Note: Descriptions are shown in the official language in which they were submitted.
CA 02491653 2008-06-19
- 1 -
Method for operation of a power generation plant
Technical Field
The present invention relates to a method for operation
of a power generation plant.
Prior Art
An air storage turbine has been disclosed in
DE 28 22 575. At times when the power consumption is
low, for example at night and at weekends, a compressor
acting as a power consuming machine uses electricity that is
generated by basic load power stations to pump air into a
storage volume, for example an underground cavern in a salt
mine. The cavern is charged, for example, to 100 bar. At
times when the electricity demand is high or if another
power station fails, the stored air is used to drive an air
turbine or a combined air/gas turbine, which generates
electrical power via a generator. This significantly
lengthens the operating life of basic load power stations
and, in liberalized electricity markets, the peak power that
can be generated in this way allows a considerable financial
saving to be achieved. Furthermore, in liberalized
electricity markets, the process of covering transient power
demands, such as those which occur when a major load is
switched on but especially when a major power station block
fails, is very highly lucrative. Even the pure provision of
appropriate capacities can save a very large amount of
money. When power generating plant fail, the capability for
frequency support is in demand. During the first fractions
of a second after the failure of an electricity supply, the
grid frequency can be kept within the permissible tolerance
without any further problems in a large grid by virtue of
the rotating
CA 02491653 2005-01-04
- 2 -
masses of the remaining power stations. However, after
this, power reserves must be available immediately in
order to prevent the frequency decreasing, and thus
failure of the entire grid. Steam power stations which
are operated slightly throttled back can provide power
amounting to the order of magnitude of around 5% of
their maximum power very quickly; however, they require
several tens of minutes for power increases beyond
this, for example up to 30% of their maximum power.
When a major load is connected to the grid, load ramps
are demanded from the power stations, in which the
provision of a considerable amount of additional power
is demanded in the region of seconds or up to ten
minutes. Gas turbine plant and combination plant allow
such increases to be coped with within minutes. An air
turbine or combined air/gas turbine in a storage power
station of the cited type has a comparable reaction. It
is also known from operating experience that rapid load
changes such as these cause severe temperature
gradients and, associated with them, damaging thermal
alternating loads and mechanical stresses particularly
in the hot gas path (which is already thermally highly
loaded in any case) of gas turbine groups or in the
steam generators for steam and combination power
stations. Geodetic hydroelectric power stations are
admittedly able to mobilize significant power reserves
within seconds; however, their availability is, of
course, restricted.
Description of the invention
The object of the present invention is thus to specify
a method of the type mentioned initially which avoids
the disadvantages of the prior art. The invention is
based in particular on the object of specifying a
capability to react quickly to transient load demands
on a power generation plant, both in terms of frequency
CA 02491653 2008-06-19
3 -
support and, very particularly, with respect to rapid
load ramps.
According to a broad aspect of the present
invention, this object is achieved by a method for
operation of a power generation plant in an electricity
network. The power generation plant comprises at least one
storage volume. It also comprises at least one power source
which can be operated with energy storage fluid which is
stored in the storage volume and which, during operation,
emits power via a generator which is connected to the power
source. It also comprises at least one process machine for
feeding energy storage fluid into the storage volume and
which consumes power during operation via a motor which is
connected to the process machine. The power output of the
power generation plant is the difference between the power
emitted by the power source and the power consumption of the
motor. The method is characterized in that, for a first
power demand, the power generation plant is operated in a
first operating mode, in which the power source is operated
at a first power output, and the process machine is operated
with a first power consumption. In the event of any changes
in the power demand to a second power demand, the net power
output of the power generation plant is in a first step
matched to the second power demand by changing the power
consumption of the process machine.
Against the background of a power generation plant
which comprises a power-consuming power consuming
machine and a power-emitting power generation machine,
the essence of the invention is thus to react to
transients in the power demand by means of appropriate
adaptation of the power which is consumed in the power
consuming machine. In practice, it has been found that,
by controlling or even shutting down the power
consuming machine, it is possible to achieve at least
CA 02491653 2008-06-19
- 3a -
around one order of magnitude greater load gradients
than by means of a control action on the power
generation machine. In one preferred method variant,
the power which is emitted from the power generation
machine is, in a first step, kept constant. This
furthermore has the advantage that the power transient
does not per se act on a thermally highly loaded power
generating structure but on a considerably less highly
loaded power-consuming structure. Air storage plant are
particularly suitable for this purpose since by virtue
of their nature they have, for example, separately
arranged turbines and compressors as well as a store in
which compressed fluid is temporarily stored for
driving the power generation machine and is available
even when the power consuming machine is at rest or is
operating at a reduced power level. The power generation
plant is in this case in a specific initial operating
state, in which the power consuming machine, or possibly
and preferably also two or more power consuming
machines, are feeding a specific mass flow of an energy
storage fluid, for example compressed air, to the
storage volume for the power generation plant while,
possibly and at the same time, a second mass
CA 02491653 2005-01-04
- 4 -
flow of the energy storage fluid is expanded via the
power generation machine; heat is advantageously
supplied to the energy storage fluid before and/or
during the expansion process. In this case, the power
consuming machine absorbs power in particular via an
electric drive motor, and the power generation machine
emits power via an electrical generator. The net power
output of the power generation plant is defined as the
difference between the power emitted by the generator
and the power consumed by the motor. In a preferred
initial operating state, the power generation plant is
operated in a steady state in such a way that the mass
flow which is fed into the storage volume from the
power consuming machine corresponds to the mass flow
flowing through the power generation machine. This
corresponds to a known gas turbine process in which,
however, the turbine and compressor are not rigidly
coupled. Depending on the grid load and the electricity
price to be achieved or to be paid for at any given
time, the initial operating state may, however, also be
a discharge mode in which the mass flowing out of the
store is greater than the mass flow that is fed by the
power consuming machine, or a charging mode, in which
the store is charged. In this case, the power
generation machine may in fact also be stationary in
the initial operating state, which corresponds to a
considerably negative net power output.
An initial operating state in which the maximum
capability to suddenly increase the power for frequency
support is available is an operating state in which all
the power consuming machines in the power generation
plant are being operated at maximum power. The entire
power consumption of the power consuming machines - as
will also be described in the following text quite
possibly a multiple of the rating during equilibrium
operation of the power generation plant - can in
principle be made available to the grid simply by
CA 02491653 2005-01-04
opening a switch. In a second step, although
considerably more slowly, the power of the power
generation machines can be increased, provided that
they are not being operated at maximum power in the
5 initial operating state. To this extent, it appears to
be desirable to make use of an initial operating state
in which the power consuming machines are running with
their full power consumption, while the power
generation machines are stationary or are running on no
load. Admittedly, when seen in absolute terms, an
initial operating state such as this results in the
greatest potential to increase the power. However, the
power component to be produced by the power generation
machines is available only with a delay since power
generation machines which are operated on no load - or
to be more precise their generators - must first of all
be synchronized to the grid. In the interest of the
maximum power dynamic response, it has therefore been
found to be advantageous to keep the power generation
machines already synchronized to the grid with a small
amount of power being emitted to the grid. In one very
particularly preferred operating method, all of the
power consuming machines are thus operated at at least
80% of their maximum power consumption. At the same
time, all of the power generation machines are
synchronized to the grid and are operated with as low a
power output as possible, preferably of less than 10%
or less than 20% of their maximum power output;
however, operational reasons may also demand a higher
minimum power. Starting from this initial operating
state, it is possible when a rapid power demand occurs
to shut down the power consuming machines by opening
switches and at the same time to pass a command to
increase the power to the power generation machines.
The power which was previously consumed by the power
consuming machines is then instantaneously available to
the grid, and the power from the power generation
machines is made available with a delay time that is
CA 02491653 2005-01-04
- 6 -
intrinsic to the system and, in particular, with a
power gradient that has an upper limit, but without
having to previously wait for synchronization. The
power dynamic response is thus maximized.
The essential feature of the invention is that the
power consuming machine is consuming power in the
initial operating state. This is because the actual
basis idea of the invention may be regarded as being to
use the power consuming machine which is feeding a
store to create a secondary power demand, which can be
-r 1-,õ
sLIU down as required, iii auuii.ivII to 4-1-11.. actual power
consumers in an electricity grid, in the form of a
bias, and to increase the net power output from a power
generation plant virtually instantaneously when
necessary by reducing or shutting down this secondary
power consumption. The power output from the power
generation machine may in this case be kept constant,
at least in a first step.
The extremely wide load control range of a storage
plant which is operated according to the invention is
significant. This is because, based on the rule of
thumb that, in the case of a gas turbine, around two
thirds of the gross turbine power is consumed in the
compressor, it can easily be seen that, on the basis of
a plant which is being operated in the steady state,
that is to say at equilibrium, 200% of the
instantaneous net power output is instantly available
just by shutting down the compressors! The entire load
control range of the plant - assuming that the
compressor design is based on steady-state operation in
equilibrium with the power generation machine as 100% -
can then roughly be estimated to have a net power
output from -200% to +300% of the rating that is
available when operating in equilibrium. This range can
also be widened even further by designing the
compressor to be correspondingly larger, in which case
CA 02491653 2005-01-04
- 7 -
a turbocompressor, for example, can be operated very
efficiently on partial load by rotation speed
regulation - in fact, the compressor need not be
operated in synchronism with the grid.
In one embodiment of the invention, the power
consumption of the power consuming machine is
controlled such that, when changes occur in the power
demand in an electricity grid in which the power
generation plant is operated, the sum of the power
consumption of the power consuming machine and the
power demand of the grid remains constant. The
constancy of this maximum power output is preferably
regulated to less than 5% of the maximum power output
of the power generation machine.
It is self-evident, for example when a very large block
is disconnected from the grid or a large load is
connected to it, that it is also necessary to take
account of situations in which the additional power
demand can no longer be satisfied purely by controlling
the power of the power consuming machines; in
situations such as these, the power consuming machines
are completely disconnected from the grid, and the
power generation machine is operated in conjunction
with other power stations that are integrated in the
grid, in a manner that is known per se, with the
maximum power gradient. Assuming that the storage
volume has been charged sufficiently, it is possible to
react very quickly and flexibly even in this situation.
The greatest flexibility and the fastest reaction can
in general be achieved when a storage power station
comprises two or more compressors which act on a
storage volume and can be controlled individually, as
well as two or more individually controllable turbines
which are fed from this storage volume. One embodiment
of a storage power station comprises four compressor
systems each having two turbocompressors and
intercoolers whose rotation speeds are controlled, as
CA 02491653 2005-01-04
- 8 -
well as two turbine sets, which each act on one
generator; the storage power station is operated with
storage pressures in the region of 30 bar, and
preferably at least 50 bar up to 100 bar.
Thus, when the power demand on the power generation
plant is increased, the power consumption of the power
consuming machine is reduced in a first step, while the
power output of the power generation machine remains
constant, in order to increase the net power output and
at the same time to satisfy the power demand at that
time. In a second step, the power output of the power
generation machine is then increased and the power
consumption of the power consuming machine is reduced,
with the net power output being kept the same as the
power demand at that time. In this case, the power
demand on the power generation machine and the power
reduction for the power consuming machine can take
place at the same time without any problems, since the
power increase occurs with a slight delay in any case,
owing to factors that are intrinsic to the system.
When the power demand on the power generation plant is
reduced, the power consumption of the power consuming
machine is increased, in a first step, while the power
output of the power generation machine remains
constant, in order to reduce the net power output and
at the same time to satisfy the power demand at that
time. In a second step, the power output of the power
generation machine is then reduced, and the power
consumption of the power consuming machine is
increased, with the net power output being kept equal
to the power demand at that time.
For long-term operation of the plant, this results in
the following possible scenario: In an initial step,
the storage volume of the power generation plant is
charged in the storage mode, for example at night. In
CA 02491653 2005-01-04
- 9 -
the process, the storage volume is not filled to the
maximum level but is charged only until it is
approximately half full - in this case, values of
between 25% and 75% of the pressure range are very
highly acceptable. These numerical figures should be
understood as meaning that the pressure range is
defined as a minimum permissible storage pressure for
operation of the power generation plant which is
defined as 0%, and a maximum permissible storage
pressure, which is defined as 100% of the pressure
range. In this case, the store filling is proportional
to the pressure. The mean store filling setting L -h at 1 s
chosen ensures in the event of corresponding power
demands a reaction capability with an increased power
consumption in the direction of storage operation and
with an increased power output in the direction of
discharge operation. During "neutral" operation, the
power generation plant is operated in equilibrium
state, as described above, such that the mass flow
flowing into the storage volume is equal to the mass
flow flowing out via the power generation machine. The
power output is between 0% and 100%, preferably 50% to
100%, of the net rating defined above. The plant has
the capability to provide power immediately by power
control or shutting down power consuming machines. At
peak load times, in which high prices are paid for the
electricity that is supplied, the plant is operated in
the discharge mode, with reduced power consuming
machine power and with high or full power generation
machine power. At times when the electricity demand is
low and the electricity price is correspondingly low,
the plant can be changed to the storage mode, in which
the power generation machines are operated at a very
low power level or are shut down, and the power
consuming machines supply storage fluid to the storage
volume. To this extent, this corresponds to the
operation of a storage power station, for example an
air storage power station, that is known per se.
CA 02491653 2005-01-04
- 10 -
However, according to the invention, the storage volume
is not completely charged, but is charged only to a
maximum of 75%. This allows the following method of
operation: At times in which the grid operator pays a
high price to provide the capability to react to the
grid demands quickly, the power generation plant will
be operated with a high power consumption, for example
of more than 800, of the power consuming machines even
when the electricity price is comparatively high. The
power generation machines are synchronized to the grid
at a minimum permissible power level, preferably of
less than 20% of their maximum power, or even at their
minimum power level. For operational reasons, the
minimum permissible power level may also be higher,
however; for example, the operating license for the
plant may forbid relatively long-term operation at such
low load levels, for various reasons. This is the
situation on the one hand when the power generation
plant is provided with additional firing in the power
generation machine, as in the case of a combined
air/gas turbine. On the one hand, the power output of
the power generation machine should sensibly be chosen
such that the firing is already in operation. This once
again allows the objective to be derived that the
firing power should be chosen such that, for example,
the combustion is sufficient in order to ensure that
approved CO and UHC emissions are not exceeded. This
frequently results in an operationally dependent
relative minimum power level which is above the cited
20%. In the described mode, incomplete charging of the
storage volume is made use of, since the plant is in
fact operated in the storage mode and more mass flows
into the store than flows out of it: For this reason,
the store is thus preferably not 100% filled. As stated
above, this initial operating state allows the greatest
possible power dynamic range. On the basis of the
numerical figures stated above, around 200% of the
plant rating can be provided in seconds by shutting
CA 02491653 2005-01-04
- 11
down or controlled deceleration of the power consuming
machines, and another roughly 300% can be provided
within minutes by loading the power generation machine.
The choice of this form of operation in a readiness
mode is primarily a financial question.
As described, the method of operation of a storage
plant according to the invention produces a major
effect on the grid by the use of an initial load and by
feeding in power, such that it is possibly worth while
when, actually for the purpose of transient operations,
control of the power demand is handed over from the
plant operator to the grid operator.
Further advantageous effects and embodiments of the
invention will become evident in the light of the
exemplary embodiment described in the following text,
or are specified in the dependent claims.
Brief description of the drawing
The invention will be explained in more detail in the
following text with reference to exemplary embodiments
which are illustrated in the drawing, in which in
detail:
Figure 1 shows an electricity grid with a storage power
station which can be operated according to the
invention;
Figure 2 shows an example of the embodiment of a
storage power station which can be operated according
to the invention;
Figure 3 shows an example of an operating concept for a
storage power station such as this as a function of the
net power output; and
Figure 4 shows an example of the power output dynamic
range which can be achieved according to the invention.
CA 02491653 2005-01-04
- 12 -
In this case, the illustrated exemplary embodiments
represent only a small instructive detail of the
invention as characterized in the claims.
Approach to implementation of the invention
Figure 1 shows an electricity grid N, highly
schematically. Loads Ml to M8 and three power stations
or their generators G1 to G3, as well as an air storage
power station S, are connected to the electricity grid
via grid switches 8. An air storage power station such
as this has been disclosed, for example, in DE 28 22
575, which disclosure represents an integral component
of the present invention. The air storage power station
S comprises at least one compressor V for filling a
storage volume 100 with an energy storage fluid, as
well as a turbine T which can be operated with the
fluid from the storage volume 100. The turbine T drives
a generator GS which generates electrical power which
can be fed into the electricity grid via the switch
112. The compressor V is driven by a motor MS which
consumes a controllable amount of electrical power via
the switch 111 and the regulator 114. The difference
between the power output of the generator GS and the
power consumption of the motor MS is fed into the grid
N via the switch 113, as the net power output from the
storage power station S. If the power consumption of
the compressor V or of its drive motor MS becomes
greater than the power that is generated in the
generator GS, the storage power station S consumes
power from the grid, via the switch 113. In a first
operation state, all the power loads M1 to M8 and all
the generators Gl to G3 as well as the storage power
station S are connected to the grid. The power
consumption of all of the loads Ml to M8 as well as
that of the drive motor MS and the power output of all
the power stations Gl to G3 as well as that of the
generator GS are balanced out at a set grid frequency.
CA 02491653 2005-01-04
- 13 -
If now, for whatever reasons, one of the power stations
Gl to G3 falls off the grid by opening the grid switch,
the power balance within the grid is actually no longer
balanced, and the grid frequency reacts to this by
dropping. The power stations that are still connected
to the grid normally react to this and increase the
power as quickly as possible in order to support the
frequency. As described initially, power stations have
widely different capabilities for rapid reaction.
Furthermore, rapid load changes such as these in power
stations result in structural loads on expensive power
station components. In addition, the appropriate power
reserves must be kept available, which means that
expensive investment is not completely utilized and
power stations do not run at their best operating point
during normal operation. Overall, these factors mean
that the maintenance and generation of power for
frequency support and for coping with fast load ramps
is very expensive. From the technical point of view, it
would, of course, in fact be desirable to first of all
disconnect appropriate power loads from the grid in the
event of failure of a power station although, for
obvious reasons, it is actually not directly possible.
In fact, the invention likewise make use of the
disconnection of loads from the grid, for example in
the event of failing of electricity generation
capacities, but this is done without affecting any of
the paying loads Ml to M8. This is done by the specific
operation of the storage power station S, as described
in the following text. As described in the
introduction, an air storage power station S as
described is operated in the storage mode at low-load
times, for example at night and at weekends. The switch
112 is opened and the switch ill is closed, such that
the motor MS drives the power consuming machine, the
compressor V, which feeds air or some other energy
storage fluid into the storage volume 100. No fluid
flows out of the storage volume 100. The storage power
CA 02491653 2005-01-04
- 14 -
station S then only consumes power from the grid. This
power consumption makes it possible to operate basic
load units such as nuclear power stations or else coal-
fired steam blocks at high power even at low-load
times, thus making better use of their high
investments. At times when there is a medium grid load,
the storage power station S is not used, and the entire
power demand is covered by the power stations Gl to G3
which are operating close to their best operating
point. At peak load times, the switch 112 is closed and
the turbine T is driven by the energy storage fluid
that is stored in the storage volume 100, and itself
drives the generator GS, from which a power demand that
cannot be covered by the power stations G1 to G3 is fed
into the grid. The invention now makes use of the
knowledge that, even at times when there is a medium or
high electricity demand, the motor MS in a storage
plant S can be operated as a secondary load,
analogously to a "bias voltage" in the electricity
grid. The storage plant S is, for example, operated
such that, during normal operation, the mass flow which
is fed from the compressor V into the storage volume
100 is equal to the mass flow flowing out via the
turbine T. Depending on the electricity price to be
achieved or to be paid for at any given time, the
storage plant may in this case, of course, also be
operated in the storage mode or in the discharge mode;
the critical factor is that the motor MS applies a load
to the electricity grid N even when the net power
output is positive, that is to say, considered from the
global point of view, the storage power station S does
not represent a load. In the event of sudden changes in
the power demand on the storage plant S, such as those
which occur for example and to a particular extent in
the event of failure of one of the power stations Gl to
G3 and when a large load is connected to the grid, this
secondary load can be changed considerably more
efficiently and quickly than it is possible to provide
CA 02491653 2005-01-04
- 15 -
additional power. If, for example, one of the power
stations Gl to G3 has to be disconnected from the grid,
the power consumption of the motor MS is, according to
the invention, reduced by simple circuitry means which
are known per se, or the switch 111 is opened entirely.
This means that additional power, which was previously
consumed by the motor MS, is available virtually
instantaneously for the loads Ml to M8. The turbine T
may in this case be operated without any problems using
energy storage fluid that is provided from the storage
volume 100. In a next step, for example, the power of
the turbine T can be increased or it can be started up
for the first time; in addition, further power stations
which are acting on the electricity grid can increase
their power, or additional resources can be connected
to the electricity grid in order to compensate for the
initial power station failure; the motor MS and hence
the compressor V of the storage palnt S can then be
starred up again successively.
The storage power station S is illustrated highly
schematically in Figure 1. Figure 2 shows an example of
an embodiment of a storage power station S. The power
consuming machine, the compressor, V in this case
comprises two compressor runs each having two
compressors and two coolers. In each compressor run, a
first compressor 11 or 13 compresses air to an
intermediate pressure. The air is cooled at an
intermediate stage in a cooler 21 or 23 and is
compressed in a second compressor 12 or 14 to a final
pressure, which is typically in the range from 30 to
100 bar or 50 to 100 bar. The compressors are driven by
drive motors MS1, MS2, MS3 and MS4. The compressed air
flows through a throttling and shut-off member 3 into
the storage volume 100. Stored air flows via a
throttling and shut-off member 4 to the turbine unit T.
Within this turbine unit T, the air first of all flows
through an exhaust gas heat exchanger 5 where, for
CA 02491653 2005-01-04
- 16 -
example, it is heated to 550 C. After this, the air is
expanded in an air turbine 6 to a pressure of around 10
to 15 bar. The state of the air at the outlet from the
air turbine 6 is normally comparable to the state at
the compressor outlet from a gas turbine group. For
this reason, the combustion chamber 7 and the turbine 8
of a gas turbine group can be arranged very
particularly advantageously downstream from the air
turbine. A fuel is burnt in the air in a manner known
per se in the combustion chamber 7, resulting in the
production of a compressed hot gas, which is expanded
approximately to the environmental pressure in the
turbine 8, carrying out work in the process. The
expanded hot gas is optionally reheated in a further
burner 9, and then flows through the exhaust gas heat
exchanger 5, in which the residual heat from the
exhaust gas is transferred to the supply air to the air
turbine 6. The air turbine 6 and the gas turbine 8 of
the turbine unit are arranged on a common shaft and
drive the generator GS. In contrast to a conventional
gas turbine group, the compressor and turbine are
mechanically completely decoupled from one another and,
owing to the intermediate storage volume in the flow
path, the fluid-mechanical coupling also has a certain
amount of elasticity. This makes it possible for the
turbine unit T and the compressor unit V to be operated
independently of one another, such that, as described
above, it is possible to react very highly flexibly to
different power demands by means of two mechanisms,
namely via the power consumption of the compressor unit
and the power output of the turbine unit, and to
increase the net power output virtually
instantaneously, in particular by shutting down power-
consuming compressors. In this case, the compressor
runs, which are arranged in parallel with the mass
flow, can likewise be controlled independently of one
another, thus further simplifying the power control for
the entire storage plant S.
CA 02491653 2005-01-04
- 17 -
It is advantageous for the controllability of the
storage plant for two or more independently
controllable compressor runs to be arranged as power
loads in parallel with the mass flow, and likewise for
two or more turbine units T to be connected as power
generators to a storage volume in parallel with the
mass flow. Figure 3 shows an example of an operating
concept for a storage power station with four
compressor runs and two turbine units. In this case,
100% power is defined as the net power output PNET which
is produced when both turbine units and all four
compressor runs are operating at maximum power in the
equilibrium state with respect to the mass balance of
the storage volume 100. The line which passes
diagonally through the diagram and is annotated PNET
represents the net power output. That portion which is
below 100% and is annotated P is the respective power
consumption of the compressors. In a first operating
region, which is annotated 4VOT, starting at -200% net
power output, that is to say 200% net power
consumption, all four compressor runs and none of the
turbines are in operation. As the power consumption
falls, the power consumption of all four compressor
runs decreases slowly until, at one point, one of the
compressor runs is taken out of operation. Three
compressor runs are then operated at full power
consumption and are likewise decelerated slowly; this
region is annotated 3VOT. This is followed with a lower
net power consumption by a region 2VOT, in which two
compressor runs and no turbine unit are operated. After
this, a first turbine unit is started up, and all four
compressor runs are operated at the same time. In the
region 3V1T, three compressor runs and one turbine unit
are in operation, in the region 1V1T, one compressor
run and one turbine unit are in operation, and so on.
At 150% net power output, the second turbine unit and
two compressor runs are started up at the same time.
CA 02491653 2005-01-04
- 18 -
The maximum peak load net power is achieved when both
turbine units are operated at full load and no
compressor run is in operation, that is to say in the
region OV2T. The net power output is then 3000. The
power which is in each case shown below 0% is the
respective power consumption of the compressors, and
represents the power which can be provided immediately
as additional net power in the manner described above.
Equilibrium operation is achieved, for example, when
both turbine units and all four compressor runs are
running on full load, thus resulting in 100% net power
output; the power consumption of the compressors is
then 200%; this means that, on the basis of the
operating method according to the invention, the
storage power station is able to compensate immediately
and without any delay for failure of a power station
block whose power corresponds to twice its own rating!
This frequency support capability and the wide control
range underscore the superiority of a storage power
station operated according to the invention.
Figure 4 schematically illustrates the power dynamic
range which can be achieved by means of the method
according to the invention. The net power output PNET is
plotted on the vertical diagram axis, with negative
values indicating a power consumption, and the time is
plotted on the horizontal diagram axis. This is based
on an initial operating state in which, as already
described a number of times, the power consuming
machines are running at full power and the power
generation machines are actually synchronized to the
grid, or are operated at a very low power level, up to
a maximum of 20% of the maximum power. Furthermore, it
is quantitatively assumed that, when the power
generation machines are operating in the steady state
on full load, 2/3 of the entire power that is generated
is required for compression of the working medium, and
that the power consuming machines are designed for
CA 02491653 2005-01-04
- 19 -
their maximum power at this operating point. It would,
of course, also be possible to design the power
consuming machines to be larger and thus to cover an
even wider power range. In the initial operating state,
the net power output is -200%; power is thus being
drawn from the grid. At the time t=To, the maximum
amount of additional power is being demanded from the
power generation plant operated according to the
invention. This power generation plant reacts to the
situation by shutting down the power consuming
machines, as a result of which 200% power is released
virtually instantaneously; the net power output is then
0%. Even when compressors that are used as power
consuming machines are run down in a controlled manner,
the typical power gradients that are achieved are
around 120% per minute, with respect to the plant
rating, as has already been defined a number of times!
At the same time, the power output of the power
generation machines is increased, which leads, although
considerably more slowly, to a further power increase
up to 300%. It must be stressed that the additional
useful power in the case of a storage plant such as an
air storage plant can intrinsically be produced very
quickly but that, in all cases, this takes place at
least one order of magnitude more slowly than is
possible by reducing or shutting down the power
consumption of the power consuming machines in the
storage power station. Typically, it can be assumed
that the power generation machine can consume power
with a gradient of around 15% per minute. A dashed line
is used to show the dynamic range with which the power
station plant can advantageously react to a falling
power demand. In this case, a falling net power output
is first of all provided by controlled acceleration of
the power consuming machine at, for example, 120% per
minute, thus making it possible to achieve a reduction
in the net power output of around 200% of the plant
rating in 100 seconds. If greater load changes occur,
CA 02491653 2005-01-04
- 20 -
the power output of the power generation machine is
also changed. This also reveals another interesting
item. A storage plant of the described type which is
operated according to the invention makes it possible
to achieve rapidly successive load cycles of up to 200%
of the plant rating without needing to subject
thermally highly loaded components to any alternating
load. Within this order of magnitude, the power control
can be carried out completely by the power consuming
machines. Reference is made once again to Figure 2, in
order to estimate the alternating thermal load on them.
Assuming that the storage pressure is 100 bar, that the
pressure ratios of each of the series-connected
compressors 11 and 12 or 13 and 14 are the same, that
the compression process takes place from the
environmental state at 15 C and intermediate cooling is
carried out in the coolers 21, 23 to environmental
temperature, and subject to isotropic compression,
maximum temperatures of little more than 300 C are
reached and, if the storage pressure is 50 bar, only
around 250 C. These temperatures are, of course,
considerably lower than those in the power consuming
machine, for which reason alternating loads result in
considerably lower loads on the structures. As
mentioned a number of times above, the power range to
be covered solely by compressor control can be widened
further by designing the compressors to be
correspondingly larger.
An additional advantage of the method according to the
invention is that the technology of air storage
turbines and their use for peak load coverage are well
known and proven in engineering. When designing a power
station to be operated according to the invention, it
is also possible to use proven standard components to a
major extent.
CA 02491653 2005-01-04
- 21 -
List of reference symbols
3 Shut-off and throttling member
4 Shut-off and throttling member
Heat exchanger, exhaust gas heat
exchanger, recuperator
6 Air turbine
7 Combustion chamber
8 Gas turbine
9 Additional firing
11 Compressor
12 Compressor
13 Compressor
14 Compressor
21 Intercooler
22 Air cooler
23 Intercooler
24 Air cooler
100 Storage volume
111 Switch
112 Switch
113 Mains switch
114 Regulator
Gl, G2, G3 Power stations
GS Generator for the power generation
machine in the storage power station
Ml, M2, M3, M4, M5, M6, M7, M8
Loads
MS Drive motor for the power consuming
machine for the storage power station
MS1, MS2, MS3, MS4
Drive motors for the power consuming
machine for the storage power station
S Storage power station
T Turbine unit, power generation machine
V Compressor unit, power consuming machine
PNET Net power output
CA 02491653 2005-01-04
- 22 -
P_ Power consumption of the power consuming
machine
