Renewable Energy 94 (2016) 10e21

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Probabilistic multi-objective optimal power flow considering

correlated wind power and load uncertainties
S. Shargh, B. Khorshid ghazani, B. Mohammadi-ivatloo*, H. Seyedi, M. Abapour
Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Increasing penetration of wind power in power systems causes difficulties in system planning due to the
Received 23 June 2015 uncertainty and non dispatchability of the wind power. The important issue, in addition to uncertain
Received in revised form nature of the wind speed, is that the wind speeds in neighbor locations are not independent and are in
11 January 2016
contrast, highly correlated. For accurate planning, it is necessary to consider this correlation in optimi-
Accepted 19 February 2016
zation planning of the power system. With respect to this point, this paper presents a probabilistic multi-
Available online 21 March 2016
objective optimal power flow (MO-OPF) considering the correlation in wind speed and the load. This
paper utilizes a point estimate method (PEM) which uses Nataf transformation. In reality, the joint
Keywords:
Optimal power flow
probability density function (PDF) of wind speed related to different places is not available but marginal
Probabilistic multi-objective optimization PDF and the correlation matrix is available in most cases, which satisfy the service condition of Nataf
Correlated wind power transformation. In this paper biogeography based optimization (BBO) algorithm, which is a powerful
Nataf transformation optimization algorithm in solving problems including both continuous and discrete variables, is utilized
in order to solve probabilistic MO-OPF problem. In order to demonstrate performance of the method,
IEEE 30-bus standard test case with integration of two wind farms is examined. Then the obtained results
are compared with the Monte Carlo simulation (MCS) results. The comparison indicates high accuracy of
the proposed method.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction economic behaviors of the power market and making compromise

Energy security concerns, environmental issues and increasing
demand have directed lots of attention to renewable energy sour-
ces. Among renewable energy sources, wind power has found high
penetration in power systems throughout the world [1]. The inte-
gration of a significant amount of wind power which has uncertain
nature has made important challenges in power system planing
and operation [2]. The main issue is that the wind speed varies
stochastically and output power of wind turbine has the same
behavior. Many researchers have demonstrated that spatial and
temporal correlation of renewable energy, i.e. wind power, which
affect power system can be predicted and modeled using efficient
methods [3,4]. Furthermore, in systems with high penetration of
wind power, spatial and temporal correlation of wind power should
be considered for extra accurate and reliable results [5,6].
Optimal power flow (OPF) is effective tool for analyzing
* Corresponding author. Smart Energy Systems Lab., Faculty of Electrical and
Computer Engineering, University of Tabriz, Tabriz, Iran.
E-mail addresses: s.shargh92@ms.tabrizu.ac.ir (S. Shargh), b.khorshidi92@ms.
tabrizu.ac.ir (B. Khorshid ghazani), bmohammadi@tabrizu.ac.ir (B. Mohammadi-
ivatloo), hseyedi@tabrizu.ac.ir (H. Seyedi), abapour@tabrizu.ac.ir (M. Abapour).
between power system economic and security [7]. This problem
sometimes decomposed into two different problem called optimal
active power dispatch and optimal reactive power dispatch (ORPD)
problems [8,9]. Probabilistic OPF (POPF) has been widely used in
planning and operation of power systems [10e12]. The methods of
solving POPF can be classified into three main categories: analytical
methods, simulation methods and approximation methods. Simu-
lation methods refer to Monte Carlo simulation (MCS) method
which is computationally expensive but popular for its simple
implementation and high accuracy [13e15]. On the other hand,
analytical methods are based on linearization and are less accurate
[16e18]. An approach using an analytical method to solve the
probabilistic load flow (PLF) considering load demand correlation
based on the use of cumulants and Gram-Charlier expansion is
proposed in Ref. [19].
Approximation methods like point estimate method (PEM),
which is proposed to solve POPF in Refs. [7,20,21], have less
computational burden and provides accurate results in the low
number of uncertain variables but is impractical in large-scale
problems. Unfortunately PEM methods are not able to handle
problems with correlated Random Variables (RVs) and some
modifications are necessary to make it capable. Two kinds of point

http://dx.doi.org/10.1016/j.renene.2016.02.064
0960-1481/© 2016 Elsevier Ltd. All rights reserved.
 S. Shargh et al. / Renewable Energy 94 (2016) 10e21 11

estimate methods (2m and 2m þ 1 scheme) are used in Ref. [22] to presents simulation results of examining proposed method on
solve the POPF. The correlated samples of wind power injections IEEE 30-bus test system and finally Section 6 concludes the remarks
and load points have been generated by Cholesky decomposition and results of the paper.
method, and the interior point method is used to solve the deter-
ministic optimal power flow calculation.
A modified PEM is proposed in Ref. [23] to solve POPF, which 2. Uncorrelated PEM for probabilistic multi-objective OPF
considers correlation among wind power, solar energy and demand
load. A review of the probabilistic techniques used for POPF studies 2.1. Probabilistic multi-objective optimal power flow formulation
and the application of unscented transformation (UT) method to
generate correlated samples is presented in Ref. [24]. Zhaos point The multi-objective OPF problem with uncorrelated input vari-
estimate method (PEM) combined with Nataf transformation applied ables can be formulated generally as below:
into correlated PLF calculation is proposed in Ref. [25]. The proposed
Minimize FðX; YÞ ¼ fF1 ðX; YÞ; F2 ðX; YÞg
method can deal with correlated input RVs with normal or non-
S:t: GðX; YÞ  0 (1)
normal probability distributions and instead of PDFs of multivariate
HðX; YÞ ¼ 0
RVs requires raw moments of the marginal distribution function of
each input RV and their correlation coefficients. In Ref. [26] traditional where F1(X,Y) and F2(X,Y) are objectives of probabilistic optimal
2m þ 1 PEM extended by using a transformation to handle correlated power flow problem (POPF). X is the vector of control variables and
RVs with arbitrary distributions in POPF problem. In [27], a method is Y is the vector of uncertain input variables of POPF problem. H(X,Y)
proposed for incorporati2m þ 1 ng the effects of spatially and and G(X,Y) are representing the equality and inequality constraints
temporally correlated input RVs within the point estimate method for of the multi-objective problem, respectively.
POPF simulation. This method is used to solve multi-period POPF In this paper, fuel cost and total emission of the system are
including correlation among time intervals of wind power generation. considered as objectives of multi-objective POPF problem, which
Most of the papers studying effect of the correlation of uncertain can be formulated as follows [29]:
variables on POPF problem presented different methods to solve Fuel cost function:
the problem but few of them have studied effect of variation of
correlation on control and output variables. For instance, a new X
NG
2
method is presented in Ref. [26] in order to observe effect of cor- F1 ðX; YÞ ¼ ai þ bi Pgi þ ci Pgi (2)
relation on uncertain variables following different distributions but i¼1
do not investigate how the variables will affect with variation of
Emission cost function:
correlation. In other case, a new method for solving POPF with
correlated variables is presented in Ref. [23] and effect of correla- NG 
X  
2
tion on accuracy of the presented method is studied, but the impact F2 ðX; YÞ ¼ ai þ bi Pgi þ gi Pgi þ εi exp li Pgi (3)
of variation of the correlation on control variables was not studied. i¼1
It should be noted that, OPF problem in accurate model, should
Where F1(X,Y) is the total fuel cost of generation units in $/h, NG
consider many objectives simultaneously, as in reality is required
is total number of generation units, Pgi (MW) is active power output
[28]. So it is important to investigate effect of RVs and their cor-
of the ith generation unit and ai, bi, ci are the fuel cost coefficients of
relation in multi objective POPF which is more closer to reality and
ith generation unit. F2(X,Y) is total emission released from ith
can be more useful in planning of system to reach different targets,
generation unit (ton/h), Pgi(pu) is active power output of ith
simultaneously.
generator and ai, bi, gi, εi and li are their emission coefficients.
In this paper, Biogeography Based Optimization algorithm with
This paper uses weighted sum method to solve multi-objective
weighted sum method is used to solve probabilistic multi-objective
optimal power flow:
optimal power flow problem and traditional PEM based Nataf
transformation is utilized to handle correlation of wind sources and F1 ðX; YÞ F ðX; YÞ
load demands. This paper studies the effect of correlation on con- F ¼ k1 þ k2 2 (4)
F1min F2min
trol and output variables of MO-POPF problem. It is noteworthy
that, the effect of correlation on the control and output variables of where, k1 and k2 are the weighting coefficient of fuel cost and
multi-objective POPF problem is studied for the first time in this emission cost objectives, simultaneously, and F1min and F2min are
paper. The standard IEEE 30-bus test system with two extended minimum amount of them which can be obtained solving single
wind farms is used to examine the proposed method. objective OPF. In this approach, for balanced weighting of objective
Contributions of the paper are as follows: functions k1 and k2 are considered to be equal to 1.
Control variables:
1. Using Hongs point estimate method and Nataf transformation to
 
solve POPF problem with correlated input variables. X ¼ Pg ; Vg ; T; Qc 1N (5)
2. Solving probabilistic multi-objective optimal power flow
considering wind power and load points correlation. h i
3. Using biogeography based optimization method as the optimi- Pg ¼ Pg1 ; Pg2 ; …; PgðNg 1Þ (6)
1ðNg 1Þ
zation method.
4. Observing effect of correlation on the control variables and both h i
objectives of the probabilistic multi-objective optimal power flow. Vg ¼ Vg1 ; Vg2 ; …; VgNg (7)
1Ng

The rest of the paper is organized as follows: Section 2 formu-  

T ¼ T1 ; T2 ; …; TNt 1Nt (8)
lates probabilistic MO-OPF and introduces PEM to solve it with
uncorrelated RVs. Section 3 illustrates BBO algorithm for solving  
MO-OPF problem. Section 4 explains Nataf transformation and PEM Qc ¼ Qc1 ; Qc2 ; …; QcNc 1Nc (9)
based Nataf transformation to handle correlated RVs. Section 5
12 S. Shargh et al. / Renewable Energy 94 (2016) 10e21

Decision variables vector (X) consists of active power output of !

generation units except slack generator (Pg), voltage magnitude of 1 ðP  mÞ2
f ðPd Þ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi exp  d 2 (19)
PV buses (Vg), tap of transformers (T) and the reactive power of ð2pÞs 2s
shunt capacitors (Qc). Furthermore N, Nt and Nc are total number of
decision variables, number of tap changing transformers and where, Pd is the load demand, m and s are its mean value and
number of shunt capacitors, respectively. standard deviation, respectively.
The above mentioned objective functions should be simulta-
neously optimized considering the following constraints. 2.2.2. Wind speed modeling
Statistical models for wind speed do not follow normal distri-
X
Nb
 
Pgi þ Pwi  Pdi ¼ Vi Vj Gij cosqij þ Bij sinqij (10) bution but rather Weibull distribution. Many studies show that
j¼1 wind speed can be modeled as two parameter Weibull distribution
[31,32]. This model is useful for both short term and long term
studies. Weibull distribution can be formulated as follows [32]:
X
Nb
 
Qgi þ Qci  Qdi ¼ Vi Vj Gij sinqij  Bij cosqij (11)
 h 
h vðh1Þ v
j¼1 f ðvÞ ¼ exp  (20)
c c c
Constraints (10) and (11) model the power flow equations. Nb is
the number of buses, Pgi and Qgi are the generated active and where, v is wind speed, c and h are the scale and shape parameters
reactive powers and Pdi and Qdi are the active and reactive power of Weibull distribution, respectively.
demands at the ith bus, respectively. Pwi and Qci are the active po- Output power of wind turbine depends on the wind speed,
wer output of the wind farm and reactive power injected by which is represented using power curves. A sample power-wind
compensation device connected to ith bus, respectively. Note that speed characteristics of a wind turbine is presented in the
reactive power generated by wind turbines is neglected in this following equation [33].
paper in line with [26]. Vi and Vj are the voltage magnitude at the 8
ith and jth buses and qij ¼ qiqj where qi and qj are the voltage angle >
> 0; 0  vi  vin;i
>
>
of these buses. >
> v  vin
>
< Pr;i i ; vin;i  vi  vr;i
min max Pwi ¼ vr;i  vin;i (21)
Pgi  Pgi  Pgi ; i ¼ 1; 2; …; Ng (12) >
>P ;
>
> vr;i  vi  vo;i
>
> r;i
>
:
min
Qgi max
 Qgi  Qgi ; i ¼ 1; 2; …; Ng (13) 0; vo;i  vi

where, vi is the wind speed of wind turbine connected at bus i. Pr,i is

min max
Vgi  Vgi  Vgi ; i ¼ 1; 2; …; Ng (14) the rated capacity of the wind turbine. vin,i, vr,i and vo,i are the cut-in,
rated and cut-out speed of the wind turbine, respectively. Fig. 1
Timin  Ti  Timax ; i ¼ 1; 2; …; Nt (15) shows power-speed characteristics of a typical wind turbine [34].

2.2.3. Point estimate method for probabilistic MO-OPF

Qcimin  Qci  Qcimax ; i ¼ 1; 2; …; Nc (16) In this paper, 2m þ 1 PEM is used to solve probabilistic MO-OPF
problem, which is proposed in Ref. [21]. The basic idea of the PEM is
VLimin  VLi  VLimax ; i ¼ 1; 2; …; Npq (17) to use some points of uncertain input variables. In 2m þ 1 PEM, it is
required to evaluate the output variables 2m þ 1 times for m
  random variables. As mentioned before, the output variables in
 
Sij   Smax
ij ; i; j ¼ 1; 2; …; NL (18) probabilistic MO-OPF model can be expressed by:

where, NL and Npq are the number of lines and PQ buses, Sij is
apparent power flowing in the line connecting bus i to bus j, and VLi
is the voltage magnitude of PQ buses. It should be mentioned that
superscripts min and max show lower and upper bounds of the
corresponding parameter.
2.2. Uncertainty modeling
One of the most important aspects of the probabilistic planning
is to propose appropriate statistical model for the uncertain vari-
ables in system. In this paper, load and wind generation are
modeled as uncertain variables which are developed as following.
2.2.1. Load modeling
Load uncertainty is commonly modeled with normal distribu-
tion [30,19]. In this paper, active load demand is modeled as normal
distributed random variable and load power factor is assumed to be
constant [30]. The corresponding PDF can be formulate as following
equation:
Z ¼ FðX; YÞ (22)
In this model X and Y are control variables and uncertain input
variables, respectively.
X ¼ ½PG ; VG ; T; QG ; Y ¼ ½Pw ; Pd  (23)
In probabilistic MO-OPF model, formulated in this paper, PEM is
used to determine 2m þ 1 points on standard normal distribution for
each of the uncertain variables. Then samples of wind speeds and
loads following specific distributions can be obtained by transforming
the samples from standard normal distribution. After running pro-
posed model 2m þ 1 times, raw moments of output variables
including total fuel cost, released emission and control variables are
calculated. According to the 2m þ 1 PEM, each random variable is
clarified with three locations and weights which are determined by:
yik ¼ mi þ xik si ði ¼ 1; 2; …; m; k ¼ 1; 2; 3Þ (24)
where, mi and si are the mean and standard deviation of yi and xik is
the coefficient representing the kth location on its probability dis-
tribution and m is the number of random input variables. Standard
 S. Shargh et al. / Renewable Energy 94 (2016) 10e21 13

Fig. 1. Output power variation for a typical wind turbine according to wind speed.

locations xik are calculated using (25). 3. Biogeography based optimization for solving MO-OPF
problem
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
li3 3
xik ¼ þ ð  1Þ3k li4  l2i3 ðk ¼ 1; 2Þ xi3 ¼ 0 (25) This method uses the concept of how a good habitat is found in
2 4
process of migration from one habitat to another. In this method, an
where, li3 and li4 are the third and fourth standardized central optimization solution vector is known as habitat, the variables of
moments of yi known as skewness and kurtosis. Weighting factors every solution vector are called suitability index variables (SIV s)
related to each location yik are calculated by: and fitness of the solution vector is known as quality for living in a
habitat, called habitat suitability index (HSI). When some species
ð1Þ3k migrate from one habitat to other habitats, the process is called
wik ¼ ðk ¼ 1; 2Þ (26) emigration and when some species migrate into one habitat from
xik ðxi1  xi2 Þ
other habitats, the process is called immigration.
The following probabilistic mathematical model represents the
1 1
wi3 ¼  (27) concept of the emigration and the immigration process in BBO
m li4  l2i3 algorithm:
Assume that at time t the probability of existence of exactly s
After calculation of concentrations for all uncertain variables,
species in a habitat is Ps(t) then at time t þ Dt the probability is:
2m þ 1 vectors of input variables Y are constructed as follows:
Ps ðt þ DtÞ ¼ Ps ðtÞð1  xs Dt  hs DtÞ þ Ps1 xs1 Dt þ Psþ1 hsþ1 Dt
Yj ði; kÞ ¼ ½m1 ; m2 ; …; mi1 ; yik ; miþ1 ; …; mm  ðk ¼ 1; 2; j
(33)
¼ 1; 2; …; 2mÞ (28)
where, hs and xs are the emigration and immigration rates at time t.
Y2mþ1 ¼ ½m1 ; m2 ; …; mm  (29) Emigration rate hk and immigration rate xk for k number of the
species can be calculated using the following equations:
Then deterministic MO-OPF problem is solved for every Yj vec-
tor, so MO-OPF is solved 2m þ 1 times and 2m þ 1 compromise Ek
hk ¼ (34)
solution vector of Z is obtained. Finally statistical moments of n
output variables of Z are calculated using weighting factors, by the

following expressions: k
xk ¼ I 1  (35)
n
X
m X
3
mn ¼ E½Zn ði; kÞ ¼ wik ðZn ði; kÞÞ (30) where, n is the total number of the species in the habitat, E and I are
i¼1 k¼1 the maximum value of emigration and immigration rates,
respectively.
h i Xm X
3   BBO algorithm uses migration and mutation mechanisms to
E Zn ði; kÞ2 ¼ wik Zn ði; kÞ2 (31) search in the feasible solutions.
i¼1 k¼1

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
h i rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  ffi 3.1. Migration
sn ¼ E Zn ði; kÞ2  ðEðZn ÞÞ2 ¼ E Z 2n  m2n (32)
In the migration mechanism, emigration and immigration rates
In these expressions, n is number of output variables. mn and sn are determine whether a solution will modify or not, which depend on
mean and standard deviation of nth output variable, respectively. other solutions. When a solution is selected to be modified, the
14 S. Shargh et al. / Renewable Energy 94 (2016) 10e21

immigration rate xi is used to probabilistically decide whether or

not to modify each SIV and when a SIV Si is chosen for modification,
the emigration rates hj of other solutions are deployed to proba-
bilistically select the solution Sj, which should migrate from a
randomly selected SIV to solution Si.

3.2. Mutation

In BBO algorithm, each habitat has an associated probability for

mutation which depends on HSI of that habitat. If this probability is
selected to be high, then the solution will has very little chance to
mutate and vice versa for less probable solutions. Mutation rate of
the solutions can be obtained using the following equation.


1  PS
mðSÞ ¼ mmax (36)
Pmax

where, mmax is a user-defined parameter and Pmax is maximum

value of mutation probability. Details about the BBO algorithm can
be found in Ref. [35].

3.3. BBO algorithm for MO-OPF problem

As mentioned, every solution vector of problem is a habitat of

the BBO algorithm and decision variables are SIV s. The step by step
description of the BBO algorithm, which is used to solve MO-OPF, is
presented in the following.

1. Define system data such as bus data, line data, load data, cost
and emission coefficient of generators and algorithm pa-
rameters such as number of iteration and habitats and so on.
2. Initialize habitats within the feasible solutions.
3. Check for equality constraints by running NR load flow pro-
gram for every habitat.
4. Calculate fitness of solutions (fuel cost, emission) known as
HSI.
5. Calculate emigration rate (h) and immigration rate (x) for
habitats.
6. Create new solutions using migration and mutation
strategies.
7. Verify feasibility of the solutions and run NR load flow
program.
8. Calculate HSI for new solutions.
9. Select the best solutions of the modified set of solutions and
the old set of solutions according to the keep rate for creating
the new set of habitats for the next iteration.
10. Go to step 5 and repeat following steps until max iteration
reach.

4. PEM using Nataf transformation to solve probabilistic MO-

OPF with correlated input variables
Fig. 2. Flowchart of the proposed method.

The PEM described to solve probabilistic MO-OPF is not useful in

problems considering correlation among input random variables. One
way to solve such problems is to use PEM and Rosenblatt trans-
formation [36]. The PEM calculates estimating points in standard
normal space and then inverse Rosenblatt transformation is
employed to map the estimating points from standard normal space
to original space [37]. Rosenblatt transformation is one of the typical
methods for solving the problems involving correlated random vari-
ables, but it has major limitations in solving most of the engineering
problems. The most important limitation of Rosenblatt trans-
formation is the requirement of joint PDF of random variables which is
impossible to gain in most of the engineering studies. This paper uses
Nataf transformation instead of Rosenblatt transformation for the
transformation from the correlated arbitrary space to uncorrelated
standard normal space [37,25].
4.1. Nataf transformation
As mentioned before, Nataf transformation is a method to map
the correlated random variables from correlated arbitrary distri-
bution to uncorrelated standard normal space. This method em-
ploys normal Copula function to obtain joint PDF of the input
random variables using marginal PDF of RVs and their correlation
matrix to obtain the respective correlation matrix in standard
normal space [38]. Assume that the input random vector Y and the
 S. Shargh et al. / Renewable Energy 94 (2016) 10e21 15

Table 1
Benchmark functiones [40,41].

Benchmark function n Search space Global optimum

Pn1 30 [30,30]n 0
f1 ðxÞ ¼ i¼1ð100ðxiþ1  x2i ÞÞ2
þ ðxi  1Þ 2
Pn P 30 [100,100]n 0
f2 ðxÞ ¼ i¼1 ð ij¼1 xi Þ2
P 30 [100,100]n 0
f3 ðxÞ ¼ ni¼1 ð½xi þ 0:5Þ2
P 30 [5.12,5.12]n 0
f4 ðxÞ ¼ ni¼1 ðx2i  10cosð2pxi Þ þ 10Þ2
f1 ðxÞ ¼ maxi fjxi j; 1  i  ng 30 [100,100]n 0

corresponding cumulative distribution functions Fi(yi) with corre-

lation matrix of Cy is available according to historical data obtained Table 3
Comparison of results for Case1.
from data sampling and numerical studies:
Method Fuel cost($/h) Emission(ton/h)
2 3
r11 r12 … r1n BBO 798.9005 0.2047
6 r21 r22 … r2n 7
Cy ¼ 6
4 «
7 (37) LTLBO [43] 799.4369 0.2047
« 1 « 5 ABC [44] 800.66 0.204826
rn1 rn2 … rnn SFLA [45] 802.21 0.2063
MSFLA [45] 802.287 0.2056

where, rij is correlated coefficient between ith and jth random

variables. The marginal transformation below is used to obtain the where, L0 is a lower triangular matrix which can be used to obtain
vector of input random variables in standard normal distribution: independent vector U from correlated vector W:
Fðwi Þ ¼ Fi ðyi Þ
wi ¼ F1 ðFi ðyi ÞÞ
where, F(.) and F1(.) are cumulative distribution function and
inverse cumulative function of the standard normal distribution.
Using marginal transformation, random vector Y is mapped to
vector W in standard normal space which will be correlated with
correlation matrix of Cw. For calculation of Cw, the following
empirical expression is proposed in Ref. [38]:
 
rij ¼ T g; r0ij r0ij
U ¼ L1
(38) 0 W (42)
Accordingly, independent vector U in standard normal space is
obtained form correlated vector Y in arbitrary space using the Nataf
transformation by Eqs (38)e(42). In this paper, inverse Nataf
transformation is used to map the uncorrelated vector of location
points from standard normal space to correlated Weibull distri-
bution and normal distribution. Fig. 2 presents flowchart of the
proposed method and explains the process of the method clearly.
(39) 5. Simulation results

where, r0ij and rij are the non-diagonal elements of the Cy and Cw,
respectively. Note that diagonal elements of correlation matrix are
1. g is the coefficient of variation, that is g ¼ s/m. In Refs. [38,39], the
authors provided 49 empirical formulas for calculating T among 10
different probability distributions.
For the two parameter Weibull distribution (wind speed), T has
the following formulation [38]:
   
T g; r0ij ¼ 1:063  0:004r0ij  0:200 gi þ gj  0:001r20ij
   
þ 0:337 g2i þ g2j þ 0:007r0ij gi þ gj  0:007gi gj
It should be mentioned that for normal distribution (load po-
wer) T is 1. After calculating the correlation matrix Cw, Cholesky
decomposition is used to factor the Cw as follows:
Cw ¼ L0 LT0
OPF problem Dou to nonlinear interaction among parameters
may have nonlinear object function and constraints and may have
non-convex feasible region. Dou to nonlinearity and non-convexity
of the OPF problem global optimal cannot be guaranteed. In order
to investigate capability of the optimization algorithms benchmark
functions are used. In this paper, In order to test the efficiency of the
proposed BBO algorithm, five benchmarks which are given in
Table 1 are employed. Mean and standard deviation of 100 times
run for each of the benchmarks are presented in Table 2. These
results are compared with results of genetic algorithm (GA) [40],
particle swarm optimization (PSO) [40] and group search optimizer
(40) (GSO) [40]. According to this table BBO converged to better result
with less standard deviation.
The standard IEEE 30-bus test case is used to illustrate the
proposed algorithm. This system has 30 buses, 41 transmission
lines, 6 generation units, 4 tap changing transformers and 9 shunt
capacitors. Bus and line data of the test case system can be found in
(41)
Ref. [42]. Coefficients used in the objective function of the emission
released from generation units can be found in Ref. [29].

Table 2
Generators cost and emission coefficients.

F. Method

GA [40] PSO [40] GSO [40] BBO

Mean Std. Mean Std. Mean Std. Mean Std.

f1 338.55 361.5 37.36 32.14 49.8 30.2 0.0257 0.0606

f2 9749.9 2594.9 1.2  103 5.78 5.78 3.68 5.2209 2.1971
f3 3.697 1.952 0.146 0.418 0.0016 0.133 3.82  105 2.32  105
f4 0.651 0.360 20.790 5.94 1.018 0.901 8.17  107 2.87  107
f5 7.961 1.506 0.412 0.250 0.108 0.004 0.0407 0.0097
16 S. Shargh et al. / Renewable Energy 94 (2016) 10e21

Table 4
Wind turbine parameters.

Parameter Rating capacity (MW) Cut in speed (m/s) Cut out speed (m/s) Rating speed (m/s)

Value 2.5 3 25 12.5

Table 5
Comparison of mean value for different wind speed correlation coefficients using PEM.

Active power generated Cl ¼ 0.1

Cw ¼ 0.0 Cw ¼ 0.1 Cw ¼ 0.3 Cw ¼ 0.5 Cw ¼ 0.7 Cw ¼ 0.9

PG1 (MW) 118.6245 118.7315 118.6186 118.6938 118.8551 118.8757

PG2 (MW) 56.8939 56.6676 56.5926 56.8394 56.5694 56.3650
PG5 (MW) 25.8901 26.0182 26.0937 25.9984 25.9784 26.1696
PG8 (MW) 35.0000 35.0000 35.0000 35.0000 35.0000 35.0000
PG11 (MW) 23.4598 23.5382 23.3867 23.6013 23.5298 23.5595
PG13 (MW) 23.0684 22.9754 23.2054 22.8209 23.0423 22.9915
Fuel cost ($/h) 804.7286 804.7404 804.7938 804.8460 804.8515 804.9104
Emission (ton/h) 0.250908 0.251014 0.250836 0.251056 0.251168 0.251265

Table 6
Comparison of standard deviation for different wind speed correlation coefficients using PEM.

Active power generated Cl ¼ 0.1

Cw ¼ 0.0 Cw ¼ 0.1 Cw ¼ 0.3 Cw ¼ 0.5 Cw ¼ 0.7 Cw ¼ 0.9

PG1 (MW) 2.9014 3.4523 3.5999 3.7354 3.8124 3.8616

PG2 (MW) 1.41412 1.6687 1.7320 1.7868 1.8354 1.877
PG5 (MW) 0.57747 0.6713 0.6923 0.7131 0.7320 0.7382
PG8 (MW) 6.10e-8 9.93e-9 2.58e-8 1.25e-7 3.44e-7 5.76e-7
PG11 (MW) 1.18793 1.4265 1.4871 1.5433 1.5775 1.6005
PG13 (MW) 1.1242 1.3351 1.3917 1.4409 1.4811 1.4863
Fuel cost ($/h) 25.7331 30.5682 31.7853 32.9117 33.6888 34.0807
Emission (ton/h) 0.00271 0.00323 0.00336 0.00348 0.00354 0.00357

Matpower4.1 is used for load flow calculations. Voltage limit and tap
of transformers' limits are considered in the range of [0.95,1.1] p.u.
and [0.9,1.1], respectively. Reactive power injected by shunt ca-
pacitors is limited to 0.05 pu with the step size of 0.01 pu.
5.1. Assessment of BBO algorithm in solving deterministic OPF
problem
In this section, the performance of BBO is evaluated in solving
deterministic OPF in introduced test system. In this case no uncer-
tainty and correlation is considered for sake of comparison of the
proposed solution method with the available methods. In this case
fuel cost and released emission of generation units of system is
considered as objectives. Algorithm parameters which are used in
this study case consist of 20 habitats, 500 iteration, keeping rate of
0.4 and mutation limit 2% of variable limitation. These results are
tabulated in Table 3 in comparison of many other research results.
This comparison demonstrates effectiveness of BBO algorithm in
solving OPF problem.
Table 7
Comparison of mean value for different load correlation coefficients using PEM.
5.2. Probabilistic MO-OPF and correlation
In this paper, the active load demand is modeled as normal
distribution with mean value of nominal load at load points of
system and standard deviation equal to 5% of mean values [25]. The
power factor assumed to be 0.85 in load points and reactive power
demand varies corresponding to active power demand according to
the power factor [25]. Two wind farms including 4 wind turbines
were added to this system and connected to buses 29 and 30 both
via a transmission line with the impedance of 0.01 þ j0.01 p.u.
Information of wind turbines are given in Table 4.
In this paper the wind speed is approximated with two
parameter Weibull distribution, the scale parameter is 7.28 and
shape parameter is 2.01 [25].
In order to demonstrate effect of correlation of wind speed and
load, separately, on control variables, fuel cost and emission, two
cases are defined.
In the first case, effect of wind speed on output variables is
observed. In this case correlation among load points is considered
Table 8
Comparison of standard deviation for different load correlation coefficients using
PEM.

Active power generated Cw ¼ 0.7 Active power generated Cw ¼ 0.7

Cl ¼ 0.0 Cl ¼ 0.5 Cl ¼ 0.9 Cl ¼ 0.0 Cl ¼ 0.5 Cl ¼ 0.9

PG1 (MW) 118.8492 118.5745 118.7102 PG1 (MW) 3.3998 5.1627 6.0256
PG2 (MW) 56.5677 56.8417 56.9682 PG2 (MW) 1.6317 2.4826 2.9083
PG5 (MW) 26.0050 26.9290 26.0275 PG5 (MW) 0.6526 0.9914 1.1499
PG8 (MW) 35.0000 35.0000 34.6831 PG8 (MW) 6.08e-8 3.23e-8 0.7086
PG11 (MW) 23.0233 23.5634 23.4872 PG11 (MW) 1.3955 2.1489 2.5233
PG13 (MW) 23.0233 22.9042 23.1068 PG13 (MW) 1.3087 2.0221 2.3728
Fuel cost ($/h) 804.7959 805.2633 805.4185 Fuel cost ($/h) 29.9023 45.7929 55.4595
Emission (ton/h) 0.251123 0.251077 0.251326 Emission (ton/h) 0.00315 0.00482 0.00568
 S. Shargh et al. / Renewable Energy 94 (2016) 10e21 17

Fig. 3. Variation of normalized mean output value and control variables with corre-
Fig. 5. The convergence of Monte Carlo results for mean values.
lation change.

increase of correlation coefficient results in the increase of standard

Fig. 4. Variation of normalized standard deviation of output and control variables with
change of correlation.
to be constant. Table 5 shows the effect of correlation on mean
value of power generated by generation units, fuel cost and emis-
sion. According to this table, increase of correlation among wind
generation affects active power generation and this changes the
total fuel cost and emission of system. It is clear that changes in
generated power due to increase of correlation coefficient causes
fuel cost and emission to increase, in most of the times.
Table 6 shows the standard deviation of output variables. In this
table, impact of correlation among wind generation is clearly re-
flected. According to this table, increase of correlation coefficient
causes standard deviation to increase in most cases. Moreover
deviation of fuel cost and emission.
Second case demonstrates effect of load correlation on the
output variables of MO-OPF problem. Table 7 gives the results of
mean value of active power generated, fuel cost and emission. Ac-
cording to this table, increase of load correlation coefficient
changes the mean value of output variables of system and Table 8
presents standard deviation of these variables. According to this
table, increase of load correlation coefficient causes standard de-
viation to increase.
Figs. 3 and 4 present normalized mean value and standard de-
viation of output and control variables considering different cor-
relation coefficients of wind power. In these figures, value of
variables are normalized based on the uncorrelated values of each
variable for better presentation of correlation effect. These figures
clearly show the effect of correlation on output and control vari-
ables of MO-OPF problem. According to Fig. 3 mean value of fuel
cost which is increased 0.14%, is affected more than mean values of
power and emission. In Fig. 4 it is clear that the increase of standard
deviation, which is increased more than 30%, is much more than
the increase of mean values. According to this graph, gradient of
standard deviation in terms of correlation coefficient decreases by
increase of correlation coefficient.
5.3. Accuracy
In order to demonstrate accuracy and effectiveness of proposed
method, the obtained results are compared with the results of
Monte Carlo simulation (MCS) which is considered as the reference
method. To do this, the mean and standard deviation of output
variables, obtained by proposed method are compared with those
of MCS with 2000 runs. In this regard, the errors for mean and

Table 9
Average error of moments for different correlation coefficients with 2000 runs of MCS.

Output variables Cw ¼ 0.0 Cw ¼ 0.7 Cw ¼ 0.7

Cl ¼ 0.0 Cl ¼ 0.1 Cl ¼ 0.9

PG (%) εm 0.2414 0.2862 0.3772

εs 2.0167 2.3136 2.4454
Fuel Cost (%) εm 0.0264 0.0281 0.0299
εs 1.0562 1.1271 1.2153
Emission (%) εm 0.0621 0.0712 0.0783
εs 0.5376 0.6064 0.6645
18 S. Shargh et al. / Renewable Energy 94 (2016) 10e21

deviation of MCSs for reported variables, which are shown in Fig. 5

and Fig. 6, demonstrate that MCS with 2000 iteration of runs is
accurate reference for this comparison. According to this table,
proposed method has acceptable error and is almost as accurate as
MCS. With respect to the errors for different values of correlation, it
is clear that the correlation affect PEM simulation error, this
method is more accurate in solving POPF with less correlation than
high correlation.

6. Conclusion

This paper, proposes a PEM based Nataf transformation to solve

probabilistic MO-OPF problem considering fuel cost and emission
as objectives. Uncertainties in the wind power output and load
demand are taken into account. The main contribution of the work
reported here is to apply the Nataf transformation to the PEM in
order to solve probabilistic MO-OPF problem with correlated input
Fig. 6. The convergence of Monte Carlo results for standard deviation. RVs. This method only requires data for the marginal distribution
function of each input RV and the correlation coefficients instead of
their joint PDF. In this paper, effect of different correlation co-
standard deviation are defined using following equations. efficients are observed on the control and output data of problem
  and on accuracy of PEM. It is concluded that correlation among
100ðmMCS  mÞ input RVs increases uncertainty in control and output data of MO-
εm ¼  ½%
 (43)
mMCS OPF problem which makes planning and forecast more complicated
than before. The effectiveness of the method is demonstrated using
  a 30-bus test system. Results of proposed method are compared to
100ðsMCS  sÞ
εs ¼  ½%
 (44) those of MCS which confirms high accuracy of the method.
s MCS

Then errors for mean and standard deviation of generated power
of generation units, fuel cost and emission are calculated using these
equations. Table 9 presents the errors of POPF results for different
amounts of correlation coefficients and also with no correlation. In
this table, average error for generation units is presented.
Convergence characteristics of both mean and standard
Appendix
The one line diagram of the IEEE 30 bus test system is shown in
Fig. 7. Bus data, branch data, generator data, tap setting data, shunt
capacitor data and cost and emission coefficients of this network
are presented in the following tables.

1 2 5 7 8

3 4 6 28

13 12 10 9 11 27

14 15 16 17 22 29 30

23 18 19 20 21 24 25

26

Fig. 7. One line diagram of the IEEE 30 bus test system.

Table 10
Bus data of IEEE 30 bus test system.

Bus Bus Load Bus voltage

Bus No Bus type Real  power (MW) Reactive power (MVar) Voltage magnitude (p.u.) Voltage angle (degree) Base voltage (KV)

1 3 0 0 1.0600 0 132.00
2 2 21.7000 12.7000 1.0430 5.4800 132.00
3 1 2.4000 1.2000 1.0210 7.9600 132.00
4 1 7.6000 1.6000 1.0120 9.6200 132.00
5 2 94.2000 19.0000 1.0100 14.3700 132.00
6 1 0 0 1.0100 11.3400 132.00
7 1 22.8000 10.9000 1.0020 13.1200 132.00
8 2 30.0000 30.0000 1.0100 12.1000 132.00
9 1 0 0 1.0510 14.3800 1.00
10 1 5.8000 2.0000 1.0450 15.9700 33.00
11 2 0 0 1.0820 14.3900 11.00
12 1 11.2000 7.5000 1.0570 15.2400 33.00
13 2 0 0 1.0710 15.2400 11.00
14 1 6.2000 1.6000 1.0420 16.1300 33.00
15 1 8.2000 2.5000 1.0380 16.2200 33.00
16 1 3.5000 1.8000 1.0450 15.8300 33.00
17 1 9.0000 5.8000 1.0400 16.1400 33.00
18 1 3.2000 0.9000 1.0280 16.8200 33.00
19 1 9.5000 3.4000 1.0260 17.0000 33.00
20 1 2.2000 0.7000 1.0300 16.8000 33.00
21 1 17.5000 11.2000 1.0330 16.4200 33.00
22 1 0 0 1.0330 16.4100 33.00
23 1 3.2000 1.6000 1.0270 16.6100 33.00
24 1 8.7000 6.7000 1.0210 16.7800 33.00
25 1 0 0 1.0170 16.3500 33.00
26 1 3.5000 2.3000 1.0000 16.7700 33.00
27 1 0 0 1.0230 15.8200 33.00
28 1 0 0 1.0070 11.9700 132.00
29 1 2.4000 0.9000 1.0030 17.0600 33.00
30 1 10.6000 1.9000 0.9920 17.9400 33.00

Table 11
Branch data of IEEE 30 bus test system.

Line No From bus To bus Resistance (p.u.) Reactance (p.u.) Susceptance (p.u.) MVA rating

1 1 2 0.0192 0.0575 0.0528 130

2 1 3 0.0452 0.1652 0.0408 130
3 2 4 0.0570 0.1737 0.0368 65
4 3 4 0.0132 0.0379 0.0084 130
5 2 5 0.0472 0.1983 0.0418 130
6 2 6 0.0581 0.1763 0.0374 65
7 4 6 0.0119 0.0414 0.0090 90
8 5 7 0.0460 0.1160 0.0204 70
9 6 7 0.0267 0.0820 0.0170 130
10 6 8 0.0120 0.0420 0.0090 32
11 6 9 0 0.2080 0 65
12 6 10 0 0.5560 0 32
13 9 11 0 0.2080 0 65
14 9 10 0 0.1100 0 65
15 4 12 0 0.2560 0 65
16 12 13 0 0.1400 0 65
17 12 14 0.1231 0.2559 0 32
18 12 15 0.0662 0.1304 0 32
19 12 16 0.0945 0.1987 0 16
20 14 15 0.2210 0.1997 0 16
21 16 17 0.0524 0.1923 0 16
22 15 18 0.1073 0.2185 0 16
23 18 19 0.0639 0.1292 0 16
24 19 20 0.0340 0.0680 0 16
25 10 20 0.0936 0.2090 0 32
26 10 17 0.0324 0.0845 0 32
27 10 21 0.0348 0.0749 0 32
28 10 22 0.0727 0.1499 0 32
29 21 22 0.0116 0.0236 0 32
30 15 23 0.1000 0.2020 0 16
31 22 24 0.1150 0.1790 0 16
32 23 24 0.1320 0.2700 0 16
33 24 25 0.1885 0.3292 0 16
34 25 26 0.2544 0.3800 0 16
35 25 27 0.1093 0.2087 0 16
36 28 27 0 0.3960 0 65
37 27 29 0.2198 0.4153 0 16
38 27 30 0.3202 0.6027 0 16
39 29 30 0.2399 0.4533 0 26
40 8 28 0.0636 0.2000 0.0428 32
41 6 28 0.0169 0.0599 0.0130 32
20 S. Shargh et al. / Renewable Energy 94 (2016) 10e21

Table 12
Generator data of IEEE 30 bus test system.

Bus No Generation power Generation limit Voltage (p.u.)

Real power (MW) Reactive power (MVar) Pmin (MW) Pmax (MW) Qmin (MVar) Qmax (MVar)

1 160.20 16.10 50.00 250.00 e e 1.0600

2 40.00 50.00 20.00 80.00 50.00 40.00 1.0450
5 0 37.00 15.00 50.00 40.00 40.00 1.0100
8 0 37.30 10.00 35.00 40.00 10.00 1.0100
11 0 16.20 10.00 30.00 24.00 6.00 1.0820
13 0 10.60 12.00 40.00 24.00 6.00 1.0710

[8] S.M. Mohseni-Bonab, A. Rabiee, B. Mohammadi-Ivatloo, Voltage stability

constrained multi-objective optimal reactive power dispatch under load and
Table 13 wind power uncertainties: a stochastic approach, Renew. Energy 85 (2016)
Transformers tap setting data. 598e609.
[9] S. Mohseni-Bonab, A. Rabiee, S. Jalilzadeh, B. Mohammadi-Ivatloo, S. Nojavan,
From bus To bus Tap setting value (p.u.)
Probabilistic multi objective optimal reactive power dispatch considering load
6 9 0.978 uncertainties using monte carlo simulations, J. Oper. Autom. Power Eng. 3 (1)
6 10 0.969 (2015) 83e93.
4 12 0.932 [10] X. Li, Y. Li, S. Zhang, Analysis of probabilistic optimal power flow taking ac-
count of the variation of load power, Power Syst. IEEE Trans. 23 (3) (2008)
28 27 0.968
992e999.
[11] M. Madrigal, K. Ponnambalam, V. Quintana, Probabilistic optimal power flow,
in: Electrical and Computer Engineering, 1998, IEEE Canadian Conference on,
Table 14 vol. 1, IEEE, 1998, pp. 385e388.
[12] H. Yu, W. Rosehart, An optimal power flow algorithm to achieve robust
Shunt capacitors data.
operation considering load and renewable generation uncertainties, Power
Bus No. Susceptance (p.u.) Syst. IEEE Trans. 27 (4) (2012) 1808e1817.
[13] A.B. Rodrigues, M.G. Da Silva, Probabilistic assessment of available transfer
10 0.19 capability based on monte carlo method with sequential simulation, Power
24 0.043 Syst. IEEE Trans. 22 (1) (2007) 484e492.
[14] H. Zhang, P. Li, Probabilistic analysis for optimal power flow under uncer-
tainty, IET Gen. Transm. Distrib. 4 (5) (2010) 553e561.
[15] G. Carpinelli, P. Caramia, P. Varilone, Multi-linear monte carlo simulation
method for probabilistic load flow of distribution systems with wind and
Table 15 photovoltaic generation systems, Renew. Energy 76 (2015) 283e295.
Generators cost and emission coefficients. [16] A. Schellenberg, W. Rosehart, J. Aguado, Cumulant-based probabilistic optimal
power flow (p-opf) with gaussian and gamma distributions, Power Syst. IEEE
Unit Cost coefficients Emission coefficients Trans. 20 (2) (2005) 773e781.
[17] L. Shi, C. Wang, L. Yao, Y. Ni, M. Bazargan, Optimal power flow solution
ai bi ci ai bi gi
incorporating wind power, Syst. J. IEEE 6 (2) (2012) 233e241.
($/MWh2) ($/MWh) ($/h) (Kg/MWh2) (Kg/MWh) (Kg/h) [18] A. Schellenberg, W. Rosehart, J. Aguado, Introduction to cumulant-based
probabilistic optimal power flow (p-opf), Power Syst. IEEE Trans. 20 (2)
1 0.00375 2.00 0 0.0126 1.1000 22.983 (2005) 1184e1186.
2 0.01750 1.75 0 0.0200 0.1000 22.313 [19] D. Villanueva, A.E. Feijo o, J.L. Pazos, An analytical method to solve the prob-
5 0.06250 1.00 0 0.0270 0.1000 25.505 abilistic load flow considering load demand correlation using the dc load flow,
8 0.00834 3.25 0 0.0291 0.0050 24.900 Electr. Power Syst. Res. 110 (2014) 1e8.
11 0.02500 3.00 0 0.0290 0.0400 24.700 [20] E. Rosenblueth, Point estimates for probability moments, Proc. Natl. Acad. Sci.
13 0.02500 3.00 0 0.0271 0.0055 25.300 72 (10) (1975) 3812e3814.
[21] H. Hong, An efficient point estimate method for probabilistic analysis, Reliab.
Eng. Syst. Saf. 59 (3) (1998) 261e267.
[22] X. Li, J. Cao, D. Du, Probabilistic optimal power flow for power systems
considering wind uncertainty and load correlation, Neurocomputing 148
(2015) 240e247.
[23] M. Aien, M. Fotuhi-Firuzabad, M. Rashidinejad, Probabilistic optimal power
flow in correlated hybrid windephotovoltaic power systems, Smart Grid IEEE
Trans. 5 (1) (2014) 130e138.
[24] M. Aien, M. Rashidinejad, M.F. Firuz-Abad, Probabilistic optimal power flow in
References correlated hybrid wind-pv power systems: A review and a new approach,
Renew. Sustain. Energy Rev. 41 (2015) 1437e1446.
[25] C. Chen, W. Wu, B. Zhang, H. Sun, Correlated probabilistic load flow using a
[1] N. Hatziargyriou, A. Zervos, Wind power development in Europe, Proc. IEEE 89 point estimate method with nataf transformation, Int. J. Electr. Power & En-
(12) (2001) 1765e1782. ergy Syst. 65 (2015) 325e333.
[2] M. Moradi-Dalvand, B. Mohammadi-Ivatloo, N. Amjady, H. Zareipour, [26] Y. Li, W. Li, W. Yan, J. Yu, X. Zhao, Probabilistic optimal power flow considering
A. Mazhab-Jafari, Self-scheduling of a wind producer based on information correlations of wind speeds following different distributions, Power Syst. IEEE
gap decision theory, Energy 81 (2015) 588e600. Trans. 29 (4) (2014) 1847e1854.
[3] E. Vladislavleva, T. Friedrich, F. Neumann, M. Wagner, Predicting the energy [27] C.S. Saunders, Point estimate method addressing correlated wind power for
output of wind farms based on weather data: important variables and their probabilistic optimal power flow, Power Syst. IEEE Trans. 29 (3) (2014)
correlation, Renew. Energy 50 (2013) 236e243. 1045e1054.
[4] A. Zagouras, H.T. Pedro, C.F. Coimbra, On the role of lagged exogenous vari- [28] M. Li, Q. Wu, T. Ji, H. Rao, Stochastic multi-objective optimization for
ables and spatioetemporal correlations in improving the accuracy of solar economic-emission dispatch with uncertain wind power and distributed
forecasting methods, Renew. Energy 78 (2015) 203e218. loads, Electr. Power Syst. Res. 116 (2014) 367e373.
[5] M.A. Abdullah, A. Agalgaonkar, K.M. Muttaqi, Probabilistic load flow incor- [29] A. Shabanpour-Haghighi, A.R. Seifi, T. Niknam, A modified teachingelearning
porating correlation between time-varying electricity demand and renewable based optimization for multi-objective optimal power flow problem, Energy
power generation, Renew. Energy 55 (2013) 532e543. Convers. Manag. 77 (2014) 597e607.
[6] I. Segura-Heras, G. Escriva -Escriv
a, M. Alc
azar-Ortega, Wind farm electrical [30] R. Azizipanah-Abarghooee, T. Niknam, M. Malekpour, F. Bavafa, M. Kaji,
power production model for load flow analysis, Renew. Energy 36 (3) (2011) Optimal power flow based tu/chp/pv/wpp coordination in view of wind
1008e1013. speed, solar irradiance and load correlations, Energy Convers. Manag. 96
[7] G. Verbic, C.A. Can ~ izares, Probabilistic optimal power flow in electricity
(2015) 131e145.
markets based on a two-point estimate method, Power Syst. IEEE Trans. 21 (4) [31] D. Lalas, H. Tselepidaki, G. Theoharatos, An analysis of wind power potential in
(2006) 1883e1893.
 S. Shargh et al. / Renewable Energy 94 (2016) 10e21
greece, Sol. Energy 30 (6) (1983) 497e505.
[32] A. Garcia, J. Torres, E. Prieto, A. De Francisco, Fitting wind speed distributions:
a case study, Sol. Energy 62 (2) (1998) 139e144.
[33] A. Rabiee, A. Soroudi, B. Mohammadi-ivatloo, M. Parniani, Corrective voltage
control scheme considering demand response and stochastic wind power,
IEEE Trans. Power Syst. 29 (6) (2014) 2965e2973.
[34] A. Soroudi, B. Mohammadi-Ivatloo, A. Rabiee, Energy hub management with
intermittent wind power, in: Large Scale Renewable Power Generation,
Springer, 2014, pp. 413e438.
[35] D. Simon, Biogeography-based optimization, IEEE Trans. Evol. Comput. 12 (6)
(2008) 702e713.
[36] M. Rosenblatt, Remarks on a multivariate transformation, Ann. Math. Stat.
(1952) 470e472.
[37] H. Li, Z. Lü, X. Yuan, Nataf transformation based point estimate method, Chin.
Sci. Bull. 53 (17) (2008) 2586e2592.
[38] P.L. Liu, A. Der Kiureghian, Multivariate distribution models with prescribed
marginals and covariances, Probab. Eng. Mech. 1 (2) (1986) 105e112.
[39] A. Der Kiureghian, P.L. Liu, Structural reliability under incomplete probability
information, J. Eng. Mech. 112 (1) (1986) 85e104.
21
[40] S. He, Q.H. Wu, J. Saunders, Group search optimizer: an optimization algo-
rithm inspired by animal searching behavior, Evol. Comput. IEEE Trans. 13 (5)
(2009) 973e990.
[41] M. Moradi-Dalvand, B. Mohammadi-Ivatloo, A. Najafi, A. Rabiee, Continuous
quick group search optimizer for solving non-convex economic dispatch
problems, Electr. Power Syst. Res. 93 (2012) 93e105.
[42] O. Alsac, B. Stott, Optimal load flow with steady-state security, Power App.
Syst. IEEE Trans. 3 (1974) 745e751.
[43] M. Ghasemi, S. Ghavidel, M. Gitizadeh, E. Akbari, An improved teach-
ingelearning-based optimization algorithm using le vy mutation strategy for
non-smooth optimal power flow, Int. J. Electr. Power & Energy Syst. 65 (2015)
375e384.
[44] M. Rezaei Adaryani, A. Karami, Artificial bee colony algorithm for solving
multi-objective optimal power flow problem, Int. J. Electr. Power Energy Syst.
53 (2013) 219e230.
[45] T. Niknam, M. Jabbari, A.R. Malekpour, et al., A modified shuffle frog leaping
algorithm for multi-objective optimal power flow, Energy 36 (11) (2011)
6420e6432.