Looping of Radial Distribution Network to Mitigate the Over Voltage Problems and to Increase the Integrated Capacity of Solar PV

Full text

Proc. of the 1
st
International Conference on Electrical, Communication and Computer Engineering (ICECCE)
24-25 July 2019, Swat, Pakistan
978-1-7281-3825-1/19/$31.00 ©2019 IEEE
Looping of Radial Distribution Network to Mitigate
the Over Voltage Problems and to Increase the
Integrated Capacity of Solar PV
Qais Ali*
US-Pakistan Center for Advanced
Studies in Energy
National University of Science and
Technology
Islamabad, Pakistan
ali.qais085@gmail.com
Hafiz Waleed Ahmad
US-Pakistan Center for Advanced
Studies in Energy
National University of Science and
Technology
Islamabad, Pakistan
waleedahmad526@gmail.com
Syed Ali Abbas Kazmi
US-Pakistan Center for Advanced
Studies in Energy
National University of Science and
Technology
Islamabad, Pakistan
saakazmi@uspcase.nust.edu.pk
Abstract— Due to the increase in demand for electrical
energy, the integration of solar PV in the grid network is
increasing at a great rate in both developing and developed
countries. However, majority developing countries have
radial power distribution networks and thus high
penetration of solar PV can cause over voltage problems.
This paper discusses the effects of solar PV integration on
end line bus voltages and the resulting power losses in radial
distribution network. A loop distribution network is
proposed to solve the problems of over voltages and increase
the integration capacity of solar PV. Seven different
scenarios have been analyzed for both radial and loop
distribution network by placing solar PV system at different
buses and of different capacity on an IEEE-33 bus system
and a comparison of the results is given using
MATLAB/Simulink environment. The results show that by
looping the radial distribution network, the problems of over
voltages and power losses are minimized, and the integration
capacity of solar PV can be increased.
Keywords— solar PV integration, radial distribution
network, loop distribution network, over voltages, renewable
energy sources
I. I
NTRODUCTION
The demand for energy is increasing day by day due to
modernization, electrification and developments in
industrial sectors. To overcome the increasing demand of
energy, integration of Renewable Energy Sources (RES)
like solar PV is increasing rapidly. According to the 2018
International Renewable Energy Agency (IREA) report,
the generation capacities of different RES is shown in fig.
1. In 2017, approximately 93GWs of energy is added from
solar PV as depicted in fig. 2, and this capacity is
expected to increase more in future. The over voltage
problem is very common when PV is integrated at the end
bus of the line in Radial Distribution Network (RDN) [1-
3].
The existing power Distribution Networks (DNs) are
designed to work in radial fashion, in which the power
flow is unidirectional from the grid/utilities to the
consumers [4].
But as the tradition of PV system
installation at distribution level is increasing, thus high
penetration of RES in the DNs can have great impact on
the power flow, voltages and stability of the DNs since
when a PV system is integrated in the DN, the power flow
no longer remains unidirectional and the existing DNs
needs to be modified and upgraded to overcome these
problems [4-6].
Fig.1. Renewable generation capacity by Energy source
(source: International Renewable Energy Agency, March 2018)
The DNs reliability, stability, voltage profile, loss
reduction and congestion of cable can be improved by
integrating PV system but it can be done by optimal
placement and sizing of PV and also needs the
modification of existing DNs [7-10]. In this paper the
IEEE-33 bus RDN is used to discuss and analyze the
problems of over voltages. In Section II, this RDN is
modified to a Loop Distribution Network (LDN) to
mitigate the problem of over voltage and to increase the
integration capacity of solar PVs in DNs. The simulation
results of both RDN and LDN have been compared and
discussed in section III and finally section IV concludes
the study. The results show that the modified LDN is
more stable with integration of PV system of different
capacities.
Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on October 21,2020 at 12:51:55 UTC from IEEE Xplore. Restrictions apply.

Fig.2. Capacity growth of RES
(source: International Renewable Energy Agency, March 2018)
II. S
YSTEM
M
ODELLING
For simulation studies the IEEE-33 bus RDN system has
been used. The only modification is made by looping the
distribution network with connection between bus 33 and
bus 18 as shown ion fig. 3
.
Fig.3. Loop distribution network
The complete model design in MATLAB/Simulink is
shown in fig. 4. The solar PV farm is connected to the
distribution network by a three-phase inverter to convert
the DC power generated by solar PV into AC power and
then output voltages from the inverter are step up using
step up transformer.
Fig.4. Complete model of the system in MATLAB/Simulink
A. Equations for total real power losses
The equation for Total Real Power Loss (TPL) is given as
follow:
(1)
where represents the current from bus to bus and
is the resistance between bus and bus .
III. S
IMULATIONS
,
R
ESULTS AND
D
ISCUSSION
Different scenarios with different penetration level and
placement of solar PV has been simulated and discussed
for both RDN and LDN one by one.
A. Case 1: When no PV is integrated in loop network
When there is no PV integrated to the system, the voltage
profile remains the same for both looped and radial
system as shown in fig. 5. Also, there is a minor decrease
in total real power losses (TPL). For RDN the TPL is
106.65 kW, while for LDN the TPL is 105.5 kW. But the
main advantage of looped distribution system is that if a
fault occurs in one feeder or any equipment of the line
will not affect the overall system.
Fig.5. Bus voltages for loop and radial network when no PV is integrated
B. Case 2: When 1500 kW PV is integrated at bus 18
When the RDN is upgraded by making it loop network,
the voltages of the buses have been significantly improved
as shown in fig. 6. For RDN the voltages have exceeded
the upper voltage limit and the bus 18 is overvoltage when
1.5 MW PV is integrated at bus 18. But in the loop
network the voltages of all buses are improved and are
brought back within limits. Also, the TPL has been
reduced considerably, for RDN the TPL is 86.5 kW, by
looping the system the TPL has been reduced to 58.89
kW.
Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on October 21,2020 at 12:51:55 UTC from IEEE Xplore. Restrictions apply.

Fig.6. Bus voltages for loop and radial network with 1.5 MW PV
integrated at bus 18
C. Case 3: PV integration at bus 18 with increased
capacity
In this case the generation from PV system has been
increased to 2.5 MW, with 40% increase in the capacity.
But still the voltages of all buses are within the limits as
shown in fig. 7. This result shows that the looped
networks are more stable and distribution generation with
greater capacity can be integrated in the LDN than that of
RDN.
Fig.7. Bus voltages of loop network with 2.5 MW PV integrated at bus
18
D. Case 4: When 1500 kW PV is integrated at bus 33
In this case we can see from figure 8, that when 1.5 MW
PV is integrated to RDN the end bus voltages of one line
has been improved i.e. bus 33, 32, 31 and 30. But on the
other hand the end bus voltages of the second line are well
below than the lower voltage limits i.e. for bus 9 to bus 18
the voltages are well below the limits. But when the
system is upgraded and modified to LDN, the voltage
profile of the system has significantly improved. Also, the
TPL has been significantly reduced, for RDN the TPL is
76.7 kW which has been decreased to 58.69 kW by
modifying the RDN to LDN.
Fig.8. Bus voltages for loop and radial network when 1.5 MW PV is
integrated at bus 33
E. Case 5: PV integration at bus 33 with increased
capacity
In this case the capacity of integrated PV is increased by
40%, from 1.5 MW to 2.5 MW. By increasing the
capacity in the LDN, the voltage profile of the system is
further improved as shown in fig. 9. Unlike for RDN,
when the capacity of the integrated PV at end buses of the
line has been increased the voltages exceeds the upper
limit. Hence this problem can be mitigated by modifying
the RDN to LDN. The LDN provide more support to RES
integration at distribution level than RDN.
Fig.9. Bus voltages of looped network when 2.5 MW PV is integrated
at bus 33
F. Case 6: When 1.5 MW PV is integrated at bus 3 or bus
6
When the DG/PV is integrated at the head bus near to the
grid i.e. bus 3 or bus 6, the effect on voltage profile is very
minor. For both LDN and RDN the voltage profile is same
as shown in fig. 10 and 11. which means that when the
DG/PV is integrated near to main generation the voltage
profile remains the same, also there is no significant
change in TPL. For RDN the TPL is 90.413 kW and is
89.28 kW for LDN, when 1.5 MW PV is integrated at bus
3. Also, TPL for RDN is 68.244 kW, while TPL is 67 kw
for LDN when 1.5 MW PV is integrated at bus 6.
Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on October 21,2020 at 12:51:55 UTC from IEEE Xplore. Restrictions apply.

Fig.10. Bus voltages for radial and loop network when 1.5 MW PV is
integrated at bus 3
Fig.11. Bus voltages for radial and loop network when 1.5 MW PV is
integrated at bus 6
G. Case 7: When multiple PV systems are integrated at
different buses
In this case two PV system are integrated in the DN. In
first scenario 1 MW PV is integrated at bus 18 and at bus
8. The voltage profile for both LDN and RDN is shown in
fig. 12. The voltage profile has been improved for the end
buses when the system is looped.
Fig.12. Bus voltages for loop and radial network when PV is integrated
at bus 8 and bus 18.
For second scenario 1MW PV is integrated at bus 33 and
at bus 18. The voltage profile for this scenario is almost
the same as shown in fig. 13. Thus, it is concluded that the
voltages of the buses depend on the DG location and
capacity for both LDN and RDN. There is significant
reduction in TPL for both scenarios with LND and RND.
Fig.13. Bus voltages for loop and radial network when PV is integrated
at bus 33 and bus 18
TABLE I. COMPARISON BETWEEN RDN AND LDN IN
TERMS OF TPL
Distributed Generation
TPL for
RDN
(kW)
TPL for
LDN
(kW)
No PV integrated
106.65
105.5
1500 kW PV at bus 18
86.5
63.85
1500 kW PV at bus 33
76.7
58.69
1500 kW PV at bus 6
68.244
67
1500 kW PV at bus 3
90.413
89.28
1000 kW PV at bus 18 and bus 33
72.833
62.3
1000 kW PV at bus 18 and bus 8
76.403
63.8
IV. C
ONCLUSION
From the simulated results shown above we conclude that
LDN is more stable than RDN and can remain stable with
high penetration level of RES. The voltage profile has
been improved significantly when the system is modified
from RDN to LDN. Also, there is enormous reduction of
TPL in LDN as can be seen from table 5.1. Thus, the
integration of RES at distribution level can be made
smooth by updating the current RDN to LDN and the
voltages can remain within limits even for high
penetration level of DG.
R
EFERENCES
[1] H. Shao, “Analysis on Voltage Profile of Distribution Network
with Distributed Generation,” IOP Conf. Ser. Earth Environ.
Sci. 113 012170, ICAESEE, 2018.
[2] M. A. Mahmud, M. J. Hossain, and H. R. Pota, “Analysis of
Voltage Rise Effect on Distribution Network with Distributed
Generation,” 18th IFAC World Congr. Milano (Italy), Int. Fed.
Autom. Control, no. 2002, pp. 14796–14801, 2011.
[3] J. Kennedy, P. Ciufo, and A. Agalgaonkar, “Over-Voltage
Mitigation within Distribution Networks with a High
Renewable Distributed Generation Penetration,”
Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on October 21,2020 at 12:51:55 UTC from IEEE Xplore. Restrictions apply.

ENERGYCON, Dubrovnik, Croat., pp. 1107–1114, 2014.
[4] R. Seguin, J. Woyak, D. Costyk, J. Hambrick, and B. Mather,
“High-Penetration PV Integration Handbook for Distribution
Engineers,” Natl. Renew. Energy Lab., 2017.
[5] M. Karimi, H. Mokhlis, K. Naidu, S. Uddin, and A. H. A.
Bakar, “Photovoltaic penetration issues and impacts in
distribution network – A review,” Renew. Sustain. Energy
Rev., vol. 53, pp. 594–605, 2016.
[6] M. Saradarzadeh, S. Farhangi, and J. L. S. P. J. D. Frey, “The
benefits of looping a radial distribution system with a power
flow controller,” IEEE Int. Conf. Power Energy, no. Kuala
Lumpur, Malaysia, pp. 723–728, 2010.
[7] M. J. E. Alam, “Performance analysis of distribution networks
under high penetration of solar PV,” Conf. Int. des Gd.
Reseaux Electriques (CIGRE, Paris), pp. 1–9, 2012.
[8] H. Sadeghian and Z. Wang, “Photovoltaic Generation in
Distribution Networks : Optimal vs . Random Installation,”
IEEE Power Energy Soc. Innov. Smart Grid Technol. Conf.,
Washington, DC, USA, 2018.
[9] M. Rezaei, “OPTIMIZATION OF SIZING AND
PLACEMENT OF PHOTOVOLTAIC ( PV ) SYSTEM IN
DISTRIBUTION NETWORKS CONSIDERING POWER
VARIATIONS OF PV AND CONSUMERS USING
DYNAMIC PARTICLE SWARM OPTIMIZATION
ALGORITHM ( DPSO ),” Indian J. Fundam. Appl. Life Sci.,
vol. 5, no. 1, pp. 3321–3327, 2015.
[10] R. Prenc, D. Škrlec, and V. Komen, “Optimal PV System
Placement in a Distribution Network on the Basis of Daily
Power Consumption and Production Fluctuation,” EuroCon,
Zagreb. Croat., no. July, pp. 777–783, 2013.
Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on October 21,2020 at 12:51:55 UTC from IEEE Xplore. Restrictions apply.