See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/237305036

A Benchmark Low Voltage Microgrid Network

Article · January 2005

CITATIONS READS

208 9,942

3 authors:

Stavros Papathanassiou Nikos D. Hatziargyriou

National Technical University of Athens National Technical University of Athens
177 PUBLICATIONS   4,669 CITATIONS    600 PUBLICATIONS   19,209 CITATIONS   

SEE PROFILE SEE PROFILE

K. Strunz
Technische Universität Berlin
131 PUBLICATIONS   3,110 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

DISTRES View project

ANEMOS.plus View project

All content following this page was uploaded by Nikos D. Hatziargyriou on 04 June 2014.

The user has requested enhancement of the downloaded file.

Presented at the CIGRE Symposium "Power systems with dispersed generation: technologies, impacts on development,
operation and performances", April 2005, Athens, Greece.

A BENCHMARK LOW VOLTAGE MICROGRID NETWORK

Stavros Papathanassiou* Nikos Hatziargyriou Kai Strunz

National Technical University of Athens University of Washington
GREECE USA

Keywords: Distributed Generation, Distribution Networks, LV Networks, Microgrids

1. INTRODUCTION

The increasing penetration of distributed generation resources to the low voltage (LV)
grids, such as photovoltaics, CHP micro-turbines, small wind turbines in certain areas and
possibly fuel cells in the near future, alters the traditional operating principle of the grids. A
particularly promising aspect, related to the proliferation of small-scale decentralized
generation, is the possibility for parts of the network comprising sufficient generating
resources to operate in isolation from the main grid, in a deliberate and controlled way. These
are called Microgrids and the study and development of technology to permit their efficient
operation has recently started with a great momentum ([1,2]).
Microgrids are foreseen within public distribution grids and therefore suitable study case
networks are required to perform simulation and analysis tasks. Moreover, standardizing
study case grids to provide “benchmark” networks suitable for Microgrid design would
further enhance their merit and utility.
The objective of this paper is to present and discuss a benchmark LV network developed
within the EU project “Microgrids”, Contract ENK5-CT-2002-00610 and later adopted as a
benchmark LV system by CIGRE TF C6.04.02: “Computational Tools and Techniques for
Analysis, Design and Validation of Distributed Generation Systems”. The network consists of
an LV feeder, while a more extended multi-feeder version is also included in the Appendix.
The emphasis is placed on the network itself, rather than on the microsources connected and
the control concepts applied. The benchmark network maintains the important technical
characteristic of real utility grids, whereas, at the same time, it dispenses with the complexity
of actual networks, to permit efficient modeling and simulation of microgrid operation.

2. THE BENCHMARK LOW VOLTAGE FEEDER

2.1 General Characteristics of the LV Network

Before presenting the benchmark network, some important technical characteristics of

* NTUA-Electric Power Division, 9 Iroon Polytechniou st., 15780 Athens, Greece - email: st@power.ece.ntua.gr
public LV distribution grids are summarized (pertaining more to European networks):
Structure. The majority of LV public distribution networks have a radial layout, with a
number of LV feeders starting from the LV busbars of the infeeding MV/LV substation. Each
feeder may include one or more spurs (branches). Consumers are connected anywhere along
the feeder or its spurs.
Symmetry. The connection of single-phase consumers makes LV networks inherently
unbalanced. In addition, single-phase lines may exist, particularly as feeder branches.
Substation. The MV/LV substation feeding the LV network typically comprises a single
transformer with a rating of a few hundred kVA up to 1 MVA. The transformer is equipped
with off-load taps at the HV winding, providing a typical regulation range of ±5%. Its
connection group is usually Dyn11, corresponding to a delta-connected primary and a wye-
connected secondary winding.
Protection. The only protection encountered in public LV networks typically consists of
simple phase overcurrent devices, most commonly fuses. The MV/LV transformer is
protected by fuse links at the MV side. A general protection element may not exist at the
output of the transformer LV winding, whereas each LV feeder is protected by its own fuses.
No other protection means are utilized along the feeder or its branches.
Line types. LV network lines are either underground cable lines, encountered mainly in urban
areas with a high load density, or most commonly overhead lines, traditionally constructed by
Al (or Cu) bare conductors. Ease of installation and environmental reasons have led to the
extensive use of twisted insulated conductors for overhead LV lines during the last decades.
Earthing. Using the classification of IEC 60364, public LV networks are either of the TN or
the TT type. The principle of each earthing scheme is illustrated in Fig. 1. More information
on the subject is provided in [3,4].

3
MV LV Ph
N PEN
PE N N
PEN
PE

TN-C TN-S TT

Figure 1. Principle of the TN and TT earthing schemes.

2.2 Description of the Benchmark LV Feeder

Based on the basic requirements discussed in the previous section, the study case LV
feeder is illustrated in Fig. 2. The feeder is an overhead line with twisted XLPE cable, serving
a suburban residential area with a limited number of consumers connected along its length, as
well as at the end of the branch at its middle. Line types are marked on the diagram and the
respective parameters are given in the Appendix. Section lengths can be determined from the
number of poles, given the fixed pole-to-pole distance of 35 m. The network neutral is multi-
grounded, at the substation, at every second pole and at each consumer connection point. At
the end of the lateral branch, a connection of the neutral may exist to an adjacent LV line (fed
by another substation).

2
 20 kV

20/0.4 kV, 50 Hz, 400 kVA

Off-load Tap Changer
uk=4%, rk=1%, Dyn11
19-21 kV in 5 steps
3Ω
0.4 kV 3+N

Overhead line
3x250 A 4x120 mm2 Al XLPE
twisted cable
Pole-to-pole distance = 35 m
Other lines
3+N
40 Ω

3+N+PE 4x6 mm2 Cu Possible neutral bridge

to adjacent LV network
30 m
Single residencial 40 Ω 2Ω
40 Ω
consumer
3Φ, Is=40 A
40 Ω
Smax=15 kVA
S0=5.7 kVA 3x70mm2 Al XLPE +
54.6mm2 AAAC 30 m
Twisted Cable 3x50 mm2 Al +35mm2 Cu XLPE

10 Ω
3+N+PE
3+N+PE 1+N+PE
4x25 mm2 Cu
40 Ω Appartment building
30 m
5 x 3Φ, Is=40 A
3+N+PE 8 x 1Φ, Is=40 A
Smax=72 kVA
Group of 4 residences 40 Ω
S0=57 kVA
4 x 3Φ, Is=40 A
Smax=55 kVA
S0=25 kVA

40 Ω

Single residencial
4x6 mm2 Cu 3+N+PE consumer
40 Ω
30 m 3Φ, Is=40 A
3+N+PE Smax=15 kVA
40 Ω S0=5.7 kVA
4x16 mm2 Cu
40 Ω
30 m
1+N+PE
LEGEND
Appartment building
1 x 3Φ, Is=40 A LV network line
6 x 1Φ, Is=40 A Connection cable
Smax=47 kVA
S0=25 kVA To consumer installation
Overhead line pole
Point of connection (supply)
40 Ω Neutral earthing
Fuses
3+N+PE 5 conductor cable (3 phases,
neutral, protective earth)

Figure 2. The benchmark LV feeder, in its standard (“non-microgrid”) form.

The arrangements at the service connection of each customer are presented in more detail
in Fig. 3. Each service connection includes the electricity meter and an overcurrent protection
element (fuse links or a miniature circuit breaker for small consumers). For the service cable,
a standard 30 m length is adopted in Fig. 2. The earthing scheme of the network may be either

3
of the TN or the TT type, depending on the connection or not of the PE conductor of the
consumer installation to the network neutral. The 40 Ω earthing resistances noted on the
diagram correspond to a standardized conductive rod, 2.5 m long by 0.02 m in diameter,
buried in homogeneous conductive earth of 100 Ω.m resistivity. The apartment building on
the lateral is supposed to have a more effective earthing arrangement (either multiple rods or
some sort of foundation earth).
MCB for small sizes
LV line Fuse links for larger Consumer installation
Possible Meter main switchboard
(TN-C network)
junction box cabinet (TN-S or TT system)
Meter to switch-
Service Meter board cable N bar

cable 3 3
M N
PEN
PE PE bar
PEN

3
Connection point
Bonding jumper used for TN DG
for DG source of
PE and N conductors isolated for TT the consumer
Consumer
ground

Figure 3. Typical service connection arrangement.

2.3 Consumer Demand Characteristics

Each consumer of the feeder is characterized by a maximum permissible current, Is, which
corresponds to the rated current of the overcurrent protection element in the connection box
(Fig. 3). The maximum demand Smax of each consumer group, also given in Fig. 2, depends on
the number of individual consumers within each group, and is found using standardized
coincidence factors for residential consumers, which become smaller as the number of
consumers increases (e.g. [5]). For this reason, the contribution S0 of each group to the
maximum demand of the feeder will be further reduced, as given in Fig. 2. The total
maximum demand of the feeder is 116.4 kVA. The power factor of all consumers may be
assumed equal to 0.85 lagging. Aggregate daily load curves are provided in the Appendix.

3. THE BENCHMARK LOW VOLTAGE MICROGRID NETWORK

Based on the LV feeder of Fig. 2, the benchmark LV microgrid network shown in Fig. 4 is
derived. It includes representative sources from all currently important (or emerging, but
promising) technologies, such as photovoltaics, microturbines (CHP generation), wind
turbines and fuel cells.
Specific technical details, models for individual sources and control concepts are beyond
the scope of this paper and will be specified in application studies. In Fig. 4, only relevant
installation locations and sizes are indicated. The total installed capacity of the microsources
is about 2/3 of the maximum load demand of the feeder (~100 kW), to provide the possibility
of simulating load management scenaria.
To support the islanded operation of the microgrid, a fast-responding central storage unit
is also considered, which may be either a battery inverter, or any other device with
sufficiently fast response to undertake the frequency regulation task upon disconnection from
the grid (e.g. a flywheel). Notably, this constitutes a centralized control approach.
Alternatively, the individual microsources might be equipped with local storage (e.g. batteries
or ultra-capacitors) and suitable controls to ensure a decentralized active power/frequency
concerted regulation ([6]).

4
 20 kV

20/0.4 kV, 50 Hz, 400 kVA

Off-load TC
uk=4%, rk=1%, Dyn11
19-21 kV in 5 steps
3Ω
0.4 kV 3+N

Overhead line
Circuit Breaker 4x120 mm2 Al XLPE
instead of fuses twisted cable
Pole-to-pole distance = 35 m
Other lines
3+N
40 Ω
Single residencial
consumer
3Φ, Is=40 A 3+N+PE 4x6 mm2 Cu Possible neutral bridge
Smax=15 kVA to adjacent LV network
30 m
S0=5.7 kVA
40 Ω 40 Ω 40 Ω 2Ω
Flywheel storage 4x16 mm2 Cu
~
(or batteries) 30 m
3Φ, 30 kW 3x70mm2 Al XLPE +
10 Ω 54.6mm2 AAAC 30 m
Twisted Cable 3x50 mm2 Al +35mm2 Cu XLPE
Circuit Breaker
Possible sectionalizing CB 10 Ω
Microturbine
3Φ, 30 kW
3+N+PE
4x25 mm2 Cu ~ ~
Group of 4 residences 3+N+PE
40 Ω ~
4 x 3Φ, Is=40 A 30 m
Smax=50 kVA Appartment building
S0=23 kVA 5 x 3Φ, Is=40 A
3+N+PE 8 x 1Φ, Is=40 A
Wind Turbine ~ ~
3Φ, 10 kW ~ Smax=72 kVA
S0=57 kVA
Photovoltaics
1Φ, 4x2.5 kW
40 Ω
40 Ω

Appartment building 3+N+PE 4x6 mm2 Cu 3+N+PE

1 x 3Φ, Is=40 A Single residencial
6 x 1Φ, Is=40 A 30 m consumer
Smax=47 kVA 3Φ, Is=40 A
S0=25 kVA Smax=15 kVA
1+N+PE 4x16 mm2 Cu
40 Ω S0=5.7 kVA
Fuel Cell 30 m
3Φ, 10 kW Photovoltaics
40 Ω 1Φ, 3 kW
40 Ω

Figure 4. Benchmark LV microgrid network.

Compared to the standard LV feeder of Fig. 2, in Fig. 4 the fuses at the feeder departure
have been replaced by a circuit breaker, in order to permit the controlled connection and
isolation of microgrid from the main grid. A second sectionalizing breaker may also be
inserted at the middle of the feeder, if selective isolation of faulted parts of the microgrid is to
be studied. However, in such a case, suitable frequency regulating means should be foreseen
in each isolated section.
The earthing arrangements of the network remain unchanged for microgrid operation.
Preliminary investigations have shown that this is acceptable ([7]), although further study
may be required on this subject. Regarding the protection philosophy, devices and settings, it
is certain that modifications will be required to the traditional LV network practice, which
have not been incorporated in the study case network of Fig. 4.

5
4. CONCLUSIONS

In this paper a benchmark LV microgrid network is presented, which is suitable for steady
state and transient simulations. The study case network is based on a standard LV feeder,
where microsources and storage devices of various types are connected. A more extended
network is also provided in the Appendix, to facilitate the simulation of multi-feeder
microgrids or multiple microgrids within the same LV grid.

5. ACKNOWLEDGEMENT

The work presented in this paper has been performed within the project “MICROGRIDS:
Large Scale Integration of Micro-Generation to Low Voltage Grids” (Contract ENK5-CT-
2002-00610), funded by the EU. The authors wish to thank Mr. N. Soultanis for calculating
the zero-sequence parameters of the lines.

6. REFERENCES

[1] EU Project “MICROGRIDS: Large Scale Integration of Micro-Generation to Low

Voltage Grids (ENK5-CT-2002-00610)”. Website: http://microgrids.power.ece.ntua.gr/.
[2] R. Lasseter, A. Akhil, C. Marnay, J. Stephens, J. Dagle, R. Guttromson, A.S. Meliopoulos,
R. Yinger and J. Eto, “White Paper on Integration of Distributed Energy Resources–The
CERTS MicroGrid Concept”. LBNL-50829, US Department of Energy, Office of Power
Technologies. Contract DE-AC03-76SF00098.
[3] B. Lacroix, R. Calvas, “Earthing systems in LV”. Cahier Technique No. 172, Schneider
Electric, 2000.
[4] B. Lacroix, R. Calvas, “Earthing systems worldwide and evolutions”. Cahier Technique
No. 173, Schneider Electric, 1995.
[5] “Standardization of Electricity Meters”, Distribution Directive No.45, Public Power
Corporation (PPC) of Greece, 1982.
[6] D. Georgakis, S. Papathanassiou, N. Hatziargyriou, A. Engler, C. Hardt, “Operation of a
prototype Microgrid system based on micro-sources equipped with fast-acting power
electronics interfaces”. Proceedings of PESC’04, June 2004, Aachen, Germany.
[7] “Safety Guidelines: Report with proposed earthing and safety procedures to ensure safe
operation of the Microgrid”, Microgrids Deliverable DE1, Dec. 2004.

7. APPENDIX

A more extended study case LV network is included in Fig. 5, which comprises two
additional LV lines, compared to the benchmark network of Fig. 2. The first is a dedicated
underground cable line, serving a workshop, whereas the other one is an overhead line serving
a small commercial district. The diagram provides for each consumer the same information as
in Fig. 2. On the commercial load feeder, where a large number of single-phase consumers are
connected, the respective phases are also noted.
The study case network of Fig. 5 permits the simulation of microgrids with multiple LV
feeders and diverse load types, or even different microgrid entities within the same LV
network (e.g. by considering that the commercial line forms a second microgrid, with CHP
microturbines as microsources). Depending on the part of the network, which forms the
microgrid (or microgrids), sectionalizing switches need to be inserted at selected locations
and suitable microsource scenaria must be adopted.

6
 20 kV
20/0.4 kV, 50 Hz, 400 kVA
Off-load TC
uk=4%, rk=1%, Dyn11
19-21 kV in 5 steps
3Ω
0.4 kV 3+N

Overhead line
4x120 mm2 Al XLPE
Underground line Overhead line
3x250 A 3x160 A 3x150 mm2 Al + 3x160 A Bare conductors
twisted cable
50 mm2 Cu XLPE cable 4x50/35/16 mm2 Al
Pole-to-pole distance = 35 m
Pole-to-pole distance = 30 m
4x35 mm2 Al
3+N conductors 3+N
3+N
40 Ω 3Φ, Is=40 A
Smax=20 kVA
Single residencial S0=11 kVA
40 Ω
consumer Possible neutral bridge 3Φ, Is=63 A 4x50 mm2 Al
3Φ, Is=40 A to adjacent LV network 40 Ω Smax=30 kVA conductors
Smax=15 kVA S0=16.5 kVA
S0=5.7 kVA 40 Ω 2Ω
1Φ, Is=40 A 4x1Φ, Is=40 A
40 Ω 200 m Phase: a Phase: abcc 4x16 mm2 Al
Residential Twisted Cable Smax=8 kVA Smax=25 kVA conductors
load 3x70mm2 Al XLPE + S0=4.4 kVA S0=13.8 kVA
54.6mm2 AAAC
40 Ω
40 Ω

Appartment building Industrial Commercial

Group of 4 residences 5 x 3Φ, Is=40 A
4 x 3Φ, Is=40 A load load 2x1Φ, Is=40 A 1Φ, Is=40 A
40 Ω 8 x 1Φ, Is=40 A Phase: ab Phase: c
Smax=55 kVA Smax=72 kVA
S0=25 kVA Smax=16 kVA Smax=8 kVA
S0=57 kVA S0=8.8 kVA S0=4.4 kVA

40 Ω 40 Ω 4x35 mm2 Al
conductors

4x1Φ, Is=40 A
40 Ω Phase: abbc
Workshop Smax=25 kVA
3Φ, Is=160 A S0=13.8 kVA
Smax=70 kVA
Single residencial S0=70 kVA 3x1Φ, Is=40 A
Appartment building 40 Ω Phase: abc
1 x 3Φ, Is=40 A consumer
Smax=20 kVA
6 x 1Φ, Is=40 A 3Φ, Is=40 A
S0=11 kVA
Smax=47 kVA Smax=15 kVA
S0=25 kVA S0=5.7 kVA
40 Ω

Figure 5. Benchmark LV network for the study of multi-feeder or multiple LV microgrids.

7
 Aggregate daily load curves for the three load types of the benchmark networks are shown in
Fig. 6. Impedance data for the various line types are provided in Table 1. Neutral resistances
are given where the neutral has a different cross section than the phases. Calculated zero
sequence impedances are quoted for selected line types, appearing in the benchmark network
of Fig. 4 (derived for combined neutral and earth return path of the current).

Table 1. Impedance data for the benchmark network lines

Rph Xph Rneutral R0 X0
Line type
(Ω/km) (Ω/km) (Ω/km) (Ω/km) (Ω/km)
1 OL - Twisted cable 4x120 mm2 Al 0.284 (1) 0.083 1.136 0.417
2 OL - Twisted cable 3x70 mm Al + 54.6 mm AAAC 0.497 (1) 0.086
2 2
0.630 2.387 0.447
3 OL - Al conductors 4x50 mm2 equiv. Cu 0.397 (1) 0.279
4 OL - Al conductors 4x35 mm2 equiv. Cu 0.574 (1) 0.294
2
5 OL - Al conductors 4x16 mm equiv. Cu 1.218 (1) 0.318
2 2
6 UL - 3x150 mm Al + 50 mm Cu 0.264 (2) 0.071 0.387 (2)
2
7 SC - 4x6 mm Cu 3.690 (3) 0.094 13.64 0.472
2
8 SC - 4x16 mm Cu 1.380 (3) 0.082 5.52 0.418
9 SC - 4x25 mm2 Cu 0.871 (3) 0.081 3.48 0.409
10 SC - 3x50 mm2 Al + 35 mm2 Cu 0.822 (2) 0.077 0.524 (2) 2.04 0.421
11 SC - 3x95 mm2 Al + 35 mm2 Cu 0.410 (2) 0.071 0.524 (2)
OL: Overhead line, UL: Underground line, SC: Service connection
(1)
: Ohmic resistance at 50 oC conductor temperature
(2)
: Ohmic resistance at temperature 90 oC for phase conductors and 20 oC for the neutral
(3)
: Ohmic resistance at temperature 70 oC for all conductors

100
90
Demand (% of maximum)

80
70
60
50
40
30
20
10 Residential Industrial Commercial
0
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00

Figure 6. Daily load curves for the three load types of the benchmark LV networks.

Summary
Microgrids are foreseen to be developed within public distribution grids and therefore
suitable study case networks are required to perform simulation and analysis tasks.
Standardizing study case grids to provide “benchmark” networks suitable for microgrid
development, further enhances their merit and utility. In the paper a benchmark LV network is
presented and discussed, consisting of a LV feeder supplying a suburban residential area. A
more extended version of the benchmark network is also included, suitable for the study of
multi-feeder or multiple microgrids. The emphasis is placed on the network characteristics,
while microsources, representative of all currently important technologies, are connected to
selected nodes. The benchmark network maintains the important technical characteristic of
real life utility grids, while dispensing with the complexity of actual networks, to permit
efficient modeling and simulation of microgrid operation.

View publication stats