AckermannTh. (ed) Wind Power in Power Systems
Подождите немного. Документ загружается.
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 11_CHA10 .3D – 231 – [197–232/36]
20.12.2004 7:41PM
Denmark (i.e. 0.3 and 0.5 million tonnes CO
2
for the low-price and the high-price
scenarios, respectively). This is primarily because of the fact that the Eltra area has
already experienced a significant expansion of wind power and that the marginal
reduction of the system emissions arising from additional expansions would be modest.
Additional wind power production would, however, reduce nondomestic CO
2
emis-
sions by 1–3 million tonnes, depending on the power price and the CO
2
emissions
resulting from the production capacity replaced abroad.
Figure 10.26 gives high and low figures for CO
2
reductions abroad. The low figures
are equivalent to replacing natural gas-based power production with an efficiency of
45 % (specific CO
2
emission: 0.456 t/MWh); the high figures are equivalent to a coal-
based power production with an efficiency of 40 % (specific CO
2
emission: 0.855 t/
MWh). For the low-price scenario, for instance, this would mean a reduction of 1.6
million tonnes and ¼ 3:0(1:6 þ 1:4) million ton nes, respectively, in nondomestic CO
2
emissions.
In an international context, it will only be attractive to furt her expand wind power
production in the Eltra area as a means of reducing CO
2
emissions if there is a price on
CO
2
and if the CO
2
is traded according to common and harmonised rules.
Another aspect is that further massive wind power expansion in the Eltra area can be
realised only if there is additional access to operational reserves and if the transmission
network is considerably reinforced. Considerable investments in air coolers, heat sto-
rage tanks, immersion heaters and/or heat pumps to decouple the power production
from the CHP production should be added to this.
10.4 Conclusions and Lessons Learned
Neighbouring TSOs traditionally offer mutual assistance based on equality and reci-
procity. Today, neighbouring assistance implies that the TSOs give each other access to
utilise their own ‘reserve power market’, and the service providers are paid.
There will be further stress on storage and regulation capabilities.
In the future, the real-time market will have a major impac t on system operation and
prices, and this market will fluctuate considerably. In fact, negative energy prices in this
market are already a reality. With this in mind, the following is a wish list for a future
electricity market:
.
improved wind forecasts;
.
enhanced control possibilities for production and load;
.
increased interconnection capacity with neighbou ring areas;
.
enhanced control and resources of Mvar;
.
an enhanced regulating power market.
The continuing increase of nondispatchable production and the very dispersed locations
on lower voltage levels require the development of a new control philosophy. Studies
made in the 1980s regarding the limits for the integration of wind energy were far off the
mark in comparison with today’s implementation, but two factors have also changed:
(1) a pool was the approach chosen for implementing the internal electricity market and
Wind Power in Power Systems 231
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 11_CHA10 .3D – 232 – [197–232/36]
20.12.2004 7:41PM
(2) start of the common exchange of regulating power, which was not effective in the
1980s.
We still have to find out what the lim its of the system are regarding operation with
ever increasing amounts of wind energy. New control systems, flexible load and add-
itional regulating capacity will be some of the new intervention measures used.
References
[1] DEFU/DTU (Danish Utilities Research Association – Technical University of Denmark) (1998)
‘Connection of Wind Turbines to Low and Medium Voltage Networks’, committee report 111-E, avail-
able at http://www.defu.dk.
[2] Degn, P. C. (2000) ‘Specifications for Connecting Wind Farms to the Transmission Network’, Eltra
document 74557; available at http://www.eltra.dk.
[3] EC (European Commission) (1986) ‘Wind Penetration in the Elsam Area (Western Denmark)’.
[4] Eltra (2003) ‘A Brief Description of Eltra’s Market Model MARS’, presented at Eltra Seminar on Market
Power, Eltra document 156884 v1, available at http://www.eltra.dk.
[5] Eriksen, P. B. (1993) ‘A Method for Economically Optimal Reduction of SO
2
/NO
x
Emissions from Power
Stations’, in Proceedings of the IEA/UNIPEDE Conference on Thermal Power Generation and Environ-
ment, Hamburg, September 1–3.
[6] Eriksen, P. B. (2001) ‘Economic and Environmental Dispatch of Power/CHP Production Systems’,
Electric Power Systems Research 57(1) 33–39.
[7] Granli, T., Gjerde, O., Lindstro
¨
m, K., Birck Pedersen, F., Pinzo
´
n, T., Wibroe, F. (2002) ‘The Transit
Solution in the Nordic Electric Power System’, feature article, in Nordel Annual Report 2001, Nordel,
http://www.nordel.org, pp. 37–42.
[8] Kristoffersen, B. B. (2003) ‘Impacts of Large-scale Wind Power on the Power Market’, in Proceedings of
the Fourth International Workshops on Large-scale Integration of Wind Power and Transmission Networks
for Offshore Wind Farms (Session 1), October 2003, Billund, Denmark, Royal Institute of Technology,
Electric Power Systems, Stockholm, Sweden.
[9] Nielsen, T. S. (1999) ‘Using Meteorological Forecasts in On-line Predictions of Wind Power’, Department
of Informatics and Mathematical Modelling, Technical University of Denmark.
[10] Nordel (1992) ‘Nordel Network Planning Criteria’, Nordel, http://www.nordel.org.
[11] Nordel (2002) ‘Nordic System Operation Agreement’, available at http://www.nordel.org.
[12] Pedersen, J. (1990) ‘SIVAEL – Simulation Program for Combined Heat and Power Production’, in
Proceedings of the International Conference on Application of Power Production Simulation, Washington.
[13] Pedersen, J. (2003) ‘Simulation Model Including Stochastic Behaviour of Wind’, in Proceedings of the
Fourth International Workshops on Large-scale Integration of Wind Power and Transmission Networks for
Offshore Wind Farms (Session 3a), October 2003, Billund, Denmark, Royal Institute of Technology,
Electric Power Systems, Stockholm, Sweden.
[14] Ravn, H. (1992) ‘Optimal Scheduling of Combined Power and Heat’, IMSOR internal publication,
Technical University of Denmark.
[15] Rønningsbakk, K., Gjerde, O., Lindstro
¨
m, K., Birck Pedersen, F., Simo
´
n, C., Sletten, T. (2001) ‘Con-
gestion Management in the Electric Power System’, Nordel Annual Report 2000, feature article in Nordel
Annual Report 2000, Nordel, http://www.nordel.org, pp. 35–44.
[16] Tonjer, A., Eriksen, E., Gjerde, O., Hilger, C, Lindstro
¨
m, K., Persson, S., Randen, H., Sletten, T. (2000)
‘A Free Electricity Market’, feature article in Nordel Annual Report 1999, Nordel, http://www.nordel.org,
pp. 35–40.
232 The Danish Power System
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 233 – [233–256/24]
20.12.2004 7:42PM
11
Wind Power in the German
Power System: Current Status
and Future Challenges of
Maintaining Quality of Supply
Matthias Luther, Uwe Radtke and Wilhelm R. Winter
11.1 Introduction
Political support and, in particular, the fact that the Renewable Energy Sour ces Act
(EEG) in Germany came into force in April 2000 have led to a dramatic increase in the
supply of power generated from wind energy. The present situation and the foreseeable
offshore developments in the North Sea and Baltic Sea require grid operators to take
supplementary measures to meet quality-of-supply requirements in the future. Power
supply quality includes quality of voltage and network frequency as well as the inter-
ruption frequency and duration, both regarding the power system and customer supply.
The integration of renewable energy sources, especially wind energy generation, changes
the power system conditions, which does not necessarily have to lead to negative
changes in the supply quality.
This chapter summarises the experience with wind power integration in Germany. We
will especially focus on the E.ON Netz control area, with an installed wind energy
capacity of 5500 MW (as of April 2003). The chapter considers various aspects of future
network development and operation wi th particular reference to system performance on
the grid. We will include information on steady-state performance. Additionally, the
results of dynamic studies and the resulting requirements regarding wind turbines are
presented. It has to be mentioned, that this has been done in close cooperation with
Wind Power in Power Systems Edited by T. Ackermann
Ó 2005 John Wiley & Sons, Ltd ISBN: 0-470-85508-8 (HB)
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 234 – [233–256/24]
20.12.2004 7:42PM
manufacturers of wind turbines
(1)
. Finally, further issues regarding the integration of
wind energy into the power system will be discussed.
11.2 Current Performance of Wind Energy in Germany
A look at the wind energy statistics reveals the extent to which this renewable energy
source is used in Germany and the dynamic nature of its development. In December
2003, around 14 000 MW were installed in a total of 15 694 wind turbines.
Between April 2002 and December 2003, in a lapse of only 20 months, 3984 additional
wind turbines with a total capacity of about 4686 MW were installed and connected to
medium-voltage (MV), high-voltage (HV) an d extra-high-voltage (EHV) systems. By
April 2003, there was a 32 % increase in installed capacity as compared with April 2002
(see Table 11.1). The rate of increase in installed capacity between April 2002 and April
2001 was 44 %.
In the future, there will be no further increase in network load in Germany or in the
E.ON Netz area. For the years 2003 and 2004, the rate of increase in newly install ed
capacity is expected to be reduced to approximately 1500 MW per year. This will be the
result of the saturation due to the limited number of areas designa ted for the use of wind
energy.
Figure 11.1 shows a recent forecast of the power supply from wind turbines in
Germany up to 2030. The main source of growth here is offs hore power generation.
At the moment, it is uncertain which projects will ultimately be implemented, and it is
also unclear what effect re-powering will have (i.e. the replacement of existing wind
turbines by more powerful onshore units). Hence, the planning certainty regarding the
future development of the grid is rather low. Several offshore projects are planned in
areas with an appropriate water depth in the North Sea and Baltic Sea. Figures 11.2 and
11.3 provide an overview of the projects that are currently under discussion.
The increase in onshore wind power combined with foreseeable offshore devel-
opments in the North Sea and Baltic Sea area require further action by the
transmission system operator (TSO) in order to guarantee the necessary quality of
Table 11.1 Development of wind energy use in Germany
Connected wind turbines Installed capacity (MW)
December 2003 15 694 13 955
April 2003 13 746 12 223
April 2002 11 710 9 269
April 2001 9 525 6 435
(1)
The authors would like to thank the wind turbine manufacturers organised in the Fo
¨
rdergemeinschaft fu
¨
r
Windenergie for fruitful discussion and for their contribution.
234 Wind Power in Germany
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 235 – [233–256/24]
20.12.2004 7:42PM
0
5 000
10 000
15
000
20
000
25 000
30
000
35 000
40 000
45 000
50 000
1990 1995 2000 2005 2010 2015 2020 2025
2030
Year
Installed wind energy capacity (MW)
Onshore and Offshore
Onshore
Figure 11.1 Forecast of power supply from wind energy in Germany (www.dewi.com) (Repro-
duced by permission of DEWI, Germany.)
Figure 11.2 Overview of planned offshore wind farms in North and Baltic Sea (Reproduced by
permission of DEWI, Germany.)
Wind Power in Power Systems 235
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 236 – [233–256/24]
20.12.2004 7:42PM
supply. The German experience is likely to be helpful for further developments in
the Union for the Coordination of Transmission of Electricity (UCTE; see also
http://www.ucte.org).
11.3 Wind Power Supply in the E.ON Netz Area
The transmission system in Germany is presently controlled by four TSOs. Figure 11.4
shows the areas of the individual TSOs as well as their share of installed wind power
generation.
In April 2003, the power supply from wind turbines installed in the E.ON Netz area
was about 5500 MW, which represents nearly half of the total installed capacity in
Germany. Most of the wind turbines are connected to the MV system in the coastal
areas of the North Sea and Baltic Sea (i.e. to the systems that are operated by regional
distribution utilities). As the specific capacity of the current wind turbines increases and
the turbine operators form groups, there is a growing need for large wind farms to be
connected to the HV and EHV system.
Figure 11.5 shows the difference between the wind power fed into the netwo rk and the
network load during a week of maximum load in the E.ON Netz control area. It is a
typical snapshot of the relationship between consumption and supply in this control
area. In the Northern part of the power system, the transmission capacity of the HV
system reaches its limits during times with low load and high wind power production.
Network bottlenecks occur because of the regional generation surplus during low
electricity consumption on peak wind power production periods.
Adlergrund
Arkona-
Becken
Pommersche
Bucht`
Kriegers
Flak
Fehmarn
westl. Ostsee
Sky 2000
DK
Bentwisch
Lumbmin
Usedom
Rügen
PL
150 MW
150 MW
350 MW
850 MW
750 MW
1.000 MW
size:
total:
MW
MW
(150–1.000)
approx. 3.200
Schleswig-Holstein
Mecklenburg-vorpommern
Hamburg
Figure 11.3 Planned offshore wind farms in North and Baltic Sea (Reproduced by permission of
DEWI, Germany.)
236 Wind Power in Germany
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 237 – [233–256/24]
20.12.2004 7:42PM
11.4 Electricity System Control Requirements
As the meteorological situation can be forecast only to a very limited extent, wind power
plants feed in a stochastic form of generation. One major factor of uncertainty here is
how precise the forecasts are regarding the exact time of a certain weather condition.
Total of installed wind energy of 12 223 MW :
E.ON Netz: 48 %
Vattenfall Europe Transmission: 37 %
RWE Net: 14 %
EnBW Transportnetze: 1 %
Figure 11.4 Distribution of installed wind capacity in Germany as of April 2003
Wind power supply
Wind power supply (MW)
4 500
4 000
3 500
3 000
2 500
2 000
1
500
1
000
500
0
18
000
16 000
14
000
12 000
10 000
8 000
6
000
4
000
2 000
0
30
17
Total load (MW)
Date (November)
Total load
Figure 11.5 Network load and aggregated wind power generation during a week of maximum
load in the E.ON Netz control area
Wind Power in Power Systems 237
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 238 – [233–256/24]
20.12.2004 7:42PM
In the E.ON Netz control area, in high winds the WTs were recorded to feed in a total of
about 4800 MW, with fluctuations amounting to 500 MW within a 15-minute period.
Occasionally, the controlling power range that is needed during such conditions already
exceeds the capacity available in the control area and therefore has to be provided from
outside. If this is the case, the controlling power has to be imported from other TSOs via
the interconnection lines. For this, the available transfer capacity of the interconnection
lines has to be taken in account.
The planned increase in wind power will lead to an increased need for controlling
power. Figure 11.6 (Haubrich, 2003) shows the results of a study on how E.ON
Netz might be affected by that. The controlling power is required to substitute the
positive or negative deviation between actual wind power supply and planned wind
power generation; in other words, it has to compensate for the wind forecast
failure. The amount of wind power on the grid increases and will continue to
require sophisticated congestion management in order to ensure system security,
quality of supply and nondiscriminatory access to the grid. Congestion management
is carried out by changing control states and, increasingly, by re-dispatching, that
is, reducing or relocating generation if limits are reached during a sudden increase
in wind power supply, for instance. The TSO has to switch from a daily bottleneck
forecast to online cycles in order to match the stochastic phenomenon of wind
power generation.
11.5 Network Planning and Connection Requirements
Given the growth rates mentioned above, the priority acceptance of energy from renew-
able power generation facilities specified in the Renewable Energy Sources (RES) Act
will face technical limits. These limits relate to the different grid criteria that apply at the
18 000
14 000
10 000
6 000
2 000
2001 2006 2011 2016
4 200
2 300
8 000
10 000
Installed wind power
Year
Power (MW)
Controlling power
range required
Figure 11.6 Forecast wind power development and controlling power range required (Haubrich,
2003). Reproduced by permission of RWTH, Germany
238 Wind Power in Germany
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 239 – [233–256/24]
20.12.2004 7:42PM
regional level and to the voltage level concerned. For the synchronously connected
overall system, the following criteria are essential for secure and reliable operation:
.
thermal overloading;
.
voltage stability;
.
frequency stability.
Thermal overloa ding occurs when the dispatch of the supply that is fed in exceeds the
regional transmission capacity of the system. This can occur, for instance, during times
with high wind speeds and low loads on the grids in the coastal regions. The local wind
turbines feeding into the MV network then exceed the regional load many times over,
which leads to feedbacks and violations of voltage standards in the HV system.
In the medium and long-term, it will be necessary to upgrade the grid of the HV and
EHV system in order to remove bottlenecks. This is very well illustrated by the situation
in the Schleswig–Holstein region, for which E.ON Netz has worked out an extensive
110 kV upgrading strategy. In view of the current licensing situation, however, it
appears doubtful whether grid upg rading will be able to keep pace with the development
of wind power. The introduction of generation managem ent co uld provide temporary
relief, however. Generation management is a method of avoiding violations of transmis-
sion system standards (e.g. overloading of equipment) by reducing the wind power that
is fed in by an adequate portion. This generation management will be limited to the
period until additional or new equipment has been installed to allow higher amounts of
wind power transfer.
The applications for the necessary grid upgrading schemes have already been sub-
mitted. These schemes will be considerably extended, though, by the development of
offshore wind power . Figure 11.7 provides an indication of possible developments
18 000
14 000
10 000
2
000
6 000
Cost
2001 2006 2011 2016
200
400
600
800
1000
1000
700
300
110 million 400 million 550 million
Installed wind capacity
Additional route kilometres
km
Installed capacity (MW)
Year
Figure 11.7 Wind power development and grid upgrading (Haubrich, 2003): installed wind
capacity and accumulated additional route kilometres, 110 kV and 380 kV (Reproduced by
permission of RWTH, Germany.)
Wind Power in Power Systems 239
//INTEGRAS/KCG/P AGIN ATION/ WILEY /WPS /FINALS_1 4-12- 04/0470855088_ 12_CHA11 .3D – 240 – [233–256/24]
20.12.2004 7:42PM
within the E.ON Netz control area. Regarding the expected offshore development,
additional EHV overhead lines will have to be built in order to transport the power
that is generated offshore to the places with high consumption density (e.g. the Ruhr
area or the region of Frankfurt-an-Main). Also, in the coastal regions of Germany, the
HV networks need to be extended in order to take in the power from the onshore wind
farms. Figure 11.7 shows the total additional line kilometres in relation to the increased
installed wind generat ion capacity (Haubrich, 2003).
Modern power systems have evolved into very large systems, stretching out over
hundreds and thousands of kilometres. These synchronously interconnected systems must
operate stably also for disturbancies. So called interarea oscillations in the low frequency
range can affect the complete system. Furthermore, the unbundling of generation, trans-
mission and supply as well as large amounts of wind power fed into the system do not
really correspond to the physical nature of synchronously interconnected power systems,
which cover a large area where different subnetworks and power plants interact. However,
in this new environment, with a possible higher loading of the transmission system, TSOs
may be forced to operate the system closer to its stability limits.
Therefore, there are detailed studies regarding the steady-state stability of the UCTE
power system. Their goal is to analyse how the reliability of system operation can be
maintained given a future environment with further system extensions and an open
market (Bachmann et al., 1999, 2002; Breulmann et al., 2000a, 2000b). As a consequence,
voltage stability as well as frequency stability and small signal stability are of increasing
importance to large interconnected power systems. The frequency of the observed inter-
area oscillations in the UCTE system lies in the range of 0.22 to 0.26 Hz, and in most cases
the damping is sufficient. In some cases, however, there was only poor damping, which
has to be considered in relation to reliable system operation. These oscillations in the low
frequency range must not excite subsynchronous resonances on the wind turbine shaft, on
the one hand. On the other hand, the turbine generator unit control must not excite power
oscillations, especially in the low frequency range.
Large-scale wind generation will also have a significant affect on system stability and
on frequency and damping of interarea oscillations. Unless appropriate measures are
taken, the increasing connection of large-scale wind farms and offshore capacities may
lead to additional limiting criteria, because of deficits in reactive power supply or
disturbances in dynamic system performance. This may result in critical situations in
the overall syst em, including cascading events and finally the loss of stability. E.ON
Netz was the first TSO in Germany to implement additional requirements for the
dynamic behaviour of wind turbines in 2003. This was necessary to guarantee system
stability in spite of changes in the generation of electrical energy (Lerch et al., 1999).
The results of the investigations show that if the HV and EHV systems are to continue
to operate reliably, additional demands will have to be made regarding the connection
of wind turbines to the system. These refer especially to providing the required support
(Lerch et al., 1999) and invo lve:
.
setting the frequency reduction protection at 47.5 Hz to provide grid support in the
event of major disturbances;
.
additional requirements of active and reactive power control of wind turbines on the
basis of a case-by-case review;
240 Wind Power in Germany