Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 022~090~ 1998-10-01
PCT~DE97/00659
A Process for Operating a Block-type Thermal Power Station
The present invention relates to a process for operating a block-
type thermal power station (BTPS), with which useful energy--such
as current as electrical energy and/or heat or cold as thermal
energy--is generated on a demand-dependent basis from primary
energy such as oil and/or natural gas, in particular from waste
0 gas.
The expression "block-type thermal power stations" is understood
to refer to current-generating plants that simultaneously utilize
heat, the primary energy, for example, various gases and light or
heavy grade heating oil, being converted by internal combustion
machines such as reciprocating engines or gas tubines. Heating
boilers are also used to cover peak demands for heat.
The particular benefit of BTPS's is the high overall efficiency,
ranging from approximately 85 per cent up to 92 per cent, with
which the primary energy that is used is converted into current
and useful heat. In particular, the approximately 40 per cent of
CA 022~090~ 1998-10-01
PCT~DE97/00659
valuable electrical energy and the simultaneous use of current
and heat are typical of the benefits obtained from BTSP's.
BWK Brennstoff-Warme-Kraft [BWK - Fuel, Heat, Power], 41 (1989),
No. 6, pp. 273 - 277 proposes a daily use optimization for power
station systems with an energy-heat linkage. Admittedly, the
feasibility and benefits of powerful software for the daily use
optimization are discussed within the framework of one study,
although the system is intended in particular for use in
lo municipal power stations that operate a thermal power station,
amongst other things. The special conditions that apply to
BTSP's are not examined in detail. Essentially, the same applies
to WO A 95/16296, which treats of control of plants that are
operated co-operatively.
BWK Brennstoff-Warme-Kraft [BWK - Fuel, Heat, Power], 47 (1995),
No. 11/12, pp. 461 - 464, and BWK Brennstoff-Warme-Kraft [BWK -
Fuel, Heat, Power], 48 (1996), No. 1/2, pp. 61 - 66 treat of the
specific problems associated with block-type thermal power
stations; also discussed in detail is the importance of the
dynamics involved in the planning and operation of block-type
thermal power stations. The point of departure in these papers
. .
CA 022~090~ 1998-10-01
PCT~DE97/00659
is that up to now there has been no verified instrument for
calculating the diurnal variation of heat demand for block-type
thermal power stations. In particular, because of a lack of
testing, it is impossible to answer the question as to whether
standardized supply profiles of different objects lead to
acceptable results. Proceeding from this, it is necessary to
determine dynamic computational aids for dynamic planning.
Comparative values that are intended to lead to optimization of
operation have been determined for practical conditions. In
o actual fact, however, to a very large extent, manual intervention
is still needed.
BWK Brennstoff-Warme-Kraft [BWK - Fuel, Heat, Power], 47 (1995),
No. 11/12, pp. 476 - 479 describes how operational optimization
of block-type thermal power station can be effected by
integration of a PC-based energy-management system into control
technology. To this end, it is important that in addition to
transmitted data, data that reflect the economic limiting
conditions also be available on the PC.
In contrast to this, it is the task of the present invention to
propose a process for operating block-type thermal power
.
CA 022~090~ 1998-10-01
PCT~DE97/00659
stations, by means of which extensively automated operational
optimization can be effected.
According to the present invention, this problem has been solved
s in that the prognosis is a long-range computation based on
various, currently existing criteria, and in that the consumption
integrals that can be derived therefrom can be determined, the
selection and/or regulation of the current and heat generation
being optimized on the basis of energy costs in the various
lo tariff phases. It is preferred that variously plotted, self-
correcting day-type curves be used as the criteria for the long-
range calculation. When this is done, a number of parameters in
day-to-day operation can be adjusted.
A well-rounded concept, especially for the operation of block-
type thermal power stations, will be pursued by way of the
present invention. At the same time, greatly improved economy is
made possible by the following changes in operating conditions:
- concentration of the plant operation in the [electrical]
current high tariff phases;
- the best possible exploitation of thermal energy that is
generated;
CA 022~090~ 1998-10-01
PCT~DE97/00659
- the greatest possible annual operating service for the
current-generating units;
- optimization of incoming and outgoing energy.
The present invention thus provides for optimized selection of
current and heat generation as based on energy costs, this being
done on the basis of both primary energy and of useful energy.
When this is done, it is advantageous that consumption curves for
the following 24 hours are determined from established 15-minute
o values that are synchronized with a maximum control of the energy
supplier.
The process according to the present invention has essentially
been implemented by software. The software for a conventional
processor or computer constitutes the means for selecting the
operational data.
This means that a new control system has been created as a
optimization program for block-type thermal power stations. It
is preferred that the appropriate software, written in assembler
code, be installed in an automating device by means of neuron
networks.
CA 022~090~ 1998-10-01
PCT~DE97/00659
Additional details and advantages of the present invention are
set out in the following description of the drawings of
embodiments, in conjunction with the patent claims. The drawings
that are appended hereto show the following:
s
Figure 1: The heat requirement for summer, transitional season,
and winter, which can be retrieved from a block-type
thermal power station;
Figure 2: Heat requirement and plant running time for the
o transition season described in Figure 1, work being
done with and without a cooling tower, as selected;
Figure 3: Automatic curve generation for the example of the
transitional season in Figure 1;
Figure 4: A structural plan for an energy optimization program.
Block-type thermal power stations (BTPS) are used to generate
current and heat by using primary energy. They are small power
generating stations that generate energy where it is to be used.
Primary energy (e.g., natural gas, diesel oil) or waste energy
(sewer gas, landfill gas, pit gas) is converted into useful
energy (heat, cold, current), efficiently and in an environ-
mentally safe manner in such plants.
CA 022~090~ 1998-10-01
PCT~DE97/00659
In addition to boilers for heat peaks and peripheral systems for
integrating current and heat, at the centre point is a gas engine
system with a synchronous generator. The waste gas, cooling
water and other sources of heat provided by the engine are used
as heat or process energy.
Block-type thermal power station systems are built to provide
electrical power outputs from 100 kW to some 20,000 kW. A block-
type thermal power station system comprises one or a plurality--
o at most six to eight--systems and peak-period boilers, should
these be necessary.
As is known, the economy achieved by block-type thermal power
stations results from the great efficiency achieved during
current and heat generation. Depending on operating conditions,
the degree of efficiency will reach 85 per cent to 90 per cent.
The potential for improving the design and grouping of the block-
type thermal power station components have thus been exhausted to
a very large extent.
Figures 1 to 3 each show the daily profiles of energy
requirements. The time, graduated in hours, is-shown along the
CA 022~090~ 1998-10-01
PCT~DE97/00659
abscissa, and the output, graduated in thousands of kW, is shown
on the ordinate. Figure 1 has three curves, numbered 1 to 3, and
these show the heat requirement for a day in summer, in the
transitional season, and in winter. One can see, in each case,
that there is a distinctive curve with peaks, in particular in
the morning, at noon, and in the evening. Such curves are known
in principle.
Figure 2 shows the curve 2 that is shown in Figure 1. The
particular requirement for heat is represented by a generator-set
running time. In particular, two or three generator sets are
operated, and each of these generates 1.7 MW. This means that in
the present example, a requirement of more than 5 MW can be
accommodated; such a case exists, for example, in the early
morning hours. Since demand falls off at mid-day, only to rise
again in the evening, when it can reach 7 MW, the three
generating sets that are in operation in the mid-day period can
activate a reservoir, from which additional heat can be drawn in
the evening.
As Curve 2 in Figure 1 and Figure 2 shows, the demand for heat
can have a specific bandwidth of 500 kW. As will be shown below,
CA 022S09OS 1998-10-01
PCT~DE97/00659
the exact curves can be determined in detail by external
correcting factors, as is shown, for example, in Figure 3. In
each instance, this results in an energy optimization program for
the day.
Essentially, the energy optimization program includes the
following functions:
1. Consumption forecast
o A reliable and extremely precise forecast for current and heat
consumption is generated continuously for the following 24-hour
period. The consumption curves are made up from measured 15-
minute values that are synchronized with the maximum control of
the energy supplier.
The forecast is generated, for example, on the basis of eight
different self-correcting day-type curves. The outside
temperature, the vectors for the last four measured values for
current and heat consumption, and a programmed calendar for the
20 day types are used as parameters when generating the forecast.
CA 022~090~ 1998-10-01
PCT~DE97/00659
The following parameters are used;
a) Storage of day-type curves for current and heat;
- normal working day
- transition day (early close of business)
- day prior to a working day, e.g., Sunday
- day after a working day, e.g., Saturday
- working day: two-shift operation
- special day 1 (bridge day)
- special day 2 (interim operation)
All day types are corrected constantly and automatically with the
measured, actual values for the last five days of the same type.
b) Correction functions for day curves:
- Outside temperature
A comparative value, which is dependent on time of day,
modifies the heat-demand curve;
- Heat output hourly average
- The steepness of electrical power output changes is
compared to the output curve and corrects the current-
demand curve;
CA 022~090~ 1998-10-01
PCT~3E97/00659
- Atmospheric humidity
If required, atomospheric humidity is measured and
matched to the heat-demand curve relative to a
comparative value.
- Daytime brightness (sunny~ overcast)
If required, daytime brightness is measured and matched
to the heat-demand curve relative to a comparative value.
c) Calendar function
- Running time 1 year (12 months)
o - Input of day types.
2. Assessment of energy costs
The types of energy and the periods in which these types of
energy are required are arranged in a priorities list that can
include up to 20 degrees. This means, in particular, that the
most costly form of energy, e.g., peak-hour current, has a value
of "1," and the cheapest form of energy, e.g., night-time heat,
has a value of "20."
The changeover of the priority steps is determined by the system
on the basis of the stored high-tariff and low-tariff periods and
the consumption curves that have been generated.
CA 022~090~ 1998-10-01
PCT/DE97tO0659
Example for the assessment scale with energy-cost points
Points
1. Current generation - peak period 40
2. High-tariff current generation 22
3. High-tariff heat generation 8
4. High-tariff heat storage for low-tariff periods 7
5. Low-tariff heat reservoir discharge 6
6. Low-tariff heat generation 5
o 7. Peak-load boiler useful output 2
8. Gas reservoir discharge (primary energy) 2
9. Gas reservoir charging (primary energy) o
10. Peak-load boiler o
11. Cooling tower for heat removal -6
12. Recooling by fresh water -4
13. Gas (primary energy) -4
14. Waste gas (primary energy) -4
15. Diesel (primary energy) o
16. Load shedding Rank 1 ..... 16 -10
17. Peak-load gas (primary energy) -40
3) Demand-curve adaptation
The actual measured correction factors are compared with the
factors for the curve that is stored in the program and the curve
25 iS then updated by an appropriate parallel shift, either upward
or downward. The weighting of the correcting factors is
predetermined in advance for each factor individually. Run-time
optimization always works with most recent, up-to-date curves.
. .
CA 022~090~ l998-lO-Ol
PCT/DE97/00659
4) Energy data
The number and function of the specific plant components are
input into the operator surface of the computer program. The
availability of the individual systems is automatically corrected
5 by the external block-type thermal power station-process
measuring and control technology. Ideal operating modes are
regulated on the basis of generation and consumption data as well
as of partial-load performance of the individual components.
o 5) Run-time optimization
The energy optimization program determines the integral for the
heat demand during the current low-tariff phases on the basis of
the forecast heat demand for the coming 24 to 48 hours, and then
compares the future demand with the heat reservoir capacity
15 available and necessary to satisfy this.
Then, to the extent that this is possible, the current/heat
generating aggregates are operated in such a way that the total
heat demand is generated by the generating sets during the high-
20 tariff phase for current.
In the event that the generating capacity of the generating sets
or the capacity of the heat reservoir make this impossible, then
in keeping with the shortfall in thermal energy and the
_ _
CA 022~090~ l998-lO-Ol
PCT/DE97/00659
generating sets or boilers that are dependent on energy valency
for base load generation, operation will be carried on
continuously in the low-tariff phase so that the reservoir
contents will then be sufficient at the lower demand level.
The efficient use of all generating and recooling components will
be regulated in the same way, and with similar formation of the
integral and computation of the particular energy costs. Thus,
the energy optimization program permits a reduction of the
o installed power, i.e, a reduction of the total investment, by the
ideal use of the generating sets and boilers.
For the remainder, operation in the cost optimum economizes in
primary energy and more higher-value energy is generated, and the
15 profit deriving from operation of the plant is thus improved.
Using the criteria described in detail heretofore, it is possible
to produce software that can be used to run an existing and
available processor. Critical in this is that all relevant data
20 for the routine operation of the block-type thermal power station
be recorded. The structure of such an energy optimization can be
seen, in particular, in the diagram shown in Figure 4, which is
essentially self-explanatory.
CA 022~090~ 1998-10-01
PCT/DE97/00659
In Figure 4, of special importance for a block-type thermal power
station 10 et seq are those units that relate in particular to
the heat demand and, 20 et seq, those units that relate in
particular to the demand for current. In particular, 11 involves
5 the formation and storage of day-type curves for heat demand, for
example, nine day types, and 12 involves determination of the
correcting factors for the heat demand, from which, in unit 13, a
demand forecast for heat covering, for example, 24 to 48 hours is
derived. In unit 24 the latter are used to determine the heat
o that can be generated with the power-heat coupling, when together
with the unit 13, a unit 25 for forecasting and exploitation of
the heat reservoir capacity is activated with the demand forecast
for heat. In the same line, a unit 26 for [electrical] current
tariff intervals with network synchronized maximum control and,
15 optionally, requirement demands for peak-load current of the
power supplier is also available.
This is followed by a unit 30 for run-time optimization of the
generating sets in order to maximize the current generation in
20 high-tariff phases. This unit 30 is followed by a unit 32 for
selecting the required power-heat coupling generating sets, and
by a unit 33 for determining the power-heat coupling generating
sets that are available but not required. In the same line there
is a unit 34 in which one list of assessed energy types, for
CA 022~090~ l998-lO-Ol
PCT/DE97/00659
example, 20 values, is compiled, and a unit 35 for setting out
supplier and consumer capacities, such as status, capacity, and
so on, for example, for 20 values. Unit 23 to unit 35 control a
subsequent unit for energy value optimization, whilst taking into
5 account all outflowing energy types and their values. The
individual demands for the generating sets 1 to n, such as
boilers, reservoirs, cooling towers, and so on, can be called
from unit 40.
o The units shown in the structural plan at Figure 4 are formed
from suitable software that can be generated in a simple manner
in Assembler Code.
In as much as no precise analytical connections can be determined
15 for the energy optimization program described above, it is
possible to use neural networks, based on approximations, to
advantage; these can be so trained that they can be set to
particular conditions so as to be autodidactic.
