Journal of Electrical and Electronics Engineering 9

_____________________________________________________________________________________________________________

Optimal Power Flow Analysis of IEEE 14 System

with Distributed Generators
DULĂU Lucian Ioan1
1
“Petru Maior” University of Tîrgu Mureú, Romania
Department of Electrical and Computer Engineering, Faculty of Engineering
N. Iorga St., no. 1, 540088, Tîrgu Mureú, Romania, [email protected]

Abstract –This paper describes the mathematical model

for the optimal power flow analysis of the IEEE 14 bus
system. In the IEEE 14 bus system were connected two
distributed generators (DGs): a hydro generator and a
photovoltaic power plant. The two objective functions
which are minimized, are the generation costs and the
total power losses of the system, while there are
satisfied equality and inequality constraints. The
optimal power flow analysis will be performed for the
steady-state condition of the system and for the whole
day, for two cases: the DGs are connected to the system
and the DGs are not connected to the system.
Fig.1. Central generation (a) and distributed generation
Keywords: optimal power flow; IEEE 14 system; (b).[2]
generation costs; power losses.
The power supplied by the distributed generation
I. INTRODUCTION sources can vary according to the availability of their
primary energy source (e.g. wind speed), so the best
Distributed generation (DG) refers to the production
location for the installation of a distributed energy
of power near or at the consumption place. The
source is where the primary energy source has the best
distributed generation resources are the cogeneration
potential.
(combined heat and power - CHP) units and the
In this paper will be performed an optimization
renewable energy sources (RES).[1-11]
(optimal power flow analysis) for the IEEE 14 system,
The renewable energy sources include wind energy,
in which were connected two distributed generation
solar energy, hydro energy, geothermal energy and the
sources. The optimization will consider the
energy from biomass. These sources don’t produce CO2
minimization of the power losses and of the generation
or greenhouse emissions, so there are environment
costs.
friendly. Also, the power produced by these sources is
relatively small.
II. OPTIMAL POWER FLOW
In Fig. 1 is presented a central generation system and
MATHEMATICAL MODEL
a distributed generation system. In the central generation
system the power produced by large power plants is
The objective of the optimal power flow is to
delivered to the consumers using the transmission and
minimize two functions, the generation cost (G(C)) and
distribution system. In the central generation system the
total power losses of the system (ǻS), while there are
power flow is unidirectional, from the power plants to
satisfied inequality and equality restrictions:
the consumers. In the distributed generation system, at
min {F= G(C)+ǻS} (1)
distribution or consumers level can be connected wind
where:
turbines, photovoltaic panels or CHP units. In this
• G(C) – generation costs;
system, the power flow can be bidirectional, from the
consumer to the distribution level. Due to the connection • ǻS – active and reactive power losses (total
of these distributed generation sources, the power losses power losses of the system).[2,11,12]
are lower and the power quality and reliability of the The generation cost G(C) can be expressed as:
C(P)= a•PGi2+ b•PGi +c (2)
grid are improved.
The distributed generation sources can also be used where:
to supply power to an isolated consumer, which is not • PGi - generating unit output;
connected to the power grid. • a, b, and c- constants.[2,11,12]
 10 Volume 9, Number 1, May 2016
_____________________________________________________________________________________________________________

The total power losses of the system can be The generators data are presented in table 1, the cost
expressed as: coefficients for the generating units are presented in
ǻS= PL+jQL (3) table 2, while the load data are presented in table 3.
where: The data for the generators, loads and the cost
• PL- active power losses of the system; coefficients are presented in [12,13].
• Q L- reactive power losses of the
system.[2,11,12] TABLE 1. Generators data.[12]
The function is subjected to the following
restrictions: Generator PGimin PGimax QGimin QGimax
[MW] [MW] [MVAr] [MVar]
Pi (V,ș)=PGi-PDi
1 0 232.4 0 10
Qi (V, ș)= QGi-QDi
PGmin ” PG ”PGmax 2 0 40 -40 50
QGmin ” QG ”QGmax
Vimin ” Vi ”Vimax 3 0 0 0 40
Plmin ” Pl ”Plmax (4)
where: 6 0 0 -6 24
• PGi- active power supplied by the generator
connected to bus i; 8 0 0 -6 24
• QGi- reactive power supplied by the generator
HG 0 2 0 0
connected to bus i;
• PDi- active power demand of the load connected PPP 0 2 0 0
to bus i;
• QDi- reactive power demand of the load
connected to bus i; TABLE 2. Cost coefficients.[13]
• Pi- total active power injection at bus i;
• Qi- total reactive power injection at bus i; Generator a b c
• Vi- voltage magnitude at bus i; [EUR/MWh2] [EUR/MWh] [EUR/h]
1 0.0016 2.00 150
• Pl- the power flow for the line l from bus j to bus
k.[2,11,12] 2 0.0100 2.50 25
The optimal power flow will be performed with the
Neplan software [14]. 6 0.0083 3.25 0

III. IEEE 14 BUS SYSTEM DATA HG 0 3 0

The optimal power flow analysis will be performed PPP 0 8 0

for the IEEE 14 bus system (Fig. 2), in which were
connected two distributed generators:
• a 2 MW hydro generator (HG) at bus 10; The power output of the distributed generators for
• a 2.3 MW photovoltaic power plant (PPP) at bus the whole day is presented in Fig. 3.
14. 


 

 





 










Fig. 3. Distributed generators power output

Fig. 2. IEEE 14 bus system [12]
 Journal of Electrical and Electronics Engineering 11
_____________________________________________________________________________________________________________

TABLE 3. Loads data.[12] IV. OPTIMAL POWER FLOW RESULTS

Load PC [MW] QC [MVAr] The active power losses and the generation cost for
the steady-state condition of the system, for the case
1 0 0 when the distributed generators are connected to the
system are:
2 21.7 12.7 • active power losses: 12.86 MW;
• total generation cost: 853.39 EUR/h.
3 94.2 19
The active power losses and the generation cost for
4 47.8 0 the steady-state condition of the system, for the case
when the distributed generators are not connected to the
5 7.6 1.6 system are:
6 11.2 7.5 • active power losses: 13.59 MW;
• total generation cost: 842.755 EUR/h.
7 0 0 The active power losses and the generation cost after
8 0 0 the optimization was performed (optimal power flow),
for the case when the distributed generators are
9 29.5 16.6 connected to the system are:
• active power losses: 2.35 MW;
10 9 5.8
• total generation cost: 513.718 EUR/h.
11 3.5 1.8 The active power losses and the generation cost after
the optimization was performed (optimal power flow),
12 6.1 1.6 for the case when the distributed generators are not
connected to the system are:
13 13.5 5.8 • active power losses: 3.12 MW;
• total generation cost: 584.41 EUR/h.
14 14.9 5
The optimal power flow analysis results for the
whole day are presented in Fig. 5 (active power losses)
and in Fig. 6 (total generation cost).
For the IEEE 14 bus system, the installed load power
is 259 MW, corresponding to 1 load factor. 
The load demand for the whole day is presented in
Fig. 4.


! "#
$ 
 
% &&&
'(&
 )) 

 

 
 &*& +,
  

! "#

$ 
 % &&&
'(&)
  )) 
 
 &*& +,










  
 







Fig. 4. Load demand

Fig. 5. Active power losses
 12 Volume 9, Number 1, May 2016
_____________________________________________________________________________________________________________

IEEE 14 system it can be concluded that the active


power losses were lower when the two distributed
generators were connected to the system.
Also, the total generation cost of the IEEE 14 system

 was higher when the two distrbuted generators were not
connected to the system due to the high cost coefficients
1 % of the generator G1.
 2) " )
 & '(& REFERENCES
 )) 
) " ) & -./0

  [1] T. Ackermann, G. Andersson and L. Soder, „Distributed

 &*& +, generation: a definition”, Electric Power Systems
Research, Nr. 57, 2001.
1 % [2] L.I. Dulău, M. Abrudean and D. Bică, “Automation of a
2) " ) distributed generation system”, 49th Universities’ Power
  & '(& Engineering Conference (UPEC 2014), IEEE Xplore, pp.
) 1-5, 2014.
 ))  [3] L.I. Dulău, C. Bondici and D. Bică, “Distributed
  generation and distributed energy resources”, Journal of
 &*& +, Computer Science and Control Systems, Vol. 7, Nr. 1, pp.
29-34, May 2014.
[4] L.I. Dulău, M. Abrudean and D. Bică, “Effects of
Distributed Generation on Electric Power Systems”,

Procedia Technology, Vol. 12, pp. 681–686, 2014.
[5] S. Golestani and M. Tadayon, “Distributed generation
dispatch optimization by artificial neural network trained
 by particle swarm optimization algorithm”, 2011 8th
International Conference on the European Energy Market
  (EEM), pp. 543-548, 2011.







[6] P.P. Padhi, R.K. Pati and A.A. Nimje, Distributed

Fig. 6. Generation cost Generation: Impacts and Cost Analysis, International
Journal of Power System Operation and Energy
V. CONCLUSIONS Management, Vol. 1, No. 3, 2012.
[7] R. Firestone and C. Marnay, “Distributed Generation
Distributed generation (DG) refers to the production Dispatch Optimization under Various Electricity Tariffs”,
of power near or at the consumption place using International Journal of Electronic Business Management,
Vol. 5, No. 3, 2007.
cogeneration (CHP) units and renewable energy sources
[8] N. Jenkins, J.B. Ekanayake and G. Strbac, „Distributed
(RES). Generation”, Institution of Engineering and Technology,
The optimization presented in this paper minimized 2010.
two objective functions, the power losses and the [9] M. Bollen and F. Hassan, “Integration of Distributed
generation costs, while there were stafisfied inequality Generation in the Power System”, 1st Edition, John Wiley
and equality restrictions. & Sons, 2011.
The analyzes were performed for the IEEE 14 bus [10] A.M. Borbely and J.K. Kreider, „Distributed Generation:
system, in which were connected two distributed The Power Paradigm for the New Millennium”, CRC
generators (DGs): a hydro generator and a photovoltaic Press, 2011.
[11] L.I. Dulău, “Optimization of a Power System with
power plant.
Distributed Generation Source”, 9th International
First, the active power losses and the generation cost Symposium on Advanced Topics in Electrical Engineering
were determined for the steady-state condition of the (ATEE 2015), IEEE Xplore, pp. 903-906, 2015.
IEEE 14 system, for the situation when the DGs were [12] J. Zhu, “Optimization of power system operation”, 2nd
connected to the system and for the situation when the Edition, John Wiley & Sons, Inc., 2015.
DGs were not connected to the system. [13] B. Sahu, A. Lall, S. Das and T.J. Patra, „Economic Load
Second, the optimal power flow analyzes were Dispatch in Power System using Genetic Algorithm”,
performed for the steady-state condition of the IEEE 14 International Journal of Computer Applications, Vol. 67,
system for the situations when the DGs were connected No.7, 2013.
[14] www.neplan.ch
to the system and when the DGs were not connected to
the system.
Third, the optimal power flow analyzes were
performed for the whole day for two situations:
• the DGs were connected to the system;
• the DGs were not connected to the system.
From the steady-state analysis, respectively from the
optimal power flow analysis for the whole day of the