1

Building A New Overlay Grid For Europe

Hakan Ergun, Graduate Student Member, IEEE, Jef Beerten, Graduate Student Member, IEEE,
Dirk Van Hertem, Senior Member, IEEE
AbstractThe strong growth and the expected further increase
in renewable energy generation in Europe requires a fundamental
upgrade of the transmission system. A potential option is to
realize these upgrades at a higher voltage by constructing a
new overlay grid or supergrid. Such a supergrid is likely to be
built using VSC HVDC technology, especially when considering
that a significant part will be offshore. The paper will first
discuss how such a grid should be conceived. Aspects that will
be covered are the potential topology of such a grid and the
balance between mesh size and the connection to the AC system.
It will be discussed how such a new overlay grid with multiple
controllers interacts with the existing system. The second aspect
that will be covered is the stepwise development of such a system.
A new overlay grid will not be conceived in a single phase, but
link by link. The differences between green field and brown
field approach are highlighted and shown using an example.
Index TermsVSC HVDC, DC Grid, Integrated approach,
Transmission system investments, supergrid

mention that the way the future grid will look like is in a large
part detirmened by the framework in which it is constructed.
This includes the ownership of the grid (who will invest in
the transmission assets) as well as the remuneration for these
investments in the form of a fair tariffication. The manner in
which the international, or rather inter-zonal, cooperation will
evolve are clearly of importance as well.
This paper will focus on the development or planning of an
overlay grid in Europe. Such an overlay system transmission
system will be consequence of the supply and demand that
interact, which will be discussed in section II. Section III
discusses the requirements of the future grid. In section IV,
the interaction between the new overlay grid and the existing
system is discussed. The stepwise development of the grid is
dealt with in section V and the effects of timed investments
is shown in section V-A. Section VI summarizes the paper.

I. I NTRODUCTION

II. F ORSEEN CHANGES IN ENERGY SUPPLY AND DEMAND

HE move towards a society with a smaller CO2 footprint in Europe has a considerable effect on the share
of renewable energy sources within the European electricity
supply.
These changes will result in a fundamental increase in the
transmission of large amounts of power over longer distances.
Furthermore, these flows are of an increasingly variable nature.
Rather than extending the current highly loaded 400 kV
AC system, a fundamental upgrade is required, with new
energy corridors or electricity highways with a transmission
capacity that is one order of magnitude higher than that of the
system which is now in use. This evolution is receiving a lot of
support. Amongst others, the European Commission [1], the
power sector [2], [3] and presure groups such as environmental
organizations [4] support this development which eventually
can lead to a pan-European overlay grid.
There are many visions of pan-European overlay grids
proposing different grid topologies and transmission technologies. Most visions propose VSC HVDC (Voltage Source
Converter High Voltage Direct Current) transmission as the
preferred technology for such transmission grids [4][7]. As
the design and construction of such an overlay grid is not only
a major investment but also a completely new technological
approach, it will fundamentally alter the European electricity
system in the way it is operated and planned. Therefore the
topology and technology of such a new overlay grid must
be well conceived and analyzed in depth. It is important to

In Germany alone, 28,3 GW of wind and 26 GW of

solar power is already installed which equals to 16,5% resp.
15,1 % of the installed capacity in Germany [8]. Especially for
wind and solar power generation a strong further increase of
generation capacity is expected until the mid of this century
[9][11]. A strong increase in generation from renewables
is also expected in the rest of Europe, with an anticipated
increase in installed capacity from offshore wind power to
exceed 150 GW [12]. At that moment, the installed onshore
wind power generation will amount to about 250 GW. As such,
a big part of the electricity generation mix will be provided
by wind power [12]. The best sites for wind power plants are
situated in the North Sea region and the western part of Europe
(Fig. 1a). On the other hand the best sides for large scale
solar power plants are situated around the Mediterranean Sea:
in Spain and Northern Africa (Fig. 1b). Classical generation
(fossil fuels and nuclear) will not disappear and completely be
replaced by RES, but the system will evolve towards a new
generation mix, with the generation not necessarily located at
the same location as it is now. Where the generation originally
was placed near the load center or near the source of primary
energy, a shift is noticed towards the latter. For the European
energy supply this means a shift of generation towards the
borders of the system, specifically towards harbors.
Secondly, the generation mix will contain more renewable
energy sources, with a more pronounced need for balancing
services (not necessarily local) and the limited capacity factor
of these energy sources. While the transmission investments
need to accommodate the generation capacity (power rating),
the transmission investments connecting towards that generator
will be less efficient as the link will have a lower amount
of energy to transport. This effect is not only valid for
direct connections to renewable energy sources, but throughout

H. Ergun, J. Beerten and D. Van Hertem are with the Department of

Electrical Engineering (ESAT/ELECTA), KU Leuven, Kasteelpark Arenberg
10, 3001 Heverlee, Belgium, http://www.esat.kuleuven.be/electa.
Jef Beerten holds a Ph.D. fellowship from the Research Foundation
Flanders (FWO).
Contact: [email protected], [email protected],
[email protected]

978-1-4673-2729-9/12/$31.00 2012 IEEE

(a) Average wind profile in Europe (source: EEA)

the grid a higher reserve transmission capacity is needed

to accommodate for an equal amount of energy transferred.
When hydro energy from Scandinavia or the Alps is used
to balance the fluctuating generation from renewable energy
sources, they also need sufficient transmission capacity to
perform this balancing.
Because the load centers are situated in central Europe (the
region roughly comprising of the Benelux, the north of France
(including Paris), (Western) Germany and north of Italy) and
England (Fig. 2), it will be inevitable to build new north-south
and west-east connections within Europe to secure electricity
supply to the demand centers. Furthermore, the load centers
are getting more concentrated as there is a noticeable shift of
the population and industry towards the city.
Although the energy consumption is getting increasingly
efficient, the consumption of electrical energy is expected to
continue to rise over the next decades. Not only is there a
general tendency towards a higher level of comfort, there
is also a fundamental shift of energy consumption towards
electricity, e.g. through the anticipated introduction of electric
vehicles.
III. R EQUIREMENTS FOR THE FUTURE GRID

(b) Average solar radiation in Europe (source: SolarGIS)

Fig. 1: Location of wind and solar resources

To achieve the aforementioned visions of a new interconnected system, different technical requirements must be
fulfilled before such an overlay grid can be built. Within this
framework, this paper addresses first the requirements of the
future grid with regards to energy transfer and mesh sizes. The
available technologies to achieve such vision are compared
with these requirements. Besides the layout of the grid also
the technology used for the overlay grid is not necessarily
determined in advance.
A. Basic requirements for the overlay grid

Fig. 2: Population density in Europe (source: IIASA)

As mentioned in section II, the future transmission system

must accommodate all generation and loads in the system, and
this generation and load will fundamentally change. Both the
location and the nature of these elements will change, resulting
in a need for more transmission capacity. The current energy
system is operating up to its limits, and although the current
system can be extended to a certain degree, a fundamental
upgrade is needed on the long run.
Such a grid must comply to some requirements in order to
fundamentally upgrade the system.
First of all, a new overlay grid should be a fundamental
upgrade with regards to power transfer capacities. A current
single circuit 400 kV transmission line can carry 12 GVA, or
even more, and multiple circuits can be connected to a single
pylon to form a single transmission corridor. This fundamental
upgrade requires a technology that is able to transmit electrical
power at a significantly higher rating, up to around 4 8 GVA
per circuit. This would require that the transmission happens
at a higher voltage level than currently used in order to avoid
excessive losses.
Secondly, the new system requires long distance transmission of electrical energy. From remote energy sources in the
North Sea, Scandinavia or the Mediterranean to the load

centers. This high power, long distance transmission should are an order of magnitude lower than the ones that might
happen at a high efficiency in order to reduce transmission be of interest for an overlay grid. Also with respect to the
losses.
operation and control of the DC grid, a lot of questions remain
Because offshore wind is expected to form a considerable unanswered. Current research [15], [16] is ongoing and most
part of the future generation mix, a third requirement is the of the remaining questions relate to the objective of the control
ability to connect to those offshore generation sites. Therefore, rather than the control algorithms. Lastly, the weakest point in
the grid needs a technology that allows undersea connections. VSC HVDC transmission might be the protection. Although a
Furthermore, the overall system security should not be commercial HVDC breaker has recently been introduced [17],
affected or ideally even increase. Concepts such as N-1 and it is still not commercially available at transmission levels.
coordinated frequency and voltage control must apply to the Furthermore, a large number of protection issues have to be
overlay grid as well.
solved to ensure a safe operation of a meshed HVDC grid
Lastly, the new grid should be compatible with the one that [18].
is currently in place. This includes operational procedures that
An alternative to the overlay grid is the extension or
are compatible, but also implies that the resulting energy flows upgrade of the existing grid infrastructure. Because finding
through the overlay grid should not cause any problems in the new transmission corridors is difficult, this can typically be
existing system.
done by making better use of the available Right-of-Way. This
Of course, the solution should be cost effective in compar- can be done through the use of new materials for the conison to other alternatives.
ductors which can carry higher currents, or by using a higher
transmission voltage on existing corridors. Another technology
which has gotten much attention lately is the use of power flow
B. Technology for the overlay grid
controlling devices (phase shifting transformers and FACTS)
The technology used is determined by different investment
to make optimal use of the existing grid infrastructure. This
decisions at a certain point in time as well as on the availability
technology enables the grid operator to shift power flows to
of different technologies at that moment.
different corridors and as such unload certain lines. The result
Currently the focus of discussion about technological chalis a system that flexibly can adapt to different generation
lenges is centered around the HVDC technology. However,
patterns. This solution extends the transmission capacity of the
the HVAC technology should not be discarded as it remains a
grid, but is not a long term solution towards grid extension.
competitive technology. The backbone of the current transmisHowever, in a multi-zonal grid, power flow controlling devices
sion system is built using 400 kV overhead lines. If the HVAC
can form an important aspect of grid extension planning.
system has to be upgraded and expanded in a fundamental way,
System upgrades through making better use of the capacity or
it is very unlikely that this can be realized with overhead lines,
through flow control are a very valuable means to extend the
at least in Europe. The main reason for this is the growing
current system and stretching the use of the infrastructure,
public resistance against them. If the overlay grid has also
but they do not form a structural upgrade which will suffice
to rely on long and high capacity HVAC connections, than
on the long run [19]. Nevertheless, the overlay grid has to
considerable parts of these connections will have to be built
compete with these technologies that focus on a short term
as underground cables or gas insulated lines. This brings new
solution and this will influence the long term grid planning as
technical challenges and comes at a cost. At this moment,
well.
HVAC XLPE cable technology allows to construct cables up to
The conversion of existing HVAC overhead lines to HVDC
500 kV and approximately 2 GVA [13]. The charging currents
technology can be one of the first steps towards a DC overlay
limit the distance of the connection when AC technology is
grid [20]. That way the transmission capabilities of existing
used, although it is possible to provide reactive compensation.
corridors can be increased significantly.
Joints and cable placement are far from trivial, specifically
In summary, although AC technology is a possible techbecause of the limited length of a cable section that can be
nology for a new overlay grid, it is very unlikely that this
transported on a drum ( 1 km). Worldwide only a limited
technology is the preferred one in Europe. The reason are
number of HVAC cables are installed in the EHV grid, and
the difficulties with permitting and public opposition against
most of them are relatively short.
overhead lines and the problems with long distance, high
Based on these considerations, a multi-terminal HVDC grid
power energy transmission from offshore connections. AC
has been proposed for the future overlay grid [6]. Especially
cable technology has inherent issues with regards to long
for an HVDC meshed overlay grid still a lot of challenges exist
distance energy transmission and is not considered to be
and new developments or even complete new technologies are
suitable for an overlay grid. VSC HVDC is a much more
1
needed. Firstly, the VSC HVDC technology requires further
likely contender although that it still has its limitations.
development in terms of transmission capacity. HVDC XLPE
cables are currently available for voltages up to 320 kV and
powers up to approximately 1200 MW [14]. Manufacturers C. Ratings and mesh sizes
offer converter stations with ratings up to 1200 MW. It is
The transmission system is built in a meshed layout, with
clear that, as with AC technology, ratings for cable technology
different voltage levels that each have their specific power
transmission range. In between these voltage levels, transform1 LCC HVDC is currently available for higher power ratings, but is less
suitable for multi-terminal grids and the use in offshore applications [6].
ers are placed which allow energy to flow to and from the more

4
LOLG = n LEHV

IV. I NTERACTION WITH THE EXISTING SYSTEM

A. Effect of building a new overlay grid on the existing one

1-3 GW

Another important aspect not to be overlooked, is the fact

that
the underlaying AC grid will require updates when an
L
=nL
overlay grid is being built. The connection points to the
underlaying AC level might not be able to cope with the
high power injections or withdrawals by the converters. When
discussing a future grid update, it will not be limited to the
newly built layer. Connecting a node in the existing grid to
the overlay grid would require a strong AC node, which can
evacuate or absorb the necessary power, possibly in the order
L
of several gigawatt. At this moment, there are only a limited
number of potential connection points that are able to do
400-800 MVA
this. Significant reinforcements in the AC system are needed.
L
Furthermore,
the overlay grid might result in a different grid
Fig. 3: Grid layout with an additional overlay grid. Mesh sizes will increase
setup compared to before. Much of the generation is planned
and a limited number of converter stations will connect to the existing extra
high voltage (EHV) AC grid (black) with the new overlay grid (OLG) in DC.
at the exterior of the system. The energy flows will therefore
change, with flows from shore to inland. The grid has not been
conceived in this manner, and upgrades are needed.
efficient higher voltage lines. As mentioned, the transmission
Compared to the existing (400 kV) overlay systems, there
lines in the new overlay grid should have a power rating will be fewer connections to the underlying systems. The
which is an order of magnitude higher than the current existing existing AC to AC connections are made by cheap transsystem, meaning somewhere in the range of 48 or even up to formers while the DC connections require more complicated
10 GW per circuit. Also the capacity between the overlay grid and expensive converter stations. As there will be fewer
and the 400 kV grid will be an order of magnitude higher than connections, each connection will have a large influence on the
that between the 400 kV system and the layer below. Where security of the system when taking N-1 security into account.
400 800 M V A transformers (possibly several in parallel)
The controllable power injections of each DC converter
are typically used for the transformation to lower levels, the makes the power flows in the AC system also more controlconverters between the current EHV grid and the new overlay lable. The control of both systems needs to be integrated even
grid will probably be in the range of 1 3 GW , possibly during the planning stages.
with multiple units placed in parallel. These ratings are not
One of the main research topics [21] is the DC voltage
unrealistic for the near future, and are also of a suitable size control in the DC systems. The DC voltage can somehow
to connect large wind farms or clusters of them to the overlay be compared to what the frequency is for AC systems: As
grid.
soon as a converter fails, the voltage will increase or decrease
When looking at the mesh sizes, it is inevitable that for due to the active power dificit. The main difference, however,
the higher levels of energy transmission also the meshes will is that the dynamics of the voltage control are much faster
become bigger. Mesh sizes depend on the generation and load than those of the frequency control in AC systems. Point-tobalance and the energy intensity in the region and can vary point systems have one converter controlling the DC voltage
considerably in size. At the 400 kV level, the diameter of a at its bus and the other converter controlling the active power.
mesh can vary from some tens up to hundreds of kilometers. In a multi-terminal set-up, the DC voltage control ideally
Within the mesh there is the possibility to have tappings should be spread out over multiple converters, allowing them
to the lower voltage levels. The overlay grid will require a to collectively stabilize the voltage after a fault or converter
similar structure, with mesh sizes that are spanning multiple outage, e.g. through droop control. An outage of a converter
underlaying meshes. Converter stations will have a similar will not only lead to a significant change on the local level, but
function as transformers in the current system. An additional will also influence the power flows throughout both the overlay
control parameter, the power flowing through the converter, DC grid and the underlaying AC system. The cause are active
can be used to control the flows in the overlay and the power injections at different converters which are influenced
underlaying grid. Here as well, the possibility to have tappings when an contingency takes place on the DC system. The
in between nodes of the mesh exists. For smaller countries influence of the operation of the grid on the AC grid should
such as those in the Benelux, the large mesh sizes might result not be overlooked when addressing contingency analyses and
in only a few nodes, if any, of the overlay grid that connect in that way, they can lead to a different grid topology.
to the local grid. The actual layout will clearly depend on
When the overlay grid connects different asynchronous
the expected geration and load interactions, but a conceptual systems, the effect of a single outage might no longer be
scheme is shown in Fig 3.
neglected. For instance, what rating is suitable for a connection
With an additional overlay grid on top of the existing system, to a smaller synchronous zone such as Ireland (7, 8 MW
it is not unlikely that the existing EHV grid will be operated installed capacity) and even Scandinavia (61 GW installed
with some connections opened in normal condition.
capacity) and how many connections should be made to such
OLG

EHV

EHV

EHV

a zone?
The study and control of a combined AC-DC system is
complicated by the fact that the overlay system is controllable
concerning the active power injections and withdrawals. With
multiple converters connected to the same synchronized system, loop-flows have to be avoided by selecting the appropriate
power set-points.

grid operators in the different countries in Europe, starting

from today until 2050. This approach is not feasible for many
different reasons. Such an overlay grid will be built step
by step and will grow naturally as result of new generation
units that are connected to the system and the natural load
evolutions that occur in the meantime. This new generation
and load requirements will result in an increased need for
transfer capacity. The investments are a natural consequence of
local and regional needs, limited by technical, economic and
B. Dependency on the framework
political boundary conditions. Furthermore, new investments
How the overlay grid must be integrated in the existing will alter future investments as a new grid situation arises after
system with several transmission system operators is uncertain. each new investment.
The aforementioned challenges are aggravated by the fact that
The individual transmission investment decisions made in
the interconnected power system in Europe now has several terms of transmission capacity by different investors (different
Transmission System Operators (TSO) responsible for their TSOs as well as potential merchant investors), will finally
own zone. Since the underlay AC transmission system forms determine the technology and topology of the final overlay
one interconnected system and due to the fact that the power grid.
injections by the converters are fully controllable, optimizing
the injections in the different zones controlled by the TSOs A. Effect of investment decisions on the final grid layout
might give rise to sub-optimal flow patterns and even loop
The planning of a new transmission line is a process that
flows between the AC and DC system. Alternatively, the AC
consists of several steps which need to be taken as shown in
system could be split up in different non-synchronized systems,
Fig. 4. Starting from a given power system topology consisting
with the DC overlay grid as a backbone structure, but doing
of a given network layout connecting different loads and generso would comprimise the advantages of an interconnected AC
ation sites, scenarios are set up on how this system will evolve
system.
in the future. These scenarios take changes in generation and
The fact that multiple TSOs operate in different regulatory
load into account, as well as ongoing changes in the network
frameworks in the different countries (or even in parts of
(e.g. aging of components). Based on these scenarios, the need
them) does not only put forward challenges with respect to
for future system changes can be identified, resulting in the
the control of the power system also poses a major challenge
identification of reinforcements (identifying nodes for reintowards the financing of the project and ownership. Different
forcements, power rating and potential technologies) in certain
legislations and procedures in the different countries compliparts of the system. In a next phase, the withheld alternative
cate the matter. The challenges from a legislative point-ofreinforcements are investigated in greater detail and compared.
view are big when compared to the ones encountered when
The second phase will provide more detailed information such
interconnecting or reinforcing the connections between two
as routing and costs involved. Note that during the entire
countries nowadays. A need arises to harmonize the different
process, e.g. for the routing of the transmission line and the
legislations on a EU level for the construction of the infrastrucchoice of the substations to be connected, not only technical
ture, as well as the legislation with regards to the integration of
but also non-technical criteria need to be taken into account.
renewable energy sources and tarrification. A simple example
Based on the resulting set of investments, a decision is made.
is the difference in approach for the connection of offshore
The following phases involve getting regulatory approval and
grids. However, a harmonized regulatory framework alone is
obtaining the necessary permits for the construction. As the
not enough. It needs to be fair to those investing in the system
permitting process can take a significant period of time (10
(correct remuneration for the investments made) and it needs
years is not an exception), the transmission investments should
to remain consistent and credible over a sufficiently long time
be made foresighted. The last phase consists of the actual
period to avoid investment risk due to regulatory uncertainty.
implementation and construction. It is clear that the entire
In the remainder of the paper, the need for a coordinated
planning process is subject to uncertainty, as many scenarios
approach when dealing with different projects is illustrated.
need to be evaluated, and not all of them will be realized (e.g.
Similarly, the development has to be coordinated between the
location and time frame of future generation).
different TSOs. Nowadays, connections to offshore wind farms
An integrated approach for transmission grid planning as
are mainly optimized with respect to one single project in shown in Fig. 5 considers reserve capacities in transmission
terms of voltage and ratings.
assets and for the connection of new generation units in the
future. This way, the potential of renewable energy sources
V. T RANSMISSION SYSTEM PLANNING FOR THE FUTURE
within a specific region can be fully exploited. On the conGRID IN STEPS
trary, overrating of transmission assets to foresee connection
In most overlay grid visions, the grid is determined for the capacities for future generation might not be feasible because
situation in 2030-2050, with the generation and load profiles of delays during the construction process of the generation
estimated in this period [4], [5], [22]. The development of units. Also lack of information about future generation units
an optimal overlay grid for 2050 requires investments, which and uncertainty of generation investments can lead to a nonmust be coordinated between all member states, or the active integrated planning approach for the transmission system.

Fig. 4: Transmission investment process for system reinforcements or transmission expansion

Fig. 6: Wind farms and points of common coupling of study case

TABLE I
A SSUMPTIONS FOR TRANSMISSION SYSTEM OPTIMIZATION

Fig. 5: Different planning approaches for connection offshore wind farms [23]

B. Timed investments illustrated

The following paragraphs show a simple study case to
investigate the difference between the integrated and nonintegrated approach of connecting new generation units. The
study case consists of two offshore wind farms (W F ) with
a rated capacity of 550 MW each and three potential points
of common coupling (P CC) as shown in Fig. 6. W F2 is
connected to the main grid five years after W F1 .
If the integrated planning approach is used, the optimal
transmission system layout can be determined using an extended version of the optimization tool developed by the
authors and presented in [24], [25]. In this case both wind
farms are treated as they are connected to the transmission
system at the same time and the share of the costs caused by
the second wind farm is discounted using the interest rate in
Table I.
While determining the optimal transmission system layout
with the non-integrated approach, first the connection layout
of W F1 can be optimized. Using structure the obtained
transmission system, W F1 can be treated as a new PCC and
the transmission layout for W F2 can be optimized separately.
Also with this approach, the obtained costs for the transmission
investments for W F2 are discounted accordingly. To calculate
the life cycle costs of the transmission system consisting
of investments, maintenance, losses and energy not supplied,
some further assumptions are made which are listed in Table I.
A further assumption is that the onshore transmission is
realized with underground cables (AC or DC). All data related
to costs, electrical parameters and reliability are obtained from

Capacity P CC1,2
Capacity P CC3
V oltage level P CC1,2,3
V oltage level W F1,2
Lif etime
Energy price
Interest rate

200 MW
1200 MW
400 kV
30 kV
25 ae
50 e/M W h
5%

the public domain [24], [25].

The optimal transmission system layout using the integrated
approach is shown in Fig. 7. In this case the transmission
system consists of a multi-terminal HVDC system between
both wind farms and P CC3 . The transmission voltage equals
320 kV and one pair of HVDC cables is used to transmit the
power. In this case the connection between W F1 and P CC3
is able to transmit the total amount of wind power, namely
1100 MW. This means that this cable is overrated, before
W F2 is connected to transmission system. The total costs of
the obtained transmission system are 669, 4 Me and the total
length of the transmission system equals 160 km, whereas the
length of offshore connections equals 109, 4 km.
For the non-integrated approach, first the transmission system layout for the first wind farm is determined, which is
shown in Fig. 8. In this case the optimal transmission system
consists of HVAC connections. The optimal transmission voltage equals 220 kV. As shown in Fig. 8, W F1 is first connected
to P CC2 via two HVAC submarine cables consisting of three
cores. 200 MW of the generated wind power are injected
in P CC2 to the main transmission system, whereas the rest,
namely 350 MW are transmitted to P CC3 via one system of
220 kV single core underground cables. The system costs of
the proposed transmission system equal 375, 3 Me. The length
of the transmission system is 139, 1 km, whereas 85, 5 km of
the transmission cables are situated offshore.
During the determination of the optimal transmission system
layout of the second wind farm, it is assumed the the load level
in the city has been increased with time and that an additional
100 MW can be injected in both P CC1 and P CC2 . Due to the

TABLE II: Break-even time between integrated and not integrated approach in dependence of interest rate
Interest rate
5%
10 %
15 %
20 %
25 %

Break even time

9, 3 ae
4, 5 ae
3 ae
2, 1 ae
1, 7 ae

equals 5, 5 %. Using the obtained system costs, the break-even

time between the integrated and the non-integrated planning
method can be calculated. This way it can be determined for
how many years the construction of the second wind farm may
be delayed, so that the overrating of the HVDC cable from the
integrated solution is still feasible. Table II shows the breakeven time for different interest rates. It can be seen that for
the given study case, with an interest rate of 5 %, the breakFig. 7: Optimal transmission layout for integrated approach
even time equals 9,3 years, which means if the connection
of the second wind farm takes place 9,3 years after the first
wind farm, it would still be cheaper to overrate the HVDC
cable connecting W F1 and P CC3 (see Fig. 7) and foresee
transmission capacities for the second wind farm.
If the integrated solution is used, the transmission system
consists of one single voltage level and transmission technology (320 kV - HVDC). With the non-integrated approach the
first wind farm is connected via 220 kV HVAC cables and the
second wind farm via a combination of 150 kV HVDC and
400 kV HVAC to the main electricity grid. This means that
at the end, three different voltage levels and two technologies
are implemented in the system, which may not be beneficial
for the operation of such a system.
The most important benefit of the integrated approach is
that the onshore transmission system has to be commissioned
only once. With the integrated approach only the offshore
connections for the second wind farm and the extension of the
Fig. 8: Optimal connection of W F1 and W F2 with non-integrated approach
converter station onshore has to be realized after five years.
With the non-integrated approach the onshore connections
of the first and second wind farm will be commissioned
used cross-section of the submarine and underground cables separately as different voltage levels and transmission techfor the connection of the first wind farm, an additional 60 MW nologies are used. In that case also two substations (P CC2
can be injected in the offshore substation of W F1 . Using these and P CC3 ) are affected by the connection of the wind farms.
values as input, the optimal transmission system for the second Summarizing, the total length of the transmission system with
the non-integrated approach is more than twice as long as
wind farm is determined and shown in Fig. 8.
The determined transmission system consists of an HVDC the integrated approach and consists of different technologies
link between W F2 and P CC2 and an HVAC connection be- and voltage levels, which require more complicated permission
tween P CC2 and P CC3 . The optimal transmission voltage of procedures than in the case of the integrated approach.
the HVDC link equals 150 kV and one pair of HVDC cables
VI. C ONCLUSIONS
is used for this link. In this case 100 MW are injected into
P CC2 and the rest of the generated wind power is transmitted
Ambitious goals of integrating renewable energy sources
to P CC3 via one system of 400 kV single core HVAC cables. in Europe will change the European transmission grid signifThe transmission system length is in total 160, 3 km, with icantly. A fundamental upgrade, forming a new overlay grid,
106, 7 km of transmission cables situated offshore. The costs of the transmission system is required, driven by the need to
for the transmission system equal 430, 3 Me.
connect remote, variable energy sources. This paper discusses
If the costs for the transmission system of the second wind how such a grid should be conceived.
Given the difficulties related to AC OHL acceptance and
farm are discounted and added to the costs of the first system,
the obtained total costs equal 708, 3 Me. Comparing the permitting and AC cable limitation, this new system will
system costs to the integrated approach, it can be seen that the probably require HVDC technology. A VSC HVDC multiintegrated approach delivers cost savings of 38, 9 Me, which terminal grid is considered to be the most optimal transmission

technology for the challenges in Europe. Transmission capacities of 48 GW will be needed, an order of magnitude higher
than the current ratings at 400 kV. Similarly to the AC system
of today, the future overlay system will be meshed.
The system should interact with the existing AC system
and provide similar services and reliability. This might require
fundamental changes in the existing AC system as well. As the
DC system will be able to control the injections in the system,
additional care is needed during the planning phases to take
this into account. Also the operation of an overlay HVDC grid
itself will affect the energy flows throughout the grid.
Furthermore, the final topology of the grid will heavily
depend on the framework in which it is installed. Questions
such as Who will invest? and how will be remuneration
for the transmission investments be done? remain unsolved.
A harmonization of the different legislations in the current
international setting is needed.
Finally, the development of a new overlay grid will not
be done in a single investment, but rather happen as a consequence of different successive investment decisions which
each affect the final layout of the grid. This effect is illustrated
with a simple example in this paper. An integrated planning
approach for the design of the overlay can help to decrease the
investment costs in the transmission system on the long term.
Even if the insecurity of construction of new generation units
is high or the generation investment is delayed for a long time,
it can still be beneficial to overrate transmission lines and to
provide reserve transmission capacities for future generation
investments. However, this requires a long term vision on grid
and generation development.

[13] ABB, XLPE Underground Cable Systems Users guide, 2008, rev. 3.
[14] ABB High Voltage Cable Unit, HVDC Light Cables-Submarine and land
power cables, Sweden, 2006.
[15] T. Haileselassie, K. Uhlen, J. Tande, and O. Anaya-Lara, Connection
scheme for north sea offshore wind integration to uk and norway: Power
balancing and transient stability analysis, in PowerTech, 2011 IEEE
Trondheim, Jun. 2011, pp. 1 5.
[16] E. Prieto-Araujo, F. Bianchi, A. Junyent-Ferre, and O. Gomis-Bellmunt,
Methodology for droop control dynamic analysis of multiterminal vschvdc grids for offshore wind farms, Power Delivery, IEEE Transactions
on, vol. 26, no. 4, pp. 2476 2485, Oct. 2011.
[17] J. Hafner and B. Jacobson, Proactive Hybrid HVDC Breakers - A
Key Innovation for Reliable HVDC Grids, in Int. Cigre Symposium,
September, 2011, Bologna-Italy.
[18] D. Van Hertem, M. Ghandari, J. Curis, O. Despouys, and A. Marzin,
Protection Requirements for a Multi-Terminal Meshed DC Grid, Int.
Cigre Symposium, September, 2011, Bologna - Italy.
[19] D. Van Hertem, The use of power flow controlling devices in the
liberalized market, Ph.D. dissertation, K.U.Leuven, Jan. 2009.
[20] P. Pramayon, S. Guerard, G. Aanhaanen, B. Kresimir, P. Cachtpole,
M. Norton, R. Puffer, A. Sorensen, M. Weibel, G. Brennan, J. Rogier,
P. Southwell, and B. Wareing, Increasing Capacity of Overhead Transmission Lines Needs and Solutions, Cigre TB JWG B2/C1.19, 2010,
iSBN: 978-2-85873-113-8.
[21] J. Beerten, D. Van Hertem, and R. Belmans, VSC MTDC systems with
a distributed DC voltage control - a power flow approach, in PowerTech,
2011 IEEE Trondheim, Jun. 2011, pp. 1 6.
[22] Eurelectric, Power Choices - Pathways to Carbon-Neutral Electricity
in Europe by 2050, 2011, www.eurelectric.org/powerchoices2050.
[23] National Grid, Offshore Development Information Statement, Technical Report, Sep. 2010.
[24] H. Ergun, D. Van Hertem, and R. Belmans, Multilevel Optimization
for Offshore Grid Planning, in Int. Cigre Symposium, September, 2011,
Bologna-Italy.
[25] , Transmission System Topology Optimization for Large Scale
Offshore Wind Integration, Submitted to: IEEE Transactions on Sustainable Energy, 2011.

R EFERENCES

Hakan Ergun (S10) received his degree in electrical engineering (Dipl.-Ing.)

at the Technical University Graz in 2009. Since 2010 is he research assistant
at the KU Leuven and persuing a PhD degree. His research interests are the
impact of integration of renewable energy sources on the electricity grid and
the development of the transmission grid for large scale renewable integration.

[1] European Commission. (2011, November) Energy infrastructure

priorities for 2020 and beyond A Blueprint for an integrated
European energy network. Brussels, Belgium. [Online]. Available:
http://ec.europa.eu/energy/infrastructure/strategy/2020 en.htm
[2] European Network of Transmission System Operators for Electricity
(ENTSO-E), ENTSO-E position paper on a framework regarding
electricity highways, ENTSO-E, Tech. Rep., Dec. 2011.
[3] , Ten-year network development plan 2010-2020, ENTSO-E,
Tech. Rep., Jun. 2010.
[4] Greenpeace and EREC, [r]enewables 24/7, Infrastructure Needed To
Save The Climate, Tech. Rep., November 2009.
[5] European Climate Foundation (ECF), Roadmap 2050, A Practical
Guide To A Prosperous, Low-Carbon Europe, Tech. Rep., 2010.
[Online]. Available: http://www.roadmap2050.eu
[6] D. Van Hertem and M. Ghandhari, Multi-terminal VSC HVDC for
the European supergrid: Obstacles, Renewable and Sustainable Energy
Reviews, vol. 14, no. 9, pp. 3156 3163, 2010.
[7] European Network of Transmission System Operators for Electricity
(ENTSO-E), Offshore transmission technology, ENTSO-E, Tech. Rep.,
Nov. 2011.
[8] Bundesnetzagentur, Auswirkungen des Kernkraft-Moratoriums auf die

Ubertragngsnetze
und die Versorgungssicherheit, Tech. Rep., May,
2011.
[9] German Energy Agency, Dena Grid Study II - Integration of Renewable
Energy Sources in the German Power Supply System from 2015 2020
with an Outlook to 2025, Nov. 2010.
[10] J. De Decker and P. Kreutzkamp, Offshore Electricity Grid Infrastructure in Europe - Final Report, Tech. Rep., Oct. 2011.
[11] European Comission, DG Energy in collaboration DC Climate Action
DG and DG Mobility and Transport, EU Energy Trends to 2030, Tech.
Rep., 2009.
[12] European Wind Energy Association. (2010) Factsheet: Statistics
& Targets. [Online]. Available: http://www.ewea.org/fileadmin/ewea
documents/documents/publications/factsheets/Factsheets.pdf

Jef Beerten (S07) was born in Belgium in 1985. He received the M.Sc.
degree in electrical engineering from the Katholieke Universiteit Leuven (KU
Leuven), Leuven, Belgium, in 2008, where he is currently working towards
the Ph.D. degree. He is a Research Assistant with the division ESAT-ELECTA
division of KU Leuven. His research interests include power system control,
the grid of the future and multiterminal VSC HVDC in particular. Mr. Beerten
is an active member of both the IEEE and Cigre.
Mr. Beerten holds a Ph.D. fellowship from the Research Foundation
Flanders (FWO).

Dirk Van Hertem (S02, SM09) graduated as a M.Eng. in 2001 from the
KHK, Geel, Belgium and as a M.Sc. in Electrical Engineering from the
KU Leuven, Belgium in 2003. In 2009, he has obtained his PhD, also from
the KU Leuven. In 2010, Dirk Van Hertem was a member of EPS group at
the Royal Institute of Technology, in Stockholm, Sweden. Dirk Van Hertem is
currently an assistant professor at the KU Leuven and working in the ELECTA
division. His special fields of interest are power system operation and control
in systems with FACTS and HVDC and building the transmission system of
the future, including offshore grids and the supergrid concept. He is an active
member of both IEEE (PES and IAS) and Cigre.