not restored during fault
α, β Shape and scale parameter of Beta Tres Duration of fault location and switching
Γ Parameters of the Weibull model Trep Duration of the fault repair
Θ, ϱ and γ CHPs coefficients Rb Resistance of the line b
λ Failure rate Lb Length of the line b

dispatch of plug-in electric vehicles,” IEEE Trans . Power

References Syst, vol. 28. no. 3, pp. 2768 – 2778, 2013.
[14] Y. Levron, J. M. Guerrero, and Y. Beck, “Optimal power flow
in microgrids with energy storage,” IEEE Trans.Power Syst.,
[1] Department of Energy Office of Electricity Delivery and
vol. 28, no. 3, pp. 3226–3234, Aug. 2014.
Energy Reliability. Summary Report: 2012 DOE Micro - grid
[15] Farhad samadi gazijahani, Ata Ajoul Abadi, Navid
Workshop.2012.
Taghizadegan, “Multi-objective Stochastic Optimal Power
[2] The U.S. Department of Energy. The Smart Grid: An
Dispatch for Active Distribution Systems Including
Introduction. [Online]. Available: http://www.oe.energy.gov/
Microgrids,” electric power engineering conference (EPEC
Documents and Media/ 2014.
2016), 158 – 170.
[3] H. Jiayi, J. Chuanwen, and X. Rong, “A review on distributed
[16] H. Huang, C. Y. Chung, K. W. Chan, and H. Chen,“Quasi-
energy resources and microgrid,” Renew. Sustain. Energy Rev.,
Monte Carlo based probabilistic small signal stability
vol. 12, no. 9, pp. 2472–2483, Dec. 2011.
analysis for power systems with plug - in electric vehicle
[4] N. Cai, N. T. T. Nga, and J. Mitra, “Economic dispatch in
and wind power integration,” IEEE Trans.Power
microgrids using multi-agent system,” in Proc. North Amer.
Syst,vol.28,no.3,pp. 3335–3343. 2013.
Power Symp. (NAPS), Champaign, IL, USA, Sep., pp.1–5.2013
[17] B. Zhao, X. Zhang, J. Chen, C. Wang, and L. Guo, “Operation
[5] Y. M. Atwa, E. F. El-Saadany, M. M. A. Salama, and R.
optimization of standalone microgrids considering lifetime
Seethapathy,“Optimal renewable resources mix for distribution
characteristics of battery energy storage system,” IEEE Trans.
system energy loss minimization,” IEEE Trans. Power Syst.,
Sustain. Energy, vol. 4, no. 4, pp. 934– 943, Oct. 2013.
vol. 25, no. 1, pp. 360–379,Feb. 2011
[18] Z. Wang, B. Chen, J. Wang , M. Begovic and C. Chen , “
[6] M. H. Moradi and M. Abedini, “A combination of genetic
Coordinated Energy Management of Networked Microgrids
algorithm and particle swarm optimization for optimal DG
in Distribution Systems,” IEEE Trans. Smart grid., vol. 6, no.
location and sizing in distribution systems,” Int. J. Elect. Power
1, pp. 3226–3234, Jan. 2015.
Energy Syst., vol. 34, no. 1, pp. 66–74, Jan.2013.
[19] Farhad samadi gazijahani, Ata Ajoul Abadi, Javad Salehi,
[7] R. H. Lasseter,“Smart distribution: Coupled microgrids,” IEEE
“Optimal Multi-Objective Operation of Multi Microgrids with
Trans. Power Syst., vol. 99, no. 6, pp. 1074–1082, Jun.2012
Considering Uncertainty,” power system conference
[8] M. E. Khodayar, M. Barati, and M. Shahidehpour, “Integration
(psc2016). Tehran, iran, pp. 425–433.
of high reliability distribution system in microgrid operation,”
[20] M. Fathi and H. Bevrani, “Statistical cooperative power
IEEE Trans. Smart Grid, vol. 3, no. 4, pp. 1997–2006,
dispatching in interconnected microgrids,” IEEE
Dec.2012.
Trans.Sustain. Energy, vol. 4, no. 3, pp. 586–593, Jul. 2013.
[9] J. Y. Kim et al.,“Cooperative control strategy of energy storage
[21] J. Wu and X. Guan, “Coordinated multi-microgrids optimal
system and microsources for stabilizing the microgrid during
control algorithm for smart distribution management system,”
islanded operation,” IEEE Trans. Power Electron., vol. 25, no.
IEEE Trans. Smart Grid, vol. 4, no. 4, pp. 2174–2181, Dec.
12, pp. 3036–3048, Dec. 2011.
2014.
[10] J. Rocabert, A. Luna, F. Blaabjerg, and P. Rodríguez,“Control
[22] R. Rajabioun, “Cuckoo Optimization Algorithm”, Applied Soft
of power converters in AC microgrids,” IEEE Trans. Power
Computing. vol. 11, pp. 5508–5518, 2011.
Syst., vol. 27, no. 11, pp. 4734–4749, Nov. 2012.
[23] S. A. Arefifar, Y. A.-R. I. Mohamed, and T. H. M. El-Fouly,
[11] Huang Y, Mao S, Nelms RM.,“Adaptive electricity scheduling
“Optimum microgrid design for enhancing reliability and
in microgrids ,” IEEE Trans Smart Grid; vol. 5, no (1): 270–
supply-security,” IEEE Trans. Smart Grid, vol. 4, no. 3, pp.
81. 2014.
1567–1575, Sep. 2013.
[12] D. P. Tuttle and R. Baldick, “ The evolution of plug-in
electric vehiclegrid interactions ,” IEEE Trans. Smart Grid,
vol. 3, no. 1, pp. 500–505, Mar . 2012.
[13] W. Yao, J. Zhao, F. Wen, Y. Xue, and G. Ledwich,“A
hierarchical decomposition approach for Coordinated

7
View publication stats