See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/282209538

An analytical method for optimal sizing of

different types of DG in a power distribution
system

Article · March 2015

DOI: 10.1109/PECON.2014.7062462

CITATIONS READS

0 105

5 authors, including:

Maryam Mirzaei Jasronita Jasni

Universiti Putra Malaysia Universiti Putra Malaysia
11 PUBLICATIONS 84 CITATIONS 119 PUBLICATIONS 460 CITATIONS

SEE PROFILE SEE PROFILE

Hashim Hizam Noor Izzri Abdul Wahab

Universiti Putra Malaysia Universiti Putra Malaysia
168 PUBLICATIONS 1,196 CITATIONS 98 PUBLICATIONS 303 CITATIONS

SEE PROFILE SEE PROFILE

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

Towards TRIZ for Electrical Engineers View project

Power System Stability and Control View project

All content following this page was uploaded by Maryam Mirzaei on 02 February 2016.

The user has requested enhancement of the downloaded file.

 2014 IEEE International Conference Power & Energy (PECON)

An Analytical Method for Optimal Sizing of Different

Types of DG in a Power Distribution System

Maryam Mirzaei¹*, Jasronita Jasni², Hashim Hizam¹, Noor Izzri Abdul Wahab¹, Salah Eldeen Gasim Mohamed¹
¹Department of Electrical Power Engineering, Faculty of Engineering, University Putra Malaysia,
43400 UPM Serdang, Selangor, Malaysia
²CELP, Faculty of Engineering, University Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia
*Email: m.mirzaei.eh@gmail.com

Abstract—This paper presented a hybrid method to find
the optimal size of the distributed generation by considering
different type of renewable sources in the distribution power
system. The proposed method is based on the combination of
two methods like the sensitivity analysis and continues power
factor method. The simulated method is done on the Matlab
environment by applying in the 16-bus distribution system.
The result illustrated the accurate and reliable size of the
different types of the distributed generation in a power system.
Keywords—continuous power flow; exact loss formula;
optimum size; optimum location; different types of DG
I. INTRODUCTION
Traditionally, electric power is produced at central station
power plants and delivered to consumers using transmission
and distribution networks. Today, there is a trend toward the
determine the optimal size and placing of DG as a high
priority issue to maximize the system efficiency.
The optimal and fast placement and assignment DG units
have become mandatory to properly assess the state of the
system by engineer with the necessary tools and techniques
to ensure that the power system operates within the specified
design parameters. So, presents an optimal and fast method
for placing and sizing of DGs is very important in recent
years. As matter of fact, the number of this researches are
considerable that it is shows the significant impact of the
discussion on this area. The Fig. 1, illustrations the number
of manuscript that was provided by authors. As you can see
in this figure, in the last 2 years, number of publication is
rose which it means the important of this subject by wide
increasing distributed renewable resources.
22
Number of citations

use of distributed generation (DG) units, for economic, 20

technical and environmental reasons, in addition to the
18
traditional large generators connected to the transmission
16
system [1]. Accordingly, it is expected that in the near future
14
DGs will have a significant contribution in electrical power
12
systems. DG units are smaller scale with different
10
penetration level based on locally available resources and
8
therefore they are mostly connected at the distribution level.
When the DG penetration is high, the generated power of 6

DG units changes the power flow in the distribution system, 4

and also in the transmission system as well. DGs have 2

several advantages that the major advantages of DG are as 0

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

power loss reduction, voltage profile improvement,
reliability and security improvement, power quality
improvement, stability enhancement, environmental
concerns and investment risk reduction [2].
Moreover, the DG connection to the network may
influence the stability of the power system, i.e., angle,
frequency, and voltage stability [3, 4]. Also, finding the
suitable placement to connect the DG to the power system
lead to take the many advantages of the DG integrated power
systems in comparison to the classical power systems [5].
Besides, the non-optimal DG locating and sizing may lead to
bad results in the system which are include increasing system
losses and having a bad effect on the voltage profile and
growing costs, and consequently an opposite effect from
what is expected. Therefore, it is important to allocate and
Fig. 1. The number of manuscript citations
Basically, there are different approaches with existing
methods on the basis of voltage profile improvement,
reduction in power system losses, maximization of system
loadability, find the optimal power factor and maximization
of bus and line stability. Generally, these methods can be
classified in three broad categories, which are conventional
optimization methods, meta-heuristic based optimization
methods and hybrid method. There are different
conventional approaches to find the optimal size and
locations of DG units in power distribution systems, which
are considered here to include the analytical approaches [6-
12], Continuation Power Flow methods [13, 14], the grid

978-1-4799-7297-5/14/$31.00 ©2014 IEEE 309

 2014 IEEE International Conference Power & Energy (PECON)

search method [15], Reactive power optimization [16],
Optimal Power Flow [17], Mixed Integer Nonlinear
Programming [18], Primal-Dual Interior Point Method [19],
and Deterministic optimization techniques [20]. On the other
hand, the meta-heuristic methods have been widely used by
researchers in recent years as these approaches are intuitive,
easy to understand, simple to implement and address the
integer variable very well as compared to the conventional
method and analytical programming. However, the results
produced are not guaranteed to be optimal and the process
speed can be slower.
Some examples of the meta-heuristic optimization
classes are genetic algorithms (GAs) [21, 22], The
Evolutionary Programming [23], tabu search [24], simulated
the important DG influence is on the power loss, which its
increasing lead to the serious effects on voltage profile and
growing the costs. This effect is shown in Fig. 2. According
to the figure, the effect of the DG size on system losses is a
few at first, and then the power loss is very sensitive to the
size of the DG at the lateral buses. As you can see, by
growing the DG penetration, the power loss is increased
again. Therefore, it can conclude that the correctly estimated
the size of DG has the important rule to decrease the power
loss in the system.
0.4
0.35
0.3

annealing [25], particle swarm optimization methods [26- 0.25

PLoss (p.u.)
28], ant colony optimization [29], Cuckoo Search [30],
0.2
bacterial foraging optimization [31] which have all been
applied in most optimization problems as well as DG 0.15

optimal placement and sizing problems. The Tabu Search is 0.1

an efficient combinatorial method which can be used to
achieve an optimal or a sub optimal solution within a
reasonably short duration. The Evolutionary Programming,
Ant Colony Algorithm, Simulated Annealing and Cuckoo
Search have still not been paid much attention by
researchers. GA is competent at obtaining a solution near
the global minima which is computationally intensive.
The hybrid methods are also a combination of the above
approaches which includes the Genetic-Fuzzy [32], Genetic-
Particle Swarm [33], Genetic-Optimal Power Flow [34],
Particle Swarm Optimization-Optimal Power Flow [35], and
Analytical-Fuzzy [36] techniques. The performance of the
hybrid algorithm is a fast and efficient approach because of
the improved versions. Actually, the analytical techniques
are easily implemented with high efficiency in terms of
computational time and best precision factor while OPF,
CPF and meta-heuristic approaches are the next advanced
level.
This paper presents an analytical method to evaluate the
impact of optimal sizing of different types of DG units and
then placing the DG units. This method is based on the
analysis of the exact loss formula and continuous power
flow (CPF) in a radial distribution system. The simulation
results are obtained on a 16-bus radial distribution system in
the Matlab environment to determine the location and size
of different types of DG units.
0.05
0
0 10 20 30 40 50 60
DG penetration level (%)
Fig. 2. Effect of size and location of DG on power loss
Another important aspect of DG is on the voltage
stability. To determine the most sensitive bus to voltage
collapse or maximum loading, continuation power-flow
program is used. Voltage stability margin is defined from
current loading level or operating point (λ) to voltage
collapse point or the critical point (CP) to evaluate the
performance of the system as a Fig. 3.
By increasing the active power injected with DG unit, the
reactive power loss is decreased. So, by rising the voltage to
the operating point from (V1) to (V2), then (CP1) is
increased to (CP2). Therefore, the maximum loadability
(λmax1) is a growth to (λmax2). Consequently, it is lead to
improve the voltage stability of the system.

II. DIFERENT IMPACT OF DG

The distribution system was initially designed such that
the power flows from the sending end to the load, conductor
sizes were gradually decreased from the substation to the
consumer side. So, the size of the distribution system in term
of load (MW) plays an important role in selecting the size of
DG units.
Fig. 3. The behavior of P-V curves and the relationship to voltage stability
A. DG can have a significant impact on:
The power-flow, losses, voltage profile, stability and
quality of power supply for customers and electricity
suppliers, power factor limit and reliability. In fact, one of

310
 2014 IEEE International Conference Power & Energy (PECON)

III. PROPOSED METHODOLOGY 1  N


Elgerd [37] presented a popular based formula to
PDGi = ( PDioi + λi .PDCPF
i )+  β ii Qi −
α ii 
∑ (α ij Pj − βij Q j ) 
j =1, j ≠ i 
calculate the exact power loss, which is used as a reference
in many researches. The total real power losses ( PL ) with N (6)
buses is calculated in a distribution system. On the other hand, DG units are classified to four types
based on the capable of injecting or consuming P and Q. So,
N N based on the type of DG and operation condition, the power
PL = ∑∑ α ij ( Pi .Pj + Qi .Q j ) + β ij (Qi .Pj − Pi .Q j )  factor of DG and optimal size of DG units are changed. If pf
i =1 j =1 is equal to the DG operating power factor; so, relationship
(1) between the active and reactive power injection from DG
placed to the node i, is:

where α = rij cos(δ − δ ) and β = rij sin(δ − δ ) , N 1 − pf 2

ij
VV
i j ij
VV
i j x=
i j i j pf 2
is the bus number and Pi , Qi , Pj andQ j are the active and
reactive power injections at buses i and j respectively. Q DGi = x. PDGi (7)
Rij + jX ij = Z ij is the ijth element of impedance matrix Therefore, based on the different types of DG, the
[Zbus]=[Ybus]¯ ˡ optimal size to minimize the power loss is calculated as:
The sensitivity factor of real power loss with respect to
real power injection from DG is given as follows: A. Type one of DG is capable of injecting P only. Some
example s of this type is like: Photovoltaic, micro
turbines, fuel cells. Because of the power factor is at
N unity, so, the optimum size of the DG at bus i is
∂PL
αi = = ∑ (α ij Pj − β ij Q j ) (2)
∂Pi i =1 calculated by PDGi as (6).

If the partial derivative of Equation (2) becomes zero, then B. Type two of DG is capable of injecting Q only like
the total active power loss will be minimized. synchronous compensators such as gas turbines. The DG
The active power injection of the bus i; where the DG unit is power factor is zero and x=∞, so the optimal DG size is
installed, is given by: given by:
1 N 
Pi = PDGi − PDi (3) QDGi = (QDioi + λi .QDCPF
i )+ βii Pi − ∑ (αijQj − βij Pj ) (8)
αii  j =1, j ≠i 
where, PDi is the load demand at node i and PDGi is the C. Type 3 and 4 are DG capable of injecting both P and Q
injection power from DG placed to the node i. and injecting P but consuming Q respectively.an
example of type 3 is synchronous machine as
cogeneration and type 4 example is wind farms. The DG
On the other hand, in CPF, by applying the loading power factor is based on, it is injecting (+1) or
parameter of (λ), loads and generations are increased from consuming (-1) reactive power. The optimal DG size is
their base case values as follows: given by:
 1 
PD = PDoi + λ .PDCPF (4) PDGi = ( PDioi + λi .PDCPF
i )− ×
 αii *(1 + x) 
QD = QDoi + λ.QDCPF (5)  N 
αii × x(QDi + λi .QD i ) − ∑ (αij ( Pj + xQ j ) − βij ( xPj − Q j ) 
oi CPF

 j =1, j ≠ i 
Where P CPF and Q CPF are power increment directions (9)
D D

of loads in the CPF and P oi and Q oi are powers of loads at

D D
After determine the optimal size of different types of DG
the base case [38]. Substituting Equation (3) and Equation
and calculating the losses for each case, the location is
(4) into Equation (1), the total active power losses with a DG
determined by minimum losses is for each type of DG units.
unit can be obtained the optimum size of the DG at bus I as
The flowchart of this procedure is shown in Fig. 4.
follows:

311
 2014 IEEE International Conference Power & Energy (PECON)

Start 600

500

Run the power flow and get Pl using (1) and bus

Total power loss(kW)

400
voltage profile

300

Run the continuous power flow and determine the loading

parameters 200

100

Find the optimal size of DG based on type of DG using (6,8-9) for each
bus 0
0 2 4 6 8 10 12 14 16
Bus No.

Locate the optimal bus to site the DG based on (1)

Fig. 5. Total power loss of a 16 bus distribution system

Run the load flow with the optimal size at the optimal location
with the DG unit According to the table 2, when the DG unit is installing
in the bus 9, the voltage profile is improved about 3.155%
and the power loss is decreased about the 64.02%, that it is
shown the good improving in the system condition. So, it can
No Is voltage allowable be concluded that by installing a DG unit and consider the
level? kind of DG with accurate placement, the system condition
can be improved and proposed method is effective to reach
Yes to this goal.
End

20
Fig. 4. Flowchart of the proposed method.
18

16

14
Size of DG (MVA)

12
IV. NUMERICAL RESULTS AND DISCUSSION
10

8
The proposed algorithm is tested on a 16-bus radial 6
distribution network. It is a radial system with the total load 4
of 28.7 MW and 5.9 MVAR, the load and line data are
2
shown in the appendix. The voltage level in every bus should
0
be in the acceptable voltage level (0.95 < Vbus < 1.05). 0 2 4 6 8
Bus No.
10 12 14 16

Based on the procedure presented, the system is simulated

and testing in Matlab program to reach the goal of finding Fig. 6. Optimal size of DG of a 16 bus distribution system
the optimal sizing for different types of DG and location. So,
after running the power flow, the total power loss and
optimal size of DG based on equations (1) and (6) are
calculated that are illustrated in Fig. 5, and Fig. 6, 1.02

respectively. Then, the DG is sited in the optimal place. Fig.

7, shows the voltage profile with and without DG units. 1

It can be seen that installing the DG unit, the level of the 0.98
voltage is grown to the unity and completely lands in the
V(p.u.)

allowable rang. So, the system performance in two condition 0.96

of with and without DG is illustrated the effectiveness
proposed method. In following the optimal size of DG and 0.94
power loss for different power factor in node 9 is shown without DG
with DG
table 1; so, it can conclude that the minimum power loss is 0.92
illustrated in the unity power factor. Moreover, table 2 is 2 4 6 8 10 12 14 16
presented the comparative analysis for different parameters Bus No.

with and without the DG unit.

Fig. 7. Voltage profile of a 16 bus distribution system with and without
DG units

312
 2014 IEEE International Conference Power & Energy (PECON)

TABLE I. OPTIMUM SIZES OF DG AT VARIOUS POWER FACTORS AND Load and line data in (p.u)
POWER LOSSES AT BUS 9
Bus No. R X P Q
Power Factor of DG 12 0.08 0.11 1 0.9
0.9 1.0 1.1
13 0.04 0.04 2.1 1
Optimal size (MVA) 10.206 12.943 11.005
14 0.04 0.04 _ _
Power loss (KW) 193.02 169.71 184.01
15 0.04 0.04 _ _

16 0.09 0.12 _ _
TABLE II. EFFECT OF OPTIMISATION OF DG LOCATION AND SIZING

Comparative Analysis
Without DG With DG
REFERENCES
[1] Khani D, Sadeghi Yazdankhah A, Madadi Kojabadi H. Impacts of
DG bus - 9 distributed generations on power system transient and voltage
V max 1 1 stability. International Journal of Electrical Power & Energy
Systems. 2012;43:488-500.
V min 0.9509 0.9809 [2] Hernandez J, Velasco D, Trujillo C. Analysis of the effect of the
implementation of photovoltaic systems like option of distributed
System loading 2.385 2.901 generation in Colombia. Renewable and Sustainable Energy
Reviews. 2011;15:2290-8.
Power loss (KW) 511.4 169.71
[3] Reza M, Slootweg J, Schavemaker P, Kling W, Van der Sluis L.
Power loss reduction Investigating impacts of distributed generation on transmission
- 66.82
(%) system stability. Power Tech Conference Proceedings, 2003 IEEE
Bologna: IEEE; 2003. p. 7 pp. Vol. 2.
[4] Donnelly M, Dagle J, Trudnowski D, Rogers G. Impacts of the
V. CONCLUSION distributed utility on transmission system stability. Power Systems,
IEEE Transactions on. 1996;11:741-6.
This paper proposes a method based on an analytical [5] Ugranli F, Karatepe E. Convergence of rule-of-thumb sizing and
method and continuous power flow to optimize the size of allocating rules of distributed generation in meshed power
different types of DG and find the optimal location of DG in networks. Renewable and Sustainable Energy Reviews.
a radial distribution network. A 16-bus radial distribution 2012;16:582-90.
network was used to test the placement algorithm. The [6] Acharya N, Mahat P, Mithulananthan N. An analytical approach for
DG allocation in primary distribution network. International
objective function of the proposed method was based on Journal of Electrical Power & Energy Systems. 2006;28:669-78.
reducing the power loss. The results show that by finding the [7] Borges CL, Falcao DM. Impact of distributed generation allocation
optimal DG placement, the total losses can be lowered and and sizing on reliability, losses and voltage profile. Power Tech
enhanced voltage profile increasing. Conference Proceedings, 2003 IEEE Bologna: IEEE; 2003. p. 5 pp.
Vol. 2.
[8] Gozel T, Hocaoglu MH. An analytical method for the sizing and
VI. APPENDIX
siting of distributed generators in radial systems. Electric power
systems research. 2009;79:912-8.
TABLE III. THE LOAD AND LINE DATA OF 16-BUS RADIAL [9] Hosseini RK, Kazemzadeh R. Optimal DG allocation by extending
DISTRIBUTION NETWORK an analytical method to minimize losses in radial distribution
systems. Electrical Engineering (ICEE), 19th Iranian Conference
Load and line data in (p.u) on: IEEE; 2011. p. 1-6.
Bus No. R X P Q [10] Hung DQ, Mithulananthan N, Bansal R. Analytical expressions for
1 0.075 0.1 2 1.6 DG allocation in primary distribution networks. Energy
Conversion, IEEE Transactions on. 2010;25:814-20.
2 0.08 0.11 3 1.5 [11] Hung DQ, Mithulananthan N, Bansal R. Analytical strategies for
renewable distributed generation integration considering energy
3 0.09 0.18 2 0.8 loss minimization. Applied Energy. 2013;105:75-85.
4 0.04 0.04 1.5 1.2 [12] Aman M, Jasmon G, Mokhlis H, Bakar A. Optimal placement and
sizing of a DG based on a new power stability index and line
5 0.11 0.11 4 2.7 losses. International Journal of Electrical Power & Energy Systems.
2012;43:1296-304.
6 0.08 0.11 5 3
[13] Ettehadi M, Vaez-Zadeh S, Ghasemi H. DG placement in
7 0.11 0.11 1 0.9 distribution networks considering voltage stability. Environment
and Electrical Engineering (EEEIC), 11th International Conference
8 0.11 0.11 0.6 0.1 on: IEEE; 2012. p. 41-6.
[14] Hedayati H, Nabaviniaki S, Akbarimajd A. A method for
9 0.08 0.11 4.5 2 placement of DG units in distribution networks. Power Delivery,
10 0.11 0.11 1 0.9 IEEE Transactions on. 2008;23:1620-8.
[15] Gozel T, Eminoglu U, Hocaoglu M. A tool for voltage stability and
11 0.09 0.12 1 0.7 optimization (VS&OP) in radial distribution systems using matlab

313
 2014 IEEE International Conference Power & Energy (PECON)
graphical user interface (GUI). Simulation Modelling Practice and
Theory. 2008;16:505-18.
[16] Khalesi N, Rezaei N, Haghifam M-R. DG allocation with
application of dynamic programming for loss reduction and
reliability improvement. International Journal of Electrical Power
& Energy Systems. 2011;33:288-95.
[17] Dent CJ, Ochoa LF, Harrison GP, Bialek JW. Efficient secure AC
OPF for network generation capacity assessment. Power Systems,
IEEE Transactions on. 2010;25:575-83.
[18] Ochoa LF, Harrison GP. Minimizing energy losses: Optimal
accommodation and smart operation of renewable distributed
generation. Power Systems, IEEE Transactions on. 2011;26:198-
205.
[19] Rueda-Medina AC, Franco JF, Rider MJ, Padilha-Feltrin A,
Romero R. A mixed-integer linear programming approach for
optimal type, size and allocation of distributed generation in radial
distribution systems. Electric power systems research.
2013;97:133-43.
[20] AlHajri MF, AlRashidi MR, El-Hawary ME. Improved Sequential
Quadratic Programming approach for optimal Distribution
Generation sizing in distribution networks. Electrical and
Computer Engineering (CCECE), 23rd Canadian Conference on:
IEEE; 2010. p. 1-7.
[21] Abou El-Ela A, Allam S, Shatla M. Maximal optimal benefits of
distributed generation using genetic algorithms. Electric power
systems research. 2010;80:869-77.
[22] Celli G, Ghiani E, Mocci S, Pilo F. A multiobjective evolutionary
algorithm for the sizing and siting of distributed generation. Power
Systems, IEEE Transactions on. 2005;20:750-7.
[23] Rahman T, Rahim S, Musirin I. Optimal allocation and sizing of
embedded generators. Power and Energy Conference, PECon
Proceedings National: IEEE; 2004. p. 288-94.
[24] Katsigiannis Y, Georgilakis P. Optimal sizing of small isolated
hybrid power systems using tabu search. Journal of Optoelectronics
and Advanced Materials. 2008;10:1241.
[25] Injeti SK, Prema Kumar N. A novel approach to identify optimal
access point and capacity of multiple DGs in a small, medium and
large scale radial distribution systems. International Journal of
Electrical Power & Energy Systems. 2013;45:142-51.
[26] Lalitha MP, Reddy VV, Usha V. Optimal DG placement for
minimum real power loss in radial distribution systems using PSO.
View publication stats
314
Journal of Theoretical and Applied Information Technology.
2010;13:107-16.
[27] El-Zonkoly A. Optimal placement of multi-distributed generation
units including different load models using particle swarm
optimization. Swarm and Evolutionary Computation. 2011;1:50-9.
[28] Kayal P, Chanda C. Placement of wind and solar based DGs in
distribution system for power loss minimization and voltage
stability improvement. International Journal of Electrical Power &
Energy Systems. 2013;53:795-809.
[29] Falaghi H, Haghifam M-R. ACO based algorithm for distributed
generation sources allocation and sizing in distribution systems.
Power Tech, 2007 IEEE Lausanne: IEEE; 2007. p. 555-60.
[30] Moravej Z, Akhlaghi A. A novel approach based on cuckoo search
for DG allocation in distribution network. International Journal of
Electrical Power & Energy Systems. 2013;44:672-9.
[31] Devi S, Geethanjali M. Application of Modified Bacterial Foraging
Optimization algorithm for optimal placement and sizing of
Distributed Generation. Expert Systems with Applications.
2014;41: 2772–81.
[32] Akorede M, Hizam H, Aris I, Ab Kadir M. Effective method for
optimal allocation of distributed generation units in meshed electric
power systems. Generation, Transmission & Distribution, IET.
2011;5:276-87.
[33] Moradi M, Abedini M. A combination of genetic algorithm and
particle swarm optimization for optimal DG location and sizing in
distribution systems. International Journal of Electrical Power &
Energy Systems. 2012;34:66-74.
[34] Naderi E, Seifi H, Sepasian MS. A dynamic approach for
distribution system planning considering distributed generation.
Power Delivery, IEEE Transactions on. 2012;27:1313-22.
[35] Gomez-Gonzalez M, Lopez A, Jurado F. Optimization of
distributed generation systems using a new discrete PSO and OPF.
Electric power systems research. 2012;84:174-80.
[36] Devi AL, Subramanyam B. Optimal DG unit placement for loss
reduction in radial distribution system-a case study. ARPN J Eng
Appl Sci. 2007;2:57-61.
[37] Elgerd OI. Electric energy systems theory: an introduction: Tata
McGraw-Hill Education; 1983.
[38] Milano F. An Open Source Power System Analysis Toolbox.
Power Systems, IEEE Transactions on. 2005;20:1199-206.