Nigerian Journal of Technology (NIJOTECH)

Vol. 34 No. 2, April 2015, pp. 325 – 331

Copyright© Faculty of Engineering,
University of Nigeria, Nsukka, ISSN: 1115-8443
www.nijotech.com
http://dx.doi.org/10.4314/njt.v34i2.17

THE IMPACT OF DISTRIBUTED

DISTRIBUTED GENERATION ON DISTRIBUTION
NETWORKS

A. O. Ekwue1,* and O. A. Akintunde2

1 JACOBS ENGINEERING INC/BRUNEL UNIVERSITY LONDON, UNITED KINGDOM

2 BRUNEL UNIVERSITY LONDON, UNITED KINGDOM

addresses
Email addresses:: 1 [email protected], 2 [email protected]
es

ABSTRACT
Distributed Generators (DG or embedded generators) are generators that are connected to the distribution
network. Their advantages are the ability to reduce or postpone the need for investment in the transmission and
distribution infrastructure when optimally located; the ability to reduce technical losses within the transmission
and distribution networks as well as general improvement in power quality and system reliability. This paper
distributed
highlights the benefits of distr 15--bus radial distribution network, modelled in the
ibuted generation by using a 15
DIgSILENT Power Factory software, to demonstrate the improvement in voltage profile as well as the reduction in
technical losses.

Keywords: distributed generation; wind energy; integration of renewable sources and technical losses.
Keywords

1. INTRODUCTION beyond the capacity of existing switchgear. This will

Distributed Generators (DG or embedded generators) result in the installation of new circuit breakers
are generators that are connected to the distribution which can be expensive. In order to ensure proper
network. Their advantages are the ability to reduce or operation of power system protection schemes, it is
postpone the need for investment in the transmission necessary to provide proper and cost-efficient
and distribution infrastructure when optimally solutions for limiting the fault levels to their
located;the ability to reduce technical losses within equipment capacity.
the transmission and distribution networks as well • Potential reactive power issues. This is because
asgeneral improvement in power quality and system current wind generation technology, for example,
reliability. has different electrical characteristics from
However, the development of renewable energy conventional thermal, nuclear and hydro plants
sources (e.g. wind or solar energy) and subsequent when large numbers of wind generators are
integration with the distribution network will increase installed on the network. The use of induction
the amount of generation so connected and this raises generators in place of the traditional synchronized
some technical issues such as [1]: generators will deliver a significant increase in
• inadvertent islanding where a section of a reactive power and it is important that this is
distribution network is to be split off the properly managed by the network operators. The
transmission network and is still energised by its increased reactive power on the network may
embedded generators. prevent the connection of further wind generators
• the reversal of power flows which means that to the network unless active steps are taken to
existing protection (if it is directional) cannot be absorb the reactive power.
used hence the need to install a new protection • Fault Ride-Through Capabilities - As the penetration
system. of renewable generation such as wind energy
• the connection of renewable generation can raise increases, the need to address the fault ride-through
the fault levels due to the fault contributions from capability issues will become more critical. Hitherto,
the renewable generators themselves to values the wind turbines were allowed to trip when a

*Corresponding author’s Telephone: +44 (0)7535267242.

 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

voltage dip occurs. The wind turbines will now be profile as well as the reduction in technical losses. The
expected to remain connected to the grid both modelling is presented in the next Section whereas the
during and after a fault. Upon voltage recovery, the results are discussed in Section 3.
wind turbines are not expected to consume It is pertinent to mention that losses can be classified
excessive reactive power when re-exciting the as technical or non-technical losses (comprising of
generator, as this may result in a further voltage dip. illegal connections of electricity, metering errors and
This subject has been addressed elsewhere in [1]. deficiencies, billing and processing errors etc); the
Despite these technical challenges, many countries are subject of non-technical losses have been discussed
reducing the greenhouse emissions and increasing the elsewhere in [10] by Nwodo and Ekwue (1992).
proportion of new forms of electricity generation [2];
the United Nations General Assembly has declared 2. TEST SYSTEM MODELLED IN DIGSILENT POWER
2014 – 2024 as the decade for sustainable energy. A FACTORY SOFTWARE
recent report from the International Energy Agency 2.1 DIgSILENT Power Factory software
[3] showed that the use of renewable energy grew DIgSILENT1 is a private company, headquartered in
strongest last year and now produces about 22% of Gomaringen near Stuttgart, Germany and provides
the global consumption of electricity. The UK, for power system analysis software and related
example, imports nearly 50% of the coal it uses and it consultancy services. PowerFactory is its principal
is also looking for new sustainable energy solutions software offering, and is a wide-ranging PC based
for the future. Renewables will help avoid dependence power system analysis package.
on imports and make the country less vulnerable to PowerFactory is suited to industrial, transmission and
security threats. The electricity generation via distribution networks, power plants as well as marine
renewables increased by 27% during June 2011 and and aerospace industries. It was developed in
June 2012 to 38 TWh [4]. By 2020 UK hopes to Germany where it has been used very
provide 15% of its energy consumption from successfully, especially with the modelling of wind
renewable sources. farms (their Doubly-Fed Induction Generator model
Japan is the world’s third largest economy and its was one of the first to be produced, and has been
energy policy is now centered on renewablesbecause refined since)[12].
these sources are regenerative and for practical PowerFactoryhas the following capabilities[12]:
purposes cannot be depleted. Hitherto Germany relied balanced and unbalanced power flow; short circuit
on nuclear power for 23% of its energy. However, as a based on IEC 60909, VDE 0102/0103, ANSI C37
result of the Fukushima incident and following mass standards; transient stability; optimal power flow; low
anti-nuclear protests across Germany, the government voltage network analysis etc. The software is suitable
announced a reversal of policy that will see all of their for this Study because:
nuclear plants phased out by 2022. Germany is now • ithas the ability to analyse the impact of
basing its energy policy not on nuclear but on distributed generation on the network, and its
renewables.According to Belgovic (2011) in [5], the generator models include induction machines, doubly-
growth in renewables in the first quarter of 2011 in fed induction generators, fuel cells, micro-turbines,
the USA was 25.82% whereas the growth of solar PV-cells, wind turbines, battery storage and single-
energy in 2011 was 104.8%. phase machines.
South Africa has set a target of 10,000 GWh for • it is suitable for radial distribution networks.
renewable energy whereas the Government of Kenya In radial distribution networks, the Jacobian matrix of
is working with the International Renewable Energy the load flow equationsis usually ill-conditioned as a
Agency to explore its potentials in this area [3]. In result of lack of diagonal dominance due to the high
Nigeria, the Ministry of Power [6,7] have identified the R/X ratio of distribution lines.
continued development of renewables and its
integration to the grid as part of its future strategy. 2.2 Radial Distribution Network
Against this background, the objectives of this paper A 15-node radial distribution network sourced
are to highlight the benefits of distributed generation fromreference [8] was modelledin the DIgSILENT
by using a 15-bus radial distribution network, Power Factory software as shown in Figure 1. Table 1
modelled in theDIgSILENT Power Factory
software,todemonstrate the improvement in voltage 1
http://www.digsilent.de/
Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 326
 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

represents the bus data whereas the line data is The total system calculation report before the
embedded in Table 3 below. Table 2 shows the installation of the DG is shown in Figure 2.
assumptions made for the DIgSILENT model. Some of the methods of integrating wind energy to a
Table 1: Nodal Data [8] network have been described in [9] as:
Node No Load (kW + jkVAr) • use of classical squirrel cage induction wind turbine
2, 5, 10, 12 44.10 + j 44.99 generators (WTG).
3,7, 9,11, 15 70.00 + j 71.41
4, 6, 8, 13 , 14 140 + j 142.82 • use of induction WTGs with dynamic slip control.
• induction or synchronous WTGs connected through
Table 2: Assumptions made in DIgSILENT Power Factory power electronic converters.
software • Doubly Fed Induction Generators (DFIG).
Parameter/System Assumptions A synchronous generator representing a modern off-
Distribution line Ground laying shore wind turbine (as a DG) was modelled in
Length of line 1 kilometre for each branch DIgSILENT Power Factory software(Figure 3) and
Line model Lumped parameter (Pi)
Load type Constant power placed separately on buses 2 and 4 to investigate the
Nominal frequency 50 Hz impact of losses.The choice of these buses is arbitrary
System type 3-phase AC hence the authors have identified the need for further
Rated voltage 12 kV studies to account for the optimal allocation of DG.
Nominal voltage 11 kV The wind turbine is assumed to generate 60% active
Upper limit – 1.055 per unit power and 40% of reactive power.
Voltage limits Lower limit – 0.95 per unit
Bus 1 Slack or reference bus

Figure 1: 15-node modelled in DIgSILENT Power Factory software

Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 327

 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

From Figure 2, with an external infeed of 1.29 MW, Q\W]^U XYZUT 0.774
QRRSTUVW XYZUT([) = =
1.31 MVar, the Active power of the DG = 60 percent of XYZUT _S\WYT 0.9
1.29 MW = 0.774 MW = 0.86 MVA (1)
Reactive power of the DG = 40 percent of 1.31 MVAr With the apparent power of 0.86 MVA, it was found
= 0.524 MVAr that the synchronous generator was overloaded hence
Under practical situations, the wind plant the value of 1 MVA was chosen instead. Other
manufacturer will provide the P-Q diagram of the parameters were left at default values as shown in
generator at the point of coupling with the Figure 4.
distribution network. It must be ensured that the wind Under practical purposes an optimal allocation of DG,
turbine generator is equipped with adequate reactive taking into account physical constraints such as
power capability to achieve appropriate reactive available capacity, costs, location and size etc will be
power regulation. The assumption above, in the carried out first to determine the appropriate buses;
absence of the appropriate P-Q diagram, is to ensure this is the subject of further research. Also the default
that sufficient reactive power is available to ensure values of the synchronous generator as set in the
satisfactory operation of the wind turbine generator. software were used and no attempt was made to fine
With a nominal voltage of 11kV and assumed power tune the control parameters.The dynamic interaction
factor of 0.9, the nominal apparent power of the DG between the wind turbine generator and the
can be calculated as: distribution network was not modelled as the main
interest in this paper is the steady state operation.

Figure 2: Total system calculation report before the introduction of DG

Figure 3: Synchronous Machine Basic Data

Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 328
 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

Figure 4: Synchronous Machine Data

3. RESULTS The results achieved by having no DG, placing the DG

3.1 Reduction in Technical Losses at either bus 2 or 4 are shown in Table 3.
Losses can be classified as technical or non-technical The line losses = 3 x I2R (2)
losses (comprising of illegal connections of electricity, where I is the current and R the resistance of the line.
metering errors and deficiencies, billing and The initial losses shown in Figure 2 as 0.06 MW from
processing errors etc) as shown in Figure 5. The the load flow result has been validated in Table 3.
combined losses could be up to 30% or more The technical losses have reduced from 0.06MW with
particularly in developing countries and this requires no DG to 0.03 MW when the DG was placed at bus 2
a major high level intervention. It has been shown in (i.e. 50% reduction) and to 0.02 MW when placed on
[2], using the transmission and distribution losses (% bus 4 (i.e. 67% reduction).
of output) data compiled by the World Bank from Further minimization of technical losses can be
1971 to 2011 for some arbitrarily chosen developed achieved by controlling the switchable reactive power
and developing countries, that for most developed sources, generator terminal voltages, transformer tap
countries the technical losses are between 4% and ratios and phase-shifters using a technique proposed
10%. The subject of non-technical losses have been in [13].
discussed elsewhere in [10] by Nwodo and Ekwue in
1992.

Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 329

 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

Figure 5: Typical Classification of Losses in a Distribution Network

Table 3: Comparison of line currents and technical losses with no DG, DG at bus 2 and DG at bus 4
Before DG After DG inserted in bus 2 After DG inserted in bus 4

Sendin Receivin R X Current I Technical Losses Current I Technical Current I Technical

g node g node (ohms) (ohms) (kA) (MW) (kA) Losses (MW) (kA) Losses (MW)
1 2 1.35309 1.32349 0.096 0.03741 0.047 0.00897 0.046 0.00859
2 3 1.17024 1.14464 0.057 0.01141 0.056 0.01101 0.012 0.00051
3 4 0.84111 0.82271 0.031 0.00242 0.031 0.00242 0.021 0.00111
4 5 1.52348 1.0267 0.004 0.00007 0.003 0.00004 0.003 0.00004
4 6 1.19702 0.8074 0.011 0.00043 0.011 0.00043 0.011 0.00043
4 7 2.23081 1.5047 0.006 0.00024 0.005 0.00017 0.005 0.00017
3 8 1.79553 1.2111 0.02 0.00215 0.02 0.00215 0.02 0.00215
8 9 2.44845 1.6515 0.009 0.00059 0.009 0.00059 0.009 0.00059
9 10 2.01317 1.3579 0.004 0.00010 0.003 0.00005 0.003 0.00005
2 11 2.01317 1.3579 0.009 0.00049 0.009 0.00049 0.009 0.00049
11 12 1.68671 1.1377 0.003 0.00005 0.003 0.00005 0.003 0.00005
2 13 2.55727 1.7249 0.027 0.00559 0.027 0.00559 0.027 0.00559
13 14 1.0882 0.734 0.011 0.00040 0.011 0.00040 0.011 0.00040
13 15 1.25143 0.8441 0.006 0.00014 0.005 0.00009 0.005 0.00009
Total 0.06149 0.03246 0.02027

3.2 Improvement in Voltage Profile software, todemonstrate the improvement in voltage

Before the introduction of distributed generation, profile as well as the reduction in technical losses.
lower voltages (i.e. below the stated 0.95 pu minimum Fixed (or no load, shunt or iron) losses as a result of
value stipulated) were noted at buses 6, 7, 9 and 10. hysteresis and eddy current losses in the iron core of
Bus 10 had the lowest voltage because it is farthest transformers have not been included in this model.
from the source (i.e. bus 1). They are usually treated as constant and independent
The voltages improved as the DG was inserted in bus 2 of network loading. Though only wind energy, as an
and then bus 4 as shown in Table 3. More system example of a renewable source, was considered in this
studies and results are available in [11] by Akintunde. paper, similar conclusions will be reached by using
other forms of renewable energy such as solar,
4. CONCLUSIONS biomass, geothermal etc.
This paper has highlighted the benefits of distributed
generation by using a 15-bus radial distribution
network, modelled in the DIgSILENT Power Factory

Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 330

 IMPACT OF DISTRIBUTED GENERATION ON DISTRIBUTION NETWORKS A. O. Ekwue & O. A. Akintunde

Table 3: Comparison of bus voltages with no DG, DG at Memorial Lecture, Faculty of Engineering,
bus 2 and DG at bus 4 University of Nigeria, Nsukka on 20 June 2014, pp
1-16
Before After DG inserted in bus After DG inserted in bus
Bus
DG 2 4 [3] Turkish International Model United Nations
1 1 1 1 (TIMUN’14) “Promoting Usage of Renewable
2 0.971 0.986 0.987
Energy in Africa”,
3 0.957 0.972 0.985 http://timun.gen.tr/wp-
4 0.951 0.966 0.988 content/uploads/2014/10/ECOSOC-Efe-Birkan-
5 0.95 0.965 0.987 renewable-energy-Afrika.pdf (accessed on 24th
6 0.949 0.964 0.986 November 2014)
7 0.949 0.964 0.986
[4] Department of Energy & Climate Change, “UK
8 0.95 0.965 0.978 Renewable Energy Roadmap Update 2012”, 27th
9 0.946 0.961 0.974 Dec2012, pp 1-67
10 0.945 0.96 0.973
11 0.968 0.983 0.983
[5]Begovic M., “Distributed Generation and
12 0.967 0.982 0.982
Photovoltaics”, presented in Dublin, Ireland in
September 2011.
13 0.958 0.973 0.974
14 0.956 0.971 0.971 [6] Nebo C. and Kuchi H., “Federal Ministry of Power
15 0.957 0.972 0.972 presentation at the Ministerial Platform for the
period of May 2011 to May 2013 (July 23rd 2013)”
The work described in this paper fits in with the [7] Nnaji B.,“Power Sector Outlook in Nigeria:
Nigeria’s Ministry of Power strategy of continued Government Renewed Priorities”, June 2011
development of renewable energy sources and http://www.
sec.gov.ng/files/Prof%20Nnaji%20Presentation.p
addressing high energy losses in the transmission and
df, accessed on 14th April 2014
distribution networks.Future investigations will
address the optimum allocation of DG taking into [8] Ashokkumar R.and Aravindhababu P., “An
account physical constraints (such as cost, location Improved Power flow Technique for Distribution
Systems,” Journal of Computer Science,
etc) to reduce technical losses and improve on voltage
Informatics & Electrical Engineering , vol. 3, no. 1,
profiles as well as the dynamic interaction between 2009, pp 1-8.
the wind turbine generator and the distribution
[9] Bousseau P.,Fesquet F.,Belhomme R.,Nguefeu S.
network.
and Thai T., “Solutions for the Grid Integration of
Wind Farms – a Survey”, Wind Energy published
5. ACKNOWLEDGEMENT by John Wiley vol. 9, 2006, pp 13-25
The authors are grateful to Brunel Institute of Power
[10] Nwodo T. C. and Ekwue A.O., “The effect of
Systems, Brunel University London for providing the metering deficiencies on the economics and
facilities for carrying out the studies reported in this reliability of developing power systems”,”,
paper. Proceedings of the 7th IEE International
Conference on Metering Apparatus and Tariffs for
6. REFERENCES Electricity Supply held in Glasgow, Nov. 1992 pp 1
–4
[1] Ekwue A. O.,Nanka-Bruce O., Rao J. and McCool D.,
“Dynamic Stability Investigations of the Fault Ride [11] Akintunde O. A., “Voltage Stability Analysis of
Through Capabilities of a Wind Farm”” Paper ID 99, Radial Distribution Networks”,, Unpublished M.Sc
Proceedings of the 16th Power Systems dissertation, Brunel University London.
Computation Conference, 14 -18 July 2008, September 2014 pp 1 – 71
Glasgow. [12] DIgSILENT Power Factory 14.1 Software Manual
[2] Ekwue A. O., “The role of R&D in revitalizing an (May 2011), vol. 1/2 and vol. 2/2
effective power system for sustainable [13] Ekwue A.O., “Improved Real-Power Loss
development in Nigeria”, ”, Invited Symposium Minimisation”, Nigerian Journal of Technology,
Paper presented at the 15th Herbert Macaulay vol. 9, no.1 September 1985, pp 31-37.

Nigerian Journal of Technology, Vol. 34, No. 2, April 2015 331