MATEC Web of Conferences 38 , 0 1 0 0 7 (2016 )

DOI: 10.1051/ m atec conf/ 2016 3 8 0 1 0 0 7


C Owned by the authors, published by EDP Sciences, 2016

Optimal Configuration of DG in Distribution System: An Overview

Kumar Mahesh1, Perumal Nallagownden1a, Irraivan Elamvazuthi1

1
Electrical and Electronics Engineering, Universiti Teknologi PETRONAS, 32610 Bandar Seri Iskandar, Perak Darul Ridzuan, Malaysia.

Abstract. Distributed generation (DG) has increased ever attention in the distribution system from last few years. The
main reason for DG in distribution system is increasing electric demand, deregulated power system and congested
transmission network, which eventually declines the system performance. There is also increasing pressure of
greenhouse gas emissions. For proper utilization of DG, the optimal placement and sizing is of main concern. Because
improper DG location and size will increase the losses and decrease the system performance than existing. On the
contrary, proper placement will maintain voltage profile, reduce power loss, and increase voltage stability in the
distribution system. This paper presents overview of DG, the advances in DG technology and different optimization
methods used for optimal placement and sizing problem. The key issues and challenges offered in the development of
DG is also presented in this paper.

1 Introduction feeders that fed from substation and connected to the

customer meter side. Hence incorporating distributed
Power utilities are facing major challenges in constantly generation in it has numerous benefits than other methods.
increasing power demands, fossil fuel depletion and Table 1.describes the comparison among these methods
environmental concern. The electrical energy is usually with respect to loss minimization, cost saving, voltage
supplied through transmission and distribution network, support, demand side management, protection system,
which are operating on maximum allowable limits. Further green power, load balancing and reliability of system.[5].
increase in load causes to reduce the voltage profile and
increase the power losses. The main cause for voltage drop Table 1. Loss minimization Methods in distribution system.
and power loss reduction are high resistance in distribution Methods a b c d e F g h
lines and insufficient reactive power requirement [1, 2]. If
at certain loading this voltage drop not maintain, this will DG allocation √ √ √ √ √ √ √ -
cause voltage instability and blackout [3]. From last few
Network
years, number of blackout has been observed globally.
restructuring √ - √ √ √ - √ √
Thus to improve voltage profile, power loss, enhancing
voltage stability and power quality is highly required. To Capacitor √ √ √ - - - - √
improve aforementioned parameter power system needs to Placement
build large centralize power plants to accommodate this
increasing demand. Transporting the power on aged and a- loss minimization b- cost saving c- voltage support d-
congested transmission network will threatens the security DSM e-Protection system f- green power g-load balancing
and stability of power system. On the other hand h- reliability
generating large power station or upgrading transmission
network either is not suitable due to economic and
environmental concern. The efficient and reliable power 2 Distributed Generation (DG)
delivery has forced power utilities to work on distribution
system in order to reduce the system losses[4]. Many “According to IEEE the DG is the source for generation of
arrangements can be carried out to improve system voltage electricity by facility that are sufficiently smaller than
profile and reduce the power losses, like placement of central power station and connected at nearly any point in
capacitor banks, network reinforcement and optimally a power system”[6]. The distributed generation is the
distributed generator placement etc. Mainly, all these source of electric power connected directly to distribution
techniques reduces the electric current at transmission network or on the customer site of the meter [2].
system by providing it locally at distribution system.
Distribution system has usually included radial
The distributed generation technology generally classified
a
Corresponding author: perumal@petronas.com.my into two categories 1) Renewable DG i.e. wind turbine,

This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits XQUHVWULFWHGXVH
distribution, and reproduction in any medium, provided the original work is properly cited.
Article available at http://www.matec-conferences.org or http://dx.doi.org/10.1051/matecconf/20163801007
 
    

photovoltaic cell, biomass, geo-thermal, small hydro 3.1 Analytical Method

power etc. 2) Non-renewable distributed generation i.e.
diesel engine, micro turbine, gas turbine and combine heat Optimal placement and sizing for DG using 2/3 analytical
and power (CHP) etc.[7, 8]. method has been proposed by [15]. According to this
method, author drives mathematical expressions for size
Previously DG was assumed as active source of power but and location of DG. Author suggests if the load is balance
with recent technology the DGs are available in several then at 2/3 distance from the feeder, a 2/3 size of DG will
types such as 1) injecting active power i.e. Photovoltaic, gives minimum power loss and improves the system
micro turbine, fuel cell 2) injecting only reactive power i.e. voltage performance. The analytical method for optimal
capacitors, DSTATCOM 3) injecting active power and placement of DG was used by [16] for different types of
consuming reactive power i.e. induction generator, 4) load distribution with unity power factor. The main
injecting active and reactive power i.e. synchronous objective was to reduce the power losses of the system. An
generator [9]. The recent technology in distributed exact loss formula for optimal placement and sizing of DG
generation is not only helping in technical and economic has been proposed by [17, 18]. The author's man objective
mode but also nonpolluting and sustainable. The main was to minimize the system real power loss. Author [13]
advantages form DGs are it can be helpful in reducing peak Proposed a novel Power stability index (PSI) for DG
demand, enhancing voltage stability, improving load placement and sizing. The objective functions was to
factor and deferral the transmission cost [10]. The minimize the active and reactive power losses of radial
technical issue related to DG connection in distribution distribution system. Furthermore, the candidate bus was
system depends on its rating. Therefore Table 2. describes found by load flow analysis on standard IEEE 12 and 69
the different categories of distributed generation used in buses. Finally author suggests that active and reactive
distribution system [2, 11]. power demand locally could be significantly reduced by
minimizing power losses and improving the voltage drops
in the lines. . The analytical method for optimal placement
Table 2. Different categories of distributed generation. and size of dispatch-able and non-dispatch (biomass &
Categories DG ratings wind) DGs with power factor consideration has been
studied by [19].
Micro DG ~1 W < 5 kW
Small DG 5 kW < 5 MW
Medium DG 5 MW < 50 MW
3.2 Numerical Method
Large DG 50 MW < 300 MW
The different techniques such as gradient search,
sequential Quadratic, linear programming, nonlinear
Traditional distribution was not designed to accommodate
programming, dynamic programming, ordinal
generation facility [12], but with recent research it is
optimization and exhaustive search methods have been
possible to get maximum utilizations of distribution
used for optimal DG placement and sizing problem.
system to provide customer load locally through DG.
considering all, the nonlinear programming, sequential
Today, number of power companies are investing in small
quadratic and ordinal optimization methods are most
scale distributed generation such as in wind turbines, solar
efficient [20]. DG placement problem considering voltage
photo voltaic cells, micro turbines, small scale hydro and/
stability analysis, author first used model analysis and
or CHP. Distributed generation was observed in England
continuous power flow method to find candidate bus for
and Wales are 1.2 GW during 1993-1994, according to
DG location, then an algorithm was carried on IEEE 33
[13] it reaches to 12GW recently. DG in Fenosa
bus radial system by [3]. A fixed DG size of 40% were
distribution system Spain has 2203MW connected which
placed on weakest bus and achieve satisfactory results. A
represents 120% of area’s total peak demand. Germany
new approach for simultaneous placement of different
installed renewable DG in south region is of 6128 MW,
types of DGs for maximization of system loadability, and
Italy represents highest worldwide grid connected DG of
corresponding minimizing power losses using HPSO (K
10 GW PV and northwest Ireland has 307 Mw connected
matrix & PSO) has been presented by [9]. The author also
as wind DG in distribution system[14].
compared the results with M.Etthadi’s method and found
improved results. Voltage profile and voltage stability
3 Literature Review margin (VSM) enhancement method in a distribution
system with optimal DG placement and sizing have been
The DG in distribution system has been used mainly to presented by [21]. Author consider the load as IEEE RTS
minimize power losses, improve voltage profile, and to bus and DG (wind and solar) as variable PQ buses
enhance voltage stability and reliability of system. Number Furthermore, author used candidate bus by sensitivity
of methods like analytical method, numerical method and analysis then with mixed integer nonlinear programming
artificial intelligence based method has been adopted to the voltage stability margin has been found. The
optimally configure the DG size and location which constraints were voltage limit, feeder capacity and DG
improves the system performance. penetration as constraints.

01007-p.2
 UTP-UMP SES 2015

Table 3. Different types of optimization techniques used for optimal DG placement and sizing in distribution system.
Technique Ref no. Advantages Disadvantages
x Simple and easy to use
x Non iterative in nature x Not suitable for variable load
2/3 Rule [15]
x No convergence issue x Approximate value

x Easy to implement x Lack of robustness

x non iterative in nature x Only consider single objective and
Analytical Method [16-18, 22]
x no convergence issue single distributed generation can be
x Computational time efficiency deployed at a time
x Easy to find literature example
Optimal power flow [23] x Hard to understand and implement
x Computational time efficiency
Mix integer nonlinear x High precision factor
[24] x Hard to understand and implement
programming x Computational efficiency
x This technique optimizes the DG
x The heuristic nature of this algorithm
placement at peak point. The
GA [25] results in a very good convergence
performance of the DG degrades as the
rate.
set point changes.
x Efficient performance for finding the x Harder to simulate, premature
Evolutionary algorithm [26] global minimum convergence
x Easy to find example from literature x Low precision factor
x The technique optimizes the placement
x This technique makes it yield good
and sizing of DG around a specific
results for number of DG and responds
PSO [27, 28] point.
well for variation.
x Relatively lower performance for
x Easy to code with few example
finding the global optimum.
x Non sorting genetic algorithm
other [29, 30] x Subject to successful convergence
x Mimetic algorithm

corresponding minimizing power losses using HPSO (K losses and network voltage variations, the equality and
matrix & PSO) has been presented by [9]. The author also inequality parameters are solved with Pareto
compared the results with M.Etthadi’s method and found
improved results. Voltage profile and voltage stability Frontier Differential Evaluation algorithm by [30]. Author
margin (VSM) enhancement method in a distribution also compares the result with GA/ PSO and concludes that
system with optimal DG placement and sizing have been PFDE technique has higher potential in discovering
presented by [21]. Author consider the load as IEEE RTS solution than GA/PSO. Optimal sitting and sizing problem
bus and DG (wind and solar) as variable PQ buses of DG considering uncertainty in DG and in load by [8].
Furthermore, author used candidate bus by sensitivity The objective function was to minimize the investment
analysis then with mixed integer nonlinear programming cost, operating cost, maintenance cost, capacity adequacy
the voltage stability margin has been found. The constraints cost and network loss cost. The author used new GA-PEM
were voltage limit, feeder capacity and DG penetration as method for probabilistic power flow and for optimal plan
constraints. of DG. The results are compared with GA-MCS. It is
conclude that GA-PEM method is seven time faster GA-
MCS. Simultaneous placement of DG and capacitor in
3.3 Artificial Intelligence
distribution network considering voltage stability index as
Artificial intelligence is the intelligent computer an objective function proposed by [29]. PV curve was used
programing used in vast areas of science and engineering. to find VSI then with help of Mematic algorithm optimal
The main purpose in DG placement and sizing is to DG placement has been found. Optimal placement and
optimize the objective functions technically and sizing determination of DG and DSTATCOM for
economically. Genetic algorithm (GA), Evolutionary minimization of power loss, voltage dip in radial
algorithm (EV), Differential evolution algorithm (DE) and distribution system by PSO based algorithm have been
Particle swarm optimization algorithm (PSO) etc. are used investigated by [32]. In this methodology it takes real
for optimal DG placement and sizing problem. A new power from DG and reactive power compensation by
multi-objective PSO based wind turbine and solar array DSTATCOM using loss sensitivity factor to find possible
generating units in radial distribution system for node for placement of DG and DSTATCOM.
minimization of power losses and enhancement in voltages
stability margin by [31]. According to this author, the
4 Optimization Techniques
optimal placement depends on objectives of distribution
system’s design. A new solution techniques for distribution Several optimization techniques has been used for optimal
expansion planning considering voltage stability, power DG placement and sizing problem. Such as conventional,
artificial intelligence and hybrid intelligence. The

01007-p.3
 
    

conventional technique used in where there is single power distribution networks," in Power Engineering
objective function and it is non iterative method. Whereas Society General Meeting, 2005. IEEE, pp. 557-562 (
artificial and hybrid intelligence method utilized the 2005).
iterative process to search the location and sizes of DGs and 2. T. Ackermann, G. Andersson, and L. Söder,
can be used for one or more than one objective functions. "Distributed generation: a definition," Electric power
The different techniques used for optimal sizing and systems research, vol. 57, pp. 195-204, (2001).
placement with advantage and disadvantages are presented 3. M. Ettehadi, H. Ghasemi, and S. Vaez-Zadeh,
in Table 3. "Voltage stability-based DG placement in distribution
networks," Power Delivery, IEEE Transactions on,
vol. 28, pp. 171-178, (2013).
5 Challenges and Research Issue 4. A. L'Abbate, G. Fulli, F. Starr, and S. D. Peteves,
"Distributed Power Generation in Europe: technical
Basically distribution system is made for unidirectional issues for further integration," Joint Research Center
power flow, incorporating DG, it become bidirectional Institute for Energy. WWW. CaRBoNWaRRooM.
power flow, so it bounces many complexities in the system. CoM2007,(2007).
Actually there are two main technical and economic 5. S. Kalambe and G. Agnihotri, "Loss minimization
constraints which has to be satisfy for proper utilization of techniques used in distribution network:
distribution system performances. The technical bibliographical survey," Renewable and Sustainable
constraints are power losses, voltage profile, voltage Energy Reviews, vol. 29, pp. 184-200, (2014).
stability, carbon emissions and reliability etc. Economic 6. J. D. G. Pepermans., D. Haeseldonckx, D’haeseleer
constraints are DGs cost, DGs installation cost and system W., Belmans R. , "WORKING PAPER SERIES
maintenance cost etc. Hence in the literature many n°2003-8 (K.U.Leuven - Energy Institute) "(2003).
techniques has been used with different objective functions 7. N. Roy and H. Pota, "Current Status and Issues of
to optimally place and size the DG. Concern for the Integration of Distributed Generation
The thirst to use of green energy is increased all around the Into Electricity Networks," (2014).
globe in order to decrease the oil dependency and Carbon 8. V. A. Evangelopoulos and P. S. Georgilakis, "Optimal
foot prints. Many countries are taking efforts to produce distributed generation placement under uncertainties
electricity by distributed generation in distribution system based on point estimate method embedded genetic
due to its numerous advantages. The main drawback of algorithm," IET Generation, Transmission &
renewable DG is uncertainty in its generation due to its Distribution, vol. 8, pp. 389-400, (2014).
prime sources, which ultimately affects the power quality. 9. M. Aman, G. Jasmon, A. Bakar, and H. Mokhlis, "A
This is another complex problem to deal with modern new approach for optimum simultaneous multi-DG
power system for finding the optimal placement and sizing distributed generation Units placement and sizing
of distributed generation. based on maximization of system loadability using
HPSO (hybrid particle swarm optimization)
algorithm," Energy, vol. 66, pp. 202-215, (2014).
6 Conclusion 10. W.-S. Tan, M. Y. Hassan, M. S. Majid, and H. Abdul
This paper offers the latest advances in DG technology, Rahman, "Optimal distributed renewable generation
DGs benefit along with research issues. In addition, an planning: A review of different approaches,"
overview of research work has been conducted for DG in Renewable and Sustainable Energy Reviews, vol. 18,
distribution system. From the literature it is observed that pp. 626-645, (2013).
in order to get maximum remuneration from DG, it should 11. R. Viral and D. Khatod, "Optimal planning of
be optimally located with proper sitting and sizing. distributed generation systems in distribution system:
Different researcher has contributed in this problem with A review," Renewable and Sustainable Energy
different objective functions and methods to satisfy the Reviews, vol. 16, pp. 5146-5165, (2012).
system constraints. The advantages and disadvantages of 12. Y. Huang, "Electricity Distribution Network Planning
different method for DG placement and sizing problem is Considering Distributed Generation," PhD, KTH
also discuss and presented. Finally challenges and research School of Electrical Engineering, KTH, SWEDEN,
issues are conferred. (2014).
13. M. Aman, G. Jasmon, H. Mokhlis, and A. Bakar,
"Optimal placement and sizing of a DG based on a new
Acknowledgment power stability index and line losses," International
Journal of Electrical Power & Energy Systems, vol.
The authors are thankful to Universiti Teknologi 43, pp. 1296-1304, (2012).
PETRONAS, Malaysia for financial support and 14. P. Hallberg, "Active Distribution System Management
motivation to conduct research. a key tool for the smooth integration of distributed
generation," Eurelectric TF Active System
Management, (2013).
References 15. H. L. Willis, "Analytical methods and rules of thumb
for modeling DG-distribution interaction," in Power
1. J. Federico, V. Gonzalez, and C. Lyra, "Learning Engineering Society Summer Meeting, 2000. IEEE,
classifiers shape reactive power to decrease losses in pp. 1643-1644 (2000).

01007-p.4
 UTP-UMP SES 2015

16. C. Wang and M. H. Nehrir, "Analytical approaches for 25. D. Singh and K. Verma, "Multiobjective optimization
optimal placement of distributed generation sources in for DG planning with load models," Power Systems,
power systems," Power Systems, IEEE Transactions IEEE Transactions on, vol. 24, pp. 427-436, (2009).
on, vol. 19, pp. 2068-2076, (2004). 26. W. Sheng, K. Liu, Y. Liu, X. Meng, and Y. Li,
17. N. Acharya, P. Mahat, and N. Mithulananthan, "An "Optimal Placement and Sizing of Distributed
analytical approach for DG allocation in primary Generation via an Improved Nondominated Sorting
distribution network," International Journal of Genetic Algorithm II," (2014).
Electrical Power & Energy Systems, vol. 28, pp. 669- 27. A. El-Zonkoly, "Optimal placement of multi-
678, (2006). distributed generation units including different load
18. D. Q. Hung, N. Mithulananthan, and R. Bansal, models using particle swarm optimization," Swarm
"Analytical expressions for DG allocation in primary and Evolutionary Computation, vol. 1, pp. 50-59,
distribution networks," Energy Conversion, IEEE (2011).
Transactions on, vol. 25, pp. 814-820, (2010). 28. M. Nayeripour, E. Mahboubi-Moghaddam, J. Aghaei,
19. D. Q. Hung, N. Mithulananthan, and R. Bansal, and A. Azizi-Vahed, "Multi-objective placement and
"An optimal investment planning framework for sizing of DGs in distribution networks ensuring
multiple distributed generation units in industrial transient stability using hybrid evolutionary
distribution systems," Applied Energy, vol. 124, pp. algorithm," Renewable and Sustainable Energy
62-72, (2014). Reviews, vol. 25, pp. 759-767, (2013).
20. P. S. Georgilakis and N. D. Hatziargyriou, 29. S. M. Sajjadi, M.-R. Haghifam, and J. Salehi,
"Optimal distributed generation placement in power "Simultaneous placement of distributed generation and
distribution networks: Models, methods, and future capacitors in distribution networks considering voltage
research," IEEE Trans. Power Syst, vol. 28, pp. 3420- stability index," International Journal of Electrical
3428, (2013). Power & Energy Systems, vol. 46, pp. 366-375,
21. R. Al Abri, E. F. El-Saadany, and Y. M. Atwa, (2013).
"Optimal placement and sizing method to improve the 30. M. H. Moradi, A. Zeinalzadeh, Y. Mohammadi, and
voltage stability margin in a distribution system using M. Abedini, "An efficient hybrid method for solving
distributed generation," Power Systems, IEEE the optimal sitting and sizing problem of DG and shunt
Transactions on, vol. 28, pp. 326-334, (2013). capacitor banks simultaneously based on imperialist
22. T. Gözel and M. H. Hocaoglu, "An analytical method competitive algorithm and genetic algorithm,"
for the sizing and siting of distributed generators in International Journal of Electrical Power & Energy
radial systems," Electric Power Systems Research, vol. Systems, vol. 54, pp. 101-111, (2014).
79, pp. 912-918, (2009). 31. P. Kayal and C. Chanda, "Placement of wind and solar
23. ]L. F. Ochoa, C. J. Dent, and G. P. Harrison, based DGs in distribution system for power loss
"Distribution network capacity assessment: Variable minimization and voltage stability improvement,"
DG and active networks," Power Systems, IEEE International Journal of Electrical Power & Energy
Transactions on, vol. 25, pp. 87-95, (2010). Systems, vol. 53, pp. 795-809, (2013).
24. A. Kumar and W. Gao, "Optimal distributed 32. S. Devi and M. Geethanjali, "Optimal location and
generation location using mixed integer non-linear sizing determination of Distributed Generation and
programming in hybrid electricity markets," IET DSTATCOM using Particle Swarm Optimization
generation, transmission & distribution, vol. 4, pp. algorithm," International Journal of Electrical Power
281-298, (2010). & Energy Systems, vol. 62, pp. 562-570, (2014).

01007-p.5