t

1s IEEE International Conference on Power Electronics. Intelligent Control and Energy Systems (ICPEICES-2016)

Optimal Placement and Sizing ofDG Comparison

ofDifferent Techniques ofDG Placement

i 2 3
Prabhjot Kaur ; Sandeep Kaur and Rintu Khanna
1 23
, , Electrical Engg. Dept.; P.E.C. University of Technology; Chandigarh; India
I 2 3
E-mail: prabhjotkaur.meele14@pec.edu.in; sandipsaroa@gmail.com; rintukhannal@rediffmail.com

Abstract- The increasing power demand and the High VoItage Electric System; the power generation in the
deficiency in generating capacities have set the way towards range of 50-100MW is termed as distributed generation.. "
Distributed Generation. Distributed Generation (DG) is The Department of Energy (DOE) considers the
assuming widespread popularity owing to its potential to
distributed generation as the generation from the energy
solve numerous issues like the power system de-regulation;
sources typically varying from less than a kilowatt ( kW) to
meeting the rising power demand; improving voltage profile;
tens of megawatts ( MW) in size for DG unit" [l].The
reduction in power losses etc. The Optimum Location of DG
and optimum DG Sizing are the two critical issues in the conventional power system is designed to work on
integration of DG with the electric grid because the improper conventional power flow which is unidirectional in nature;
p lacement and improper sizing of DG in power system can power flow from source to end-user or consumer but with
not only leads to the enhanced total power losses but can also the incorporation of DG technology; the power can be
damage the normal operation of power system. Optimal generated locally and in case of surplus; it can be fed back
p lacement of DG units is a nonlinear optimization problem. to system or electric grid; making it bi-directional. Hence;
This paper proposes a methodology to calculate the optimal
it requires the reconsideration of many aspects of
location and effective optimal size. The influence of variation
distribution network. The inclusion of DG into the
in location of DG with respect to the total power losses and
existing power system can prove both beneficial and
voltage in the system is also carried out. The proposed
methodology is tested on 33 bus radial distribution network. detrimental also. The incorporation of DG improves the
The obtained results are exhibited in graphical manner. The reliability of power supply; it reduces the power losses;
results achieved from the proposed methodology are provides voltage support and improves the system
paralleled with that of the exhaustive load flows. stability. But on the other hand; the inappropriate DG
Keywords-Distributed Generation; DG Placement; integration can result in increased power losses; can affect
Sensitivity analysis; Optimal Sizing the proteetion coordination and cause islanding also. The
main benefits ofDG integration includes increased overall
I. INTRODUCTION
efficiency; better quality of power; reducing the burden on
The demand for power and electricity is increasing transmission and distribution network; reduction in carbon
day by day with the increasing industrialization. The emissions and encouraged of clean energy that are
conventional methods of power generation i.e. thermal; renewable sources of generation.
hydrogenation are not able to meet the rising demand. There are numerous DG technologies; some are in use
There is disparity between the demand and supply which from long time and some are just emerging. These are
has set the way towards a new technology called basically c1assified into two categories; the conventional
Distributed Generation. DG also named as dispersed or technologies and the non-conventional technologies. The
localized generation is usually defined as the small scale conventional technology includes the use of micro­
power generation at the consumer end. It is the onsite turbine; the conventional generator etc. The non­
generation of electricity i.e. the small generating units conventional includes the renewable sources of generation
which are installed near to load end. It avoids the need of like the photovoltaic cells; wind energy; fuel technology
network expansion in order to supply new areas and new etc. [3]. The selection for these depends on the sole
load. It can be used for commercial; industrial and criteria of availability at the load end. Extensive research
residential applications. There are various definitions of is required to integrate the DG with the power system.
distributed generation. As per IEEE; distributed generation Based on the output characteristics of the DG; the DG
is the power generation by utilities which are very small resources can be modelIed as the voltage prototype;
than the conventional generating units. varying reactive power model and invariable power factor
The International Energy agency describe the model.The effect of load models on the planning of
dispersed generation as the onsite generation of electricity distributed generation (DG) is studied in [8]. The inclusion
which is linked to electric grid at the distribution voltage of DG is full of reconsiderations for the present power
level. As per 'CIGRE'; i.e. International Council on Large system as it alters the line flows; changes the energy
Electric Systems;Distributed generation is also known as losses. Before incorporating DG into the grid system; the

978-1-4673 -8587-9/16/$31.00 ©2016 IEEE [1]

 t
1s IEEE International Conference on Power Electronics. Intelligent Control and Energy Systems (ICPEICES-2016)

optimal placement and optimal sizing are the two critical Sizing; called the 'heuristic approach' [7].Hosein A. Hejazi
issues which need to be addressed. The inappropriate size in [9] has presented a Differential Evolution Algorithm to
and location can hamper the normal operation of system. find the optimal location of DG in distribution network
taking into account the various economical and technical
11. OPTIMAL LOCATION ANDDG SIZING
aspects. In [10]; Ahmad M. Tahboub et al. has proposed
The widespread use of DG is the outcome of the high a new formulation for distribution system reconfiguration
efficiencies in its technologies like co-generation; tri­ (DSR) to reduce the yearly energy losses taking into
generation; the tlexibility in choice of these technologies account the considering the variation in real and reactive
and the decreasing burden on transmission and distribution power demand and (DG) profiles employing the fuzzy C
network. The integration of DG affects the short circuit c1ustering algorithm. S.A. Hosseini in [11] has proposed a
current; voltage output; line losses; system stability and method using genetic algorithm discussing the transient
reliability. So; these parameters are required to be short circuit fault currents owing to incorporation of DG
investigated before placing DG in the system. The studies and thus precluding the unwanted effects on the
conducted so far have proved that the non-optimal coordination of protection devices.
placement of DG can overshoot the line losses which can The main objective in this referenced paper is the
exceed the line losses without DG. If DG sizing is minimization of power losses; to obtain the most optimal
enhanced; it can reduce the line losses but further rise in size-two approaches are provided. One is based on the
size can result in an increase in losses. Moreover; if the sensitivity of the real power losses and the other is based
DG is placed at a particular bus at which voItage is lowest; on the voItage improvement.
it can improve the voltage but if it is placed at a node
III. FORMULATING PROBLEM & METHODOLOGY
where the voltage is already high; it can cause violation of
permissible voltage limits. The appropriate and optimal The objective is to optimize the distribution voItage
placement of DG can be carried out using Analytical with the optimal Distributed Generation. So; initially the
methods; intelligent methods and numerical methods. The basic load tlow of 33 bus system is carried out with a
numerical methods include the Quadratic programming; radial distribution network. A basic load tlow program is
Linear and non-linear programming; MNLP ( 'mixed run to calculate the line tlows and bus voItages and bus
integer non-linear programming'). There are other currents. Appropriate method is used to carry out load
methods which use intelligence based search methods tlow analysis. There are various criteria's for the
including Ant Colony Method; Particle Swarm placement of DG such as the bus with minimum voltage
Optimization; Genetic Algorithm and Artificial Bee or the point where losses are maximum and either can be
Colony Method. selected as the point of placement of DG. After DG
In [2] the technical benefits of DG are presented in incorporation; the optimal size of the DG will be
the form of indices like VoItage improvement profile; calculated as per the load profile.
line-Iosses reduction index; environmental index etc. and The various steps involved are listed as folIows:
the results are obtained for 12 bus system. In [4] various 1. Basic Load tlow analysis using appropriate
sensitivity methods like combined loss sensitivity; voItage method
sensitivity index; index vector method based on both 2. Calculation of bus voItages; bus currents and
novel and modified novel approach are employed to find power tlows and losses
out the optimal location of DG.V.V.S.N Murthy [5] has 3. Selection of criteria for DG placement
used the methods of voltage sensitivity index; combined 4. Optimal Location ofDG
loss sensitivity index; index vector method to find out the 5. Optimal sizing ofDG as per the load profile
optimal location and sizing and has provided a Step 1: Basic Load Flow Analysis: For load tlow
comparative approach and presented the results in analysis; the conventional methods used are
graphical manner. To find the optimal location for placing Gauss seidal and newton raphson methods; but
DG; the '2/3' rule can also be used. Analytical expressions these become inefficient while dealing with
are derived by Wang and Nehrir [6] to obtain the optimal distribution network; wh ich has high R/X ratio.
DG Location in radial as weil as meshed distribution So; the modified gauss seidal or modified
systems to reduce the power losses. But they have used the newton raphson methods can be used to obtain
constant size DG. The evolutionary programming the accurate results. Moreover; there is another
optimization techniques can also be developed to find the method based on 'Kirchhoff's current law and
optimal DG sizing. Kirchhoff's voltage law' called the forward
There is another approach that can be employed to backward sweep method. It proceeds in two
choose the most optimum location and optimum DG steps. In the forward sweep; the downstream

[2]
 t
1s IEEE International Conference on Power Electronics. Intelligent Control and Energy Systems (ICPEICES-2016)

voItage is updated at each step beginning from IV. TEST SYSTEM AND SIMULATION
the source node. In the backward sweep; the t9 20 21 22
KCL and KVL are used to compute the bus
voItage from the furthest node.
Step 2: The bus voItages; bus currents and the line
tlows are calculated. The power losses are
also calculated which are later used for
locating optimal location and optimal sizing
of DG.
Step 3: Finding Optimal Location of DG: There are 23 24 25
numerous criteria's used for fmding the
optimal location of DG. These are discussed
as follow: Fig. I: 33 Bus Radial Distribution System
Optimal Location based on 'Loss Sensitivity Factor':
A list of candidate buses is find out using the formula for The test system used is IEEE 33 bus radial

sensitivity analysis; Real power loss sensitivity for distribution system; which is shown below:

locating DG. The principle involved in this method is

V. SIMULATION RESULTS
linearizing the non-linear equation near the initial
operating point which helps in reducing the workspace. In Bus Voltage witbout DG
this case; frrstly; the sensitivity factors are calculated 102

corresponding to each bus and then in the descending

0.98
order of sensitivity factors; the buses are arranged. The
095
buses with the higher sensitivity factors constitute the list '"
e 094
for candidate buses for DG placement. Now these are

11111111111 1IIIII
0
0_92
studied at alternate positions and the optimal location is
�
'0 0.9
>
obtained. ;; 088
(!l
The total active power losses in a system are given by 085

the 'Exact loss Formula'; which is as follow [1]: 084

N N 1 2 34567 8 9lO1112 131415161718192021222324252627282930313233

BusNumber

PL L L )ai/Pil}
= + QiQj) + ßij(Qil} + PiQj)] (1)
i=1 }=1 Fig. 2: Bus Voltage before DG Placement
where,
• Active Power Losses 0.2110 MW
a·· =...:
I)
ViVj
:!L..cos(o,-o,)
I ) (2) • Total Active Generation 3.93 MW

ß·· =...::!L..sin(o.-o.)
I)
ViVj 1 ) (3)
• Total Reactive Generation 2.44 MVAR
The total real Power Losses without placement of DG
The senstivity factor for the active power loss wrt the are 0.2110 MW. For 33 bus radial distribution test system;
active power injection is given by:
a set of bus es is picked as the contender buses for
a.= aP 2LN a·· ·-ß·· ·
1 aPL=
i 1=1( P Q) I)) I) ) (4) Placement of DG. Based on the loss sensitivity analysis;

A set of candidate buses is obtained by using the topmost bus is bus 8; but based on power loss

sensitivity factor and DG is placed alternately at different reduction; the most appropriate location for placement of
candidate buses to select the most appropriate bus for DG is bus 6; followed by bus 7; with total active power
locating DG. The DG is then located at the most losses as of 0.0679 MW and 0.0696MW; respectively.
appropriate bus. Thus; bus 6 is selected for DG Placement. The active
Optimal Location based on Voltage profile power loss reduction is 68%.
improvement: In this criterion; the basic load tlow is When DG is placed at bus 6
carried out and the voItage at each bus is obtained. The • Active Power Losses 0.0679 MW
buses are arran ged in ascending order of their voltage • Total Active Generation 3.78 MW
magnitudes. The bus with the least voltage magnitude is • Total Reactive Generation 2.35 MVAR
chosen as the appropriate bus for placing DG.For 33 bus The bus voItages of the 33 bus system are calculated
th
radial distribution system,bus 18 is the optimal location and the bus with the minimum voltage is considered for
based 0 voItage profile improvement. the placement of DG. The bus 18 has the minimum

[3]
 t
1s IEEE International Conference on Power Electronics. Intelligent Control and Energy Systems (ICPEICES-2016)

voltage of 0.938p.u; hence the DG is placed at the bus TABLE I: SIZING AND Loss REDUCTION

18,if considering the improvement in bus voltage as the Location of Optimum Voltage Losses(MW) Loss
main purpose. DG DG Sizing Increment Reduction
BusYoltagewithDG at Bus 18 (MW) (V)
L02 Without DG - 0.2110 -
Bus 6 2.56 5% 0.0679 68%
Bus 7 2.45 6% 0.0696 67%
Bus 8 1.8 8% 0.0828 61%
098
Bus 17 0.87 11% 0.1199 43%
Bus 18 0.86 11% 0.1236 41%

�� .. . .. .. .. . .. .. .. . .. .. .. .. ... .. .

12345678 9101112131415161718192021222324252627282930313233
. 1 111 111 VI. CONCLUSION

In this paper; different approaches and criteria's for

finding optimal DG location are discussed and calculated
for IEEE 33 bus system. It provides comparison of two
methods and finding most appropriate location for DG
Fig. 3: Bus Voltages with DG Placed at us 18 placement. The optimal size is calculated at all the buses.
The results have been carried out for finding out the
After placement of DG at bus 18; the voltage is
reduction in active power losses; the increment in voltage
improved. The voltage at bus 17 becomes 1 per unit from
of buses. The findings are presented in graphical form and
0.904 and at bus 18 it becomes l.005 per unit from 0.9038
analyzed from different point of view. The results are
per unit and the total active power losses are reduced to
compared and discussed in elaborate manner
0.1236 MW with the placement ofDG at bus 18.
• Active Power Losses 0.1236 MW REFERENCES
• Total Active Generation 3.84 MW
[I] Naresh Acharya; Pukar Mahat; N. Mithulananthan "An analytical
Total Reactive Generation 2.39 MVAR
approach for DG allocation in primary distribution network"
The optimal sizing is obtained for 33 bus system Electrical Power and Energy Systems 28; 20066; p.p. 69-678
using knitro solver shown in Fig.4. [2] P. Chiradeja and R. Ramakumar; "An approach to quantifY the
Optimum DG Sizing (MW)
technical benefits of distributed generation," IEEE Trans. Energy
Convers.,vol. 19; no. 4; pp. 764-773; Dec. 200.4
3.5 [3] Praveen Kumar Sharma; Nitin Singh "Optimal Location of DG in
Radial Distribution Network" International Journal of Research
Review in Engineering Science & Technology (issn 2278-6643).
2.5
[4] V.V.S.N Murthy; Ashwani Kumar " Comparison of optimal DG
allocation methods in radial distribution system based on sensitivity
L5 analysis" Electrical Power and Energy Systems 53; 2013.
p.p. 450-467
[5] V.V.S.N Murthy; Ashwani Kumar " Optimal Placement of DG in
0.5
radial distribution systems based on new voltage stability index
under load growth" Electrical Power and Energy Systems 69; 2015;
p.p. 246-256
[6] Wang C; et al. "Analytical approaches for optimal placement of
distributed generation sources in power systems"; IEEE Trans
Fig. 4: Optimal DG Size Power Syst 2004; 19(4):2068-76
[7] Sandeep K. et al. "A MINLP technique for optimal placement of
02 5
Losses(MW) multiple DG units in distribution systems",Electrical Power and
Energy Systems 63,2014; p.p. 609-617
[8] Devender Singh et al. "Effect of Load Models in Distributed
02
Generation Planning" IEEE Transactions on Power Systems; Vol.
22; no. 4; Nov. 2007.
01 5 [9] Hosein A. Hejazi et al. "Distributed Generation Site and Size
A1location Through a Techno Economical Multi-objective
Differential Evolution A1gorithm" 20I0 IEEE International
01
Conference on Power and Energy; Nov 29-Decl; Kuala Lumpur;
Malaysia
005 [10] Ahmad M. Tahboub et al. "Distribution System Reconfiguration
for Annual Energy Loss Reduction Considering Variable
Distributed Generation Profiles" IEEE Transactions on power
delivery; Vol. 30; NO. 4; Aug 2015.
Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus Bus
2 4 6 8 IO 12 14 16 18 20 22 24 26 28 30 32
[lI] S.A. Hosseini; et al. "Optimal Placement and Sizing of Distributed
Generation Sources Considering Network Parameters and
Protection Issues" 3rd International Conference on Renewable
Energy Research and Applications Milwakuee; USA 19-22 Oct
Fig. 5: Active Power Losses for 33 Bus System 2014.

[4]