Professional Documents
Culture Documents
a r t i c l e
i n f o
Article history:
Received 3 April 2015
Received in revised form 24 December 2015
Accepted 4 February 2016
Keywords:
Wind energy
Sensitivity analysis
Wind generators
Incremental Transmission Losses
a b s t r a c t
The study evaluates the electrical power system behaviour when wind turbines are inserted into the
power grid. The assessment is made using a sensitivity analysis technique applied to the power flow solution. Unlike the typical algorithms, the sensitivity analysis technique does not require an iterative process, resulting in a fast method with great precision. This proposed method make easy to check the
wind turbine behaviour to the changing of wind speed. Initially, the power flow solution is obtained
and identified as the base case. When there are perturbations in the generators, the new solution is
obtained directly by sensitivity analysis technique. The technique was applied in 34-bus, 70-bus and
126-bus test distribution system. The places chosen to connect the wind turbines were determined by
the Incremental Transmission Losses method. The results demonstrate the effectiveness of the methodology. When wind turbines are inserted in the studied systems, active and reactive losses are reduced and
voltage profile is improved.
2016 Elsevier Ltd. All rights reserved.
Introduction
The development of alternative energy sources are growing in
the world. Wind energy conversion takes a key role in environmental issues, being a renewable source of natural energy, clean and
efficient. Wind turbines can be used both in connection with electrical grids as in isolated places [1].
Brazil has one of the largest wind potential on the planet and
one of the most promising markets for wind power generation.
The year 2013 was a historic year for wind energy in Brazil, being
contracted 4.7 gigawatts (GW) of energy (data EPE Energy
Research Company) [2], capable of supplying energy for about
8.5 million households. Currently, there are in operation in the
Northeast 21 wind farms and, in the region Southern 27 wind
farms. The forecast for 2015 is the integration of 106 more parks
and in 2016, more 254 parks. These data refer to the plants of
the type I, which are dispatched and scheduled by the National
Electric System Operator ONS [3]. This share growing of wind
generation in the Brazilian electric power matrix, especially in
the Northeast and South, has led the ONS in recent years the find
structural solutions and the development of models and tools to
organize this generation. Even with a good generation forecast,
the inherent wind intermittency does not guarantee an exact
amount of generated energy.
Corresponding author.
E-mail address: william.moreti@ufabc.edu.br (W.M. da Rosa).
http://dx.doi.org/10.1016/j.ijepes.2016.02.002
0142-0615/ 2016 Elsevier Ltd. All rights reserved.
307
W.M. da Rosa et al. / Electrical Power and Energy Systems 80 (2016) 306311
gx; u 0
5
gx ; u 0
As a modification Du in u causes a change Dx in x , the expansion in Taylors Series of (6) until first order term gives us the
Eq. (7).
Pk V; h V k
X
Q k V; h V k V m Gkm sinhkm Bkm coshkm
"
Sx J
where:
V k voltage magnitude at bus k.
V m voltage magnitude at bus m.
hkm difference in voltage phase between the k and m bus.
Gkm real element of the matrix Y BUS associated with the bus k
and m.
Bkm imaginary element of the matrix Y BUS associated with the
bus k and m.
m 2 k the set of all m bus having connection with the bus k.
The solution of the Power Flow (PF) problem is obtained from
the balance equations of the active and the reactive power given,
respectively, by Eqs. (3) and (4).
calc
DPk Pspe
k P k V; h 0
calc
DQ k Q spe
k Q k V; h 0
3
4
@ DP
@V
@ DQ
@V
m2k
m2k
@ DP
@h
@ DQ
@h
" @ DP
Su
@P
@ DP
@Q
@ DQ
@P
@ DQ
@Q
Sx D x Su D u 0
10
Dx S1
x Su Du
11
1
As Su is the identity matrix and S1
, the expression
x is equal to J
for the correction vector Dx is given by:
Dx J1 Du
12
The Eq. (12) can be written in matrix form as (13), wherein NPQ
is the number of load bus of the distribution system.
3
Dx1
"
7
6 Dx
@ DP
2
7
6
7
6
@@h
..
7
6
DQ
5
4
.
@h
Dx2NPQ 2NPQ
2
3
Du1
7
6 Du
2
7
6
7
6
..
7
6
5
.
2NPQ2NPQ 4
Du2NPQ 2NPQ
2
@ DP
@V
@ DQ
@V
#1
13
308
W.M. da Rosa et al. / Electrical Power and Energy Systems 80 (2016) 306311
In Eq. (13), the distribution system is represented by the substation and load buses. The vector u consists of the independent variables that are the active and reactive power injections into the
buses, and the vector x that represent the controlled variables,
magnitudes and phase of the buses voltage. The matrix system
(13) can be rewritten as (14).
Dh 2
6
6 Dh 3
6
6
..
6
.
6
6
6
6 DhNPQ
6
6 DV
6
2
6
6 DV
3
6
6
6
..
6
.
4
DV NPQ
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
5
2NPQ
DPesp
2
6
6 DPesp
3
6
6
6 ..
6 .
6
6
6 DP esp
NPQ
6
1
J2NPQ2NPQ 6
6 DQ esp
2
6
6
6 DQ esp
3
6
6
6 ..
6 .
4
DQ esp
NPQ
3
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
5
where:
P Lkm active power losses in the line associated with the k and
m buses.
The sensitivity factor of the active power losses in relation to
the injection of active power of a wind generator defines the ITL
method. The incremental change of the total system losses, @PL ,
in relation to the change in net power injected in the bus k, @Pk ,
can be obtained by Eq. (18).
ITLk
14
"
@h
@PL
@V
2NPQ
xnew x Dx
15
2NPQ
1
qcp AR v 3
2
16
17
@Q
@h
@Q
@V
" @PL #
@P k
@PL
@Q k
2NPQ 2NPQ
19
2NPQ
ITLp
ITLq
The analysis of the kinetic energy conversion of wind to electricity was presented by the several authors [12,13]. The power
of the winds energy can be converted into electrical energy by:
@P
@h
@P
@V
where:
@PL
is the index ITLp , represents the incremental losses cost of
@P
18
" @P #
The matrix system of Eq. (14) has at the right side the perturbation vector (injections of active and reactive power) multiplied by
the inverse of the J matrix. The correction vector of state variables
is at the left side of the equality.
Using the matrix system (14), new solutions for the state variables of the problem, xnew , can be obtained when perturbations
are performed in the load bus as follows:
Pw
@P L
@P k
1
2NPQ
" @PL #
J T 2NPQ 2NPQ
@hk
@P L
@V k
20
2NPQ
Table 1
Ranking of best connect point ITLp .
System
Bus
Bus
34
70
126
10
62
13
26
63
120
309
W.M. da Rosa et al. / Electrical Power and Energy Systems 80 (2016) 306311
Table 2
Comparison of losses with insert of wind turbine in the 34 bus system.
Power
MW
MVAr
Active
Reactive
0.225001
0.357293
0.533335
0.759378
1.041671
1.386463
1.800007
2.000000
2.000000
2.000000
0.000000
0.000000
0.175299
0.249595
0.342380
0.455708
0.591633
0.657368
0.657368
0.657368
0.222234300
0.218128891
0.215747624
0.210666042
0.205874449
0.19998365
0.192928974
0.184671158
0.180757261
0.180757261
0.180757261
0.065261310
0.064120038
0.063457403
0.062043307
0.060707752
0.059063546
0.057091011
0.054776929
0.053678079
0.053678079
0.053678079
6
7
8
9
10
11
12
13
14
15
Results
The proposed methodology has been applied in three distribution systems, containing 34, 70 and 126 bus, analyzing the insertion of the two wind turbines in each system. The systems data
used in the simulation can be found respectively [1618]. The
Studies were performed in MATLAB environment with a LG A520
1.01
1.00
0 m/s
10 m/s
0.99
Losses
V (m/s)
the bus with the ITL higher value, will be most appropriate for the
wind farm connection.
After selecting the locations where to install the wind turbines,
the wind power are simulated according to the characteristic wind
speed of that region and the power system voltage are calculated.
The methodology was applied in this study to identify the best
buses for installation of two wind turbines, in order to minimize
losses of active power in the distribution system.
15 m/s
0.98
0.97
0.96
0.95
0.94
0.93
0.92
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
system bus
Fig. 2. Voltage profile to 34-bus system for different wind speeds.
310
W.M. da Rosa et al. / Electrical Power and Energy Systems 80 (2016) 306311
Table 3
Comparison of losses with insert of wind turbine in the 70 bus system.
Power (MW)
Losses
Power (MW)
Losses
V (m/s)
MW
MVar
Active (MW)
Reactive (MVar)
V (m/s)
MW
MVar
Active (MW)
Reactive (MVar)
0.225001
0.357293
0.533335
0.759378
1.041671
1.386463
1.800007
2.000000
2.000000
2.000000
0.000000
0.000000
0.175299
0.249595
0.342380
0.455708
0.591633
0.657368
0.657368
0.657368
0.224704988
0.217544216
0.213420025
0.204233876
0.195917404
0.185832172
0.173967295
0.160393957
0.154087007
0.154087007
0.154087007
0.102047939
0.099009510
0.097258546
0.093357948
0.089823641
0.085534198
0.080482476
0.074695274
0.072002900
0.072002900
0.072002900
0.225001
0.357293
0.533335
0.759378
1.041671
1.386463
1.800007
2.000000
2.000000
2.000000
0.000000
0.000000
0.175299
0.249595
0.342380
0.455708
0.591633
0.657368
0.657368
0.657368
0.193826086
0.189661395
0.187246376
0.181790978
0.176833197
0.170761416
0.163525949
0.155110017
0.151142749
0.151142749
0.151142749
0.218277071
0.213585860
0.210865831
0.204720912
0.199137787
0.192301225
0.184156063
0.174684501
0.170220614
0.170220614
0.170220614
6
7
8
9
10
11
12
13
14
15
a
Table 4
Comparison of losses with insert of wind turbine in the 126 bus system.
6
7
8
9
10
11
12
13
14
15
TM
26. These are the connection points where the wind power
generators presented in Section Determination of the wind turbine
power were introduced.
For each wind scenario, an active power was calculated and
added to the system, keeping the power factor constant. Table 2
shows the power system loss reduction as a function of the wind
speed increase. It can be observed that for high winds, the power
was limited by the maximum wind farm power. The active losses
achieve 20% reduction with fast winds.
Fig. 2 shows the voltage profile for wind speeds of 5, 10 and
15 m/s. After the choice of the insertion point, the proposed
methodology permits to verify the effect of wind intermittency
in the system power quality. In this simulation, the faster the wind
the better the voltage profile.
70-bus distribution system
The buses 62 and 63 cause major losses on the base case of 70bus distribution system. The wind turbines are connected in these
buses and Table 3 shows the wind effect on the total losses. As in
the 34-bus distribution system, the wind increase reduces the system power losses. The active power losses can be reduced by a 30%
factor when the wind is fast enough.
Fig. 3 shows the voltage profile for wind speeds of 5, 10 and
15 m/s. In this system, the wind intermittency is critical on the
buses near the injection point.
The ITL method has determined that the buses that cause
greater losses to 34-bus distribution system are the buses 10 and
1.02
0 m/s
1.00
10 m/s
15 m/s
0.98
0.96
0.94
0.92
0.90
0.88
0.86
9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69
system bus
Fig. 3. Voltage profile to 70-bus system for different wind speeds.
311
W.M. da Rosa et al. / Electrical Power and Energy Systems 80 (2016) 306311
1.06
0 m/s
1.04
10 m/s
15 m/s
1.02
1.00
0.98
0.96
0.94
0.92
system bus
Fig. 4. Voltage profile to 126-bus system for different wind speeds.
be on the buses 13 and 120. Table 4 shows that all the power
injected reduces the losses in the system.
Fig. 4 shows the voltage profile for wind speeds of 5, 10 and
15 m/s. In this case, the wind speed affects almost all the power
system, improving the voltage profile.
Conclusion
The Sensitivity Analysis tool can be used to simulate the insertion of a wind farm without needing to perform each variation of
these wind turbines with a new calculation of power flow. The proposed method can speed up the power quality analyses to validate
the wind speed effect on the voltage profile and on the losses.
Tests conducted in 34-bus, 70-bus and 126-bus systems
showed that the two SA techniques can be used in power distribution systems. It can be seen that the good choice of the wind turbines insertion point reduces both the losses and the voltage
deviation when the wind turbine is generating power. However,
as the wind intermittency affect the voltage profile, a fast method
as the proposed one is necessary to verify the power quality on
line.
This technique may be used in addition to meet the consumption needs, also to improve the quality of energy that will be delivery to the consumer.
References
[1] Blaabjerg F, Ionel DM. Renewable energy devices and systems research.
Electr Pow Compon Syst 2015;43:8378.
[2] Tolmasquim M. Plano Decenal de Expanso de Energia 2023. In: MME MdMeE,
editor <http://www.epe.gov.br/Estudos/Documents/PDE2023.pdf/2014>.
[3] National Electric System Operator (ONS) <http://www.ons.org.br/home/2015>.
[4] DeMeo EA, Grant W, Milligan MR, Schuerger MJ. Wind plant integration [wind
power plants]. Power Energy Mag IEEE 2005;3:3846.
[5] Lund T, Srensen P, Eek J. Reactive power capability of a wind turbine with
doubly fed induction generator. Wind Energy 2007;10:37994.
[6] Engelhardt S, Erlich I, Feltes C, Kretschmann J, Shewarega F. Reactive power
capability of wind turbines based on doubly fed induction generators. IEEE
Trans Energy Convers 2011;26:36472.
[7] Murari ALL, Tabares HG, Vargas GAL, Belati EA, Sousa VA, Salles M, et al. Study
of transmission system with wind power control and optimal reactive power
flow. Przeglad Elektrotechniczny 2014;07.
[8] Belati EA, Nascimento CF, Dietrich AB, de Faria H. Sensitivity analysis applied
to nodal technical losses evaluation in power transmission systems. Int Trans
Electr Energy Syst 2014;24:17885.
[9] Rosa WMD, Belati EA. First-order sensitivity applied in power distribution
system. Przeglad Elektrotechniczny 2014;7.
[10] Tinney WF, Hart CE. Power flow solution by Newtons method. IEEE Trans
Power Apparatus Syst 1967;PAS-86:144960.
[11] Belati EA, da Costa GRM. Transmission loss allocation based on optimal power
flow and sensitivity analysis. Int J Electr Power Energy Syst 2008;30:2915.
[12] Slootweg JG, Polinder H, Kling WL. Dynamic modelling of a wind turbine with
doubly fed induction generator. Power Engineering Society Summer Meeting,
2001. vol. 1. p. 6449.
[13] Heier S. Wind energy conversion systems. Grid Integr Wind Energy. Wiley &
Sons, Ltd; 2014. p. 31117.
[14] Vargas GAL, Tabares HG, Murari ALL, Belati EA, Sguarezi-Filho AJ. Analysis of
transmission systems with optimal power flow and Dobly-fed induction
generator. Latin America Trans IEEE (Revista IEEE America Latina)
2015;13:2195201.
[15] Santos ECD. User of losses sensitivity coefficients (ITL) in determining the best
location for buses of a generation unit distributed in radial distribution
systems [masters degree]. Distrito Federal Brasil: Federal District Brazil;
2009.
[16] Chis M, Salama MMA, Jayaram S. Capacitor placement in distribution systems
using heuristic search strategies. Generation Transm Distrib IEE Proc
1997;144:22530.
[17] Baran ME, Wu FF. Optimal capacitor placement on radial distribution systems.
Power Del IEEE Trans 1989;4:72534.
[18] Yan L, Yulei L, Buhan Z, Chengxiong M. A Modified Newton-Raphson power
flow method considering wind power. Power and Energy Engineering
Conference (APPEEC), 2011 Asia-Pacific 2011. p. 15.