DE102014223853A1 - Use of a LIDAR system for the short-term prediction of expected wind conditions and as a basis for power plant management and power plant management process based on a signal available and expected wind conditions from a LIDAR system - Google Patents

Abstract

It proposes the use of a LIDAR system (20) for short-term forecasting in a 1-10 minute forecast horizon, in particular 1-5 minutes, expected wind conditions at the location of a wind power plant (12) and power plant management based on the calculated expected wind conditions, and a method for Power plant management of a power grid unit (10), in which at least one wind power plant (12) feeds electrical energy and wherein at least one further power generator (14) is activably connected to the power grid unit (10), with the following steps: By means of the LIDAR system (20) a short-term forecast in a prediction horizon of 1-10 minutes, in particular 1-5 minutes, expected wind conditions at the location of the wind power plant (12) and on the basis of the determined expected wind conditions, activation of further energy generator (14) or another energy generator (14).

Description

The invention relates generally to a method for predicting a future situation as the basis for optimized use of renewable energy, such as wind energy. In particular, the invention relates to a method of predicting expected wind conditions and a method of using a related power plant management signal.

Regardless of which energy source / carriers or which energy sources / carriers, for example fuels based on fossil fuels or regenerative energy sources such as wind power, are used, there is always the need to generate energy, in this case electrical energy, as accurately as possible to adapt to the respective energy requirements. The poor predictability of a respective contribution to the total available amount of energy due to renewable energy is a problem. In so-called Iceland grids, ie smaller, independent power grids, individual or few wind energy and / or photovoltaic systems can make up a significant proportion of the electrical energy available within the power grid. Then, local variations, for example, in the wind energy that can be used for power generation, lead to considerable fluctuations in the electrical energy available within the power grid, which are not readily compensated by corresponding fluctuations elsewhere. On the other hand, however, especially in such smaller power grids, renewable energies are also competitive in terms of cost compared to fossil fuels, in particular compared to a typical power generation by means of, for example, diesel generators.

An optimal balancing of a respective contribution of the individual energy sources in a power grid requires a foresight. This is especially the case when energy sources are added, which feed electrical energy into the respective power grid on the basis of wind power and / or solar energy. In the boundary conditions for such a forecast, it should be borne in mind that a diesel generator usually requires a lead time of 1-5 minutes until electrical energy can be fed into the respective power grid by means of the diesel generator. Although the loss of each available amount of energy due to, for example, a sudden shading of a photovoltaic system by a cloud layer or a doldrums in a wind turbine can be compensated in principle by connecting a diesel generator, while switching on and the lead time of the diesel generator is initially only one due to the reduced contribution the photovoltaic system or wind turbine resulting reduced amount available. It would therefore be desirable to be able to detect and predict meteorological influences such as clouds and changing wind conditions on the energy available within the power grid in sufficient time, if necessary, to take measures to compensate for a decline or an increase in the energy input elsewhere can be.

Of particular importance is the wind power. When changing the wind conditions, the pitch of the rotor blades (pitch) are known to be adapted to a wind turbine, resulting in fluctuations in the amount of energy each fed into the power grid can result. However, wind gusts, which can also lead to an emergency shutdown of a wind turbine, for example, have a greater influence on the amount of energy fed in, so that the previous energy contribution of the wind power plant or a group of wind turbines disappears within a very short time and must be compensated elsewhere. Similar effects are generally associated with weather fronts. The effects of calm or a sudden decrease in wind force are less drastic. But the resulting decline in energy fed in must also be compensated for elsewhere. Conversely, even when the wind is suddenly refreshed, a resulting increased energy input must be compensated for elsewhere.

It would therefore be desirable to be able to predict with sufficient certainty the prevailing wind conditions at the location of a wind power plant or a group of wind power plants in order to be able to support a strategy for power plant management, namely for load balancing within the power grid. The prediction does not have to reach very far into the future. Sufficient is, for example, a prediction for each coming five or ten minutes, with this time value (time horizon of the prediction), for example, after a switch-on and lead time of other, for compensation purposes switchable energy generator / sources, such as a diesel generator, directed.

For measurements in the atmosphere so-called LIDAR systems are known. These are known to emit laser pulses and detect the backscattered from the atmosphere light. From the transit time of the signals and the speed of light, the distance to the location of the scattering and thus, for example, a distance of a weather front or a cloud can be calculated. By means of varying beam directions (scanning directions) a three-dimensional model of the atmospheric conditions in the investigated direction can be created. Using temporally staggered scans, it is also possible to determine a speed of cloud movements and from this the wind speed prevailing at the scanning location.

The invention is on the one hand the use of a LIDAR system for a wind power plant for a short-term forecast of expected wind conditions and on the other hand the use of such a system and the resulting forecast results for an optimized management of the power plants / power plants feeding into a respective power grid. A short-term prediction is to be understood as a prediction which predicts the wind conditions to be expected and the energy input of the respective wind power plant or a group of wind turbines individually and together here and hereinafter referred to as the wind power plant of the next approximately 5-10 minutes, in particular each coming 1-5 minutes, concerns.

The power plant management process relates primarily, but not necessarily exclusively to a power grid unit of the type mentioned, ie a smaller, independent power grid unit, in which at least one wind power plant feeds electrical energy and to which at least one other power generator, for example a diesel generator, is activated connected, and comprises the following steps: By means of a LIDAR system, a short-term forecast for wind conditions expected in a forecast horizon of 1-10 minutes, in particular a forecast horizon of 1-5 minutes, takes place at the location of the wind power plant. On the basis of the calculated expected wind conditions and / or expected due to the expected wind energy input, for example, an activation of further energy producers or another energy producer, if due to the expected wind conditions of a decreasing energy contribution of the wind power plant is assumed, or for example an activation of energy storage, such as batteries or the like, if based on the expected wind conditions from an increasing energy contribution of the wind power plant is expected.

When operating a wind power plant, predictive values are already being determined and used. So it is common to use long-term forecasts of up to four days with a 30- or 60-minute resolution based on numerical weather forecasts. In addition, within a wind power plant with several wind turbines, the future resulting from a wind turbine energy contribution based on a momentary energy contribution of a windward wind turbine of the same wind power plant can be estimated. The known distances of the wind turbines with each other and the energy contribution of each remote wind turbine or a wind speed measured there are taken into account. A so-called forecast horizon of such predictions is dependent on the respective wind speed, and in the case of a wind power plant with a diameter of, for example, 1 km and a wind speed of 25 m / s, the forecast horizon is approximately 40 seconds. In addition, the measured values of a LIDAR system for windward measurements with distances of up to 200 m are used to control or regulate a single wind turbine. This leads to a forecast horizon of the order of 20 seconds for the respective individual wind turbine.

Such predictions are all not or at least not well suited as a basis for power plant management, because just the period of each coming 1-10 minutes, especially the period of each coming 1-5 minutes, is not covered. Without a reliable prediction precisely for this time range, a large amount of reserve energy, which is often unnecessary, has to be maintained within the scope of power plant management. For this purpose, it is necessary, for example, that diesel generators, which can be used in the context of power plant management, already running or running permanently and thus at any time for an additional energy supply available and / or that may also be used for an energy feed batteries or the like are charged and maintained at a sufficient state of charge. These requirements usually limit the amount of wind energy usable in the context of hybrid power generation to about 30%.

Irrespective of this, LIDAR measurements are also used for planning of wind power plants, namely for selecting a location of a wind power plant and / or for location-specific adaptations of the wind power plant. Further applications are a performance monitoring of a wind turbine and controls or arrangements for optimized operation of a single wind turbine within each considered ultrashort period of less than 20 seconds and a detection range of about 200 m (see above). Finally, LIDAR measurements are also used in the design of wind turbines and in basic research for the development of wind turbines, for example for the validation of numerical simulations of the flow conditions of so-called wake wake turbulence.

The advantage of the approach proposed here is that using a LIDAR system also for the for the Power plant management so essential time range of 1-10 minutes, especially 1-5 minutes, a reliable statement in terms of expected within this time range wind conditions and a subsequently expected energy contribution of the wind power plant or single or multiple wind turbines comprising it can be determined.

The output signals used as part of the power plant control and generated by the LIDAR system can have a very different information content. This goes from simple binary signals that indicate, for example, a "change in wind conditions within the next 1-5 minutes" to concrete numerical estimates, such as "changing wind conditions in three minutes", "energy contribution increases by 1 MW / s within the next 3 minutes "or" Energy contribution of 8.3 MW in 5 minutes ".

For example, to automatically generate a simple binary signal indicating, for example, a change in wind conditions within the respective forecast horizon, a predetermined or predeterminable threshold is used and, as soon as a future wind force expected due to a LIDAR measurement transduces the instantaneous wind force by more than the threshold - or falls below, such a signal is generated. In a particular embodiment of such binary signals, a first binary signal is generated when the future wind strength exceeds the instantaneous wind force by more than the threshold and a second binary signal when the future wind force falls below the instantaneous wind force by more than the threshold. Then, based on the respective binary signal directly a corresponding control of another power generator of the same power plant, so for example a diesel generator done. An example of such a drive is that with a binary signal indicating a future wind speed which is less than the current wind speed by more than the threshold, the diesel generator is started, so that the moment the reduced wind strength affects the energy contribution the wind power plant or individual wind turbines, the diesel generator is available for feeding the then lack of energy.

The more accurately a signal generated on the basis of LIDAR measurements reflects the future wind conditions or their dependent variables (energy contribution per time unit, energy contribution per period absolute, energy contribution from time to absolute, etc.), the more specific can be reacted in the context of automatic power plant management. If, for example, a numerical estimated value is available for a future expected reduced energy contribution, a number of diesel generators or a diesel generator with sufficient power required to compensate for the reduced energy contribution can be started in good time or comparable measures can be initiated.

In order to generate a signal based on LIDAR measurements (control signal), a device which is also referred to below as the control unit is determined and set up, on the one hand at least one LIDAR system for generating laser beams and for evaluating the portions of the laser beams reflected in the atmosphere according to the LIDAR Principle and on the other hand comprises a signal conditioning unit. The signal conditioning unit generates, for example, the above-mentioned binary signals and / or the above-mentioned numerical estimates.

In a particular embodiment, the signal conditioning unit may also comprise one or more filters, for example based on neuronal networks, genetic algorithms and the like, which optimize the signal or LIDAR signal available from the LIDAR system, for example on the basis of historical data. This optimization is done by training the filter on the basis of historical LIDAR signals and time correlated data for actual wind conditions and / or energy contributions, etc. so that the control signal output by the control unit progressively better represents the actual future conditions as the filter progresses. Additionally or alternatively, such a filter can also take into account a model, for example a model, the surrounding topography and / or vegetation, the location (s) of a wind turbine or individual wind turbines, the effects of shading and / or a so-called CFD simulation considered. The model can be a model that has the respective boundary conditions considered in a rather fine resolution, for example in a 1 m, 2 m or 5 m grid, or in a medium resolution, for example a 100 m-raster, underlying. Additionally or alternatively, such a filter may also take into account a weather model on a macro level (1-50 km grid) to combine the local LIDAR measurements with regional long term weather effects.

Within the power plant management, the control unit acts as a controller which, on the basis of a control signal generated by the control unit, automatically activates or deactivates further energy producers, such as a diesel generator or several diesel generators.

The invention is thus likewise such a control unit and an implementation of the method described here and hereinafter, in particular in the form of software, ie a corresponding computer program with program code instructions executable by a computer and a storage medium with such a computer program, and a control unit with a processing unit in Form of or in the manner of a microprocessor and a memory in which such a computer program is loaded as means for implementing and executing the method and executable by the processing unit and executed in the operation of the control unit by the processing unit.

An embodiment of the invention will be explained in more detail with reference to the drawing. Corresponding objects or elements are provided in all figures with the same reference numerals.

The embodiment is not to be understood as limiting the invention. Rather, additions and modifications are quite possible in the context of the present disclosure, in particular those, for example, by combination or modification of individual in conjunction with the described in the general or specific description part and in the claims and / or the drawing features or method steps for those skilled in the art with regard to the solution of the problem can be removed and lead by combinable features to a new subject or to new process steps or process steps.

Show it

1 a power grid unit with a wind power plant supplying the power grid unit as well as a diesel generator which can be activated and connected to the power grid unit and likewise connected consumers,

2 a control unit with a LIDAR system for the automatic activation of the diesel generator depending on future expected wind conditions,

3 . 4 and 5 Illustrations explaining a determination of wind direction and wind speed using a LIDAR system,

6 Snapshots of changing wind conditions,

7 an alternative embodiment of the control unit 2 such as

8th an illustration to explain a physical model for the determination of future expected wind conditions.

1 shows in a simplified schematic form a smaller power grid unit 10 (island grid) with individual power plants / energy producers included. Accordingly, the power grid unit includes 10 at least one wind power plant 12 with at least one wind turbine or a plurality of wind turbines and at least one further power generator 14 , for example a diesel generator 14 , To the power grid unit 10 is at least one consumer in addition to these dining units 16 , usually a plurality of consumers 16 , connected.

Due to the energy demand of the or each to the power grid unit 10 connected consumer 16 results in a in the power grid unit 10 necessary amount of energy. For simple conditions it is assumed for the further description that the wind power plant 12 can normally provide this amount of energy. If that of the wind power plant 12 produced amount of energy, for example, due to currently insufficient wind conditions, not sufficient, is the respective other energy producer 14 , so here's the diesel generator 14 or a diesel generator 14 , then connected, which then additional electrical energy in the power grid unit 10 feed and thus ensure the availability of the necessary amount of energy.

Such an automatic influence on the wind power plant 12 and / or the diesel generator 14 is referred to in accordance with the usual terminology as power plant management and automatic power plant management is a control unit 18 intended. This includes in a conventional manner means for detecting each of the wind power plant 12 provided electrical energy and means for detecting the respective energy needs of the consumer 16 or the consumer 16 , When the energy needs of the wind power plant 12 provided amount of energy exceeds, carried out by the control unit 18 an automatic activation of the diesel generator 14 ,

In the approach for optimized power plant management presented here, the central aspect is that the control unit 18 the diesel generator 14 or other power generators activated in time if that of the wind power plant 12 provided electrical energy, for example due to a decreasing wind strength decreases. So far, the control unit activated 18 automatically the diesel generator 14 if a quantity of injected electrical energy for that reason is the energy requirement of the or each consumer 16 below. The mere activation of the diesel generator 14 But does not directly lead to the provision of the now in the power grid unit 10 lack of electrical energy, because the diesel generator 14 requires a certain startup and lead time until a power supply can take place. Consequently, the diesel generator would have to 14 be activated sufficiently in advance of that future time, starting from that of the wind power plant 12 provided electrical energy returns. A typical startup and lead time of a diesel generator 14 is in the range of 3-5 minutes. The automatic activation of a diesel generator 14 due to a decreasing amount of through the wind power plant 12 fed electrical energy through the control unit 18 must be done accordingly at least 3-5 minutes before the actual decrease in the amount of energy fed.

This includes the control unit 18 further functional units and accordingly is the control unit 18 in the illustration in 2 shown with further details. Accordingly, the control unit comprises 18 a fundamentally known and therefore not specifically illustrated LIDAR system 20 for the generation of laser beams and for the evaluation of the portions of the laser beams which are reflected in the atmosphere according to the LIDAR principle and on the other hand a signal conditioning unit 22 , The LIDAR system 20 generates continuously or regularly at least one LIDAR signal 24 as a measure of wind conditions determined at the sampling location and thus also as a measure of a change in the wind force expected within a forecast horizon or as a measure of within a forecast horizon at the location of the wind power plant 12 expected wind conditions. The forecast horizon is a time span and is based on a forecast horizon of up to ten minutes, in particular 1-5 minutes. By means of the LIDAR system 20 is in the necessary distances a sampling of the atmospheric conditions possible. The signal conditioning unit 22 generated, for example, based on the LIDAR signal 24 one or more binary signals 26 . 28 , for example, a first binary signal 26 when through the LIDAR signal 24 encoded within the prediction horizon expected wind strength exceeds a momentary wind strength by a predetermined or predeterminable threshold and a second binary signal 28 when through the LIDAR signal 24 coded, within the forecast horizon expected wind speed falls short of the current wind strength at least by the threshold. The second binary signal 28 or one based on the second binary signal 28 generated control signal 30 can then activate the diesel generator 14 be used. The control unit 18 or the signal conditioning unit 22 become a signal that encodes the current wind force 32 and a signal encoding the threshold 34 fed.

The signal conditioning unit 22 can - as previously explained - only the LIDAR signal 24 or control signals from the LIDAR system 20 evaluate or in addition to the LIDAR signal 24 also still - as before by means of 1 explained - a signal from the wind power plant 12 Evaluate, which encodes the current energy supply, so that due to a in the signal conditioning unit 22 correspondingly applied link the second binary signal 28 and the control signal 30 for controlling the diesel generator 14 whenever generated either by the LIDAR signal 24 encoded within the forecast horizon expected wind strength falls below a momentary wind strength around the threshold value or that by the wind power plant 12 received signal coded energy input under the energy needs of the consumer 16 or the consumer 16 falls.

• – bei einem zum Beispiel eine prozentuale Abnahme der Windstärke kodierenden LIDAR-Signal wird auf Basis des LIDAR-Signals oder einer auf Basis des LIDAR-Signal ermittelten erwarteten Energiemenge ein Zeitpunkt zum Aktivieren eines Dieselgenerators oder dergleichen ermittelt;
• – und/oder bei einem zum Beispiel eine prozentuale Abnahme der Windstärke kodierenden LIDAR-Signal wird auf Basis des LIDAR-Signals oder einer auf Basis des LIDAR-Signal ermittelten erwarteten Energiemenge eine notwendige Energiemenge und eine zum Einspeisen der notwendigen Energiemenge erforderliche Anzahl von zum Beispiel Dieselgeneratoren ermittelt und der oder jeder Dieselgenerator wird gestartet;
• – und/oder bei einem zum Beispiel eine prozentuale Abnahme der Windstärke kodierenden LIDAR-Signal werden auf Basis des LIDAR-Signals eine minimal notwendige Reserveenergiemenge und/oder eine minimal notwendige Schwungmasse und auf deren Basis bevorzugte Betriebsmodi (netzgeführt, netzbildend, aus) von in der Stromnetzeinheit 10 vorhandenen Umrichtern von zum Beispiel Photovoltaikanlagen oder Batterieanlagen ermittelt, um einen kostenoptimalen Betrieb bezüglich eines Hilfsenergieverbrauchs und/oder einer Komponentenalterung zu erreichen.

In addition to such, for explanatory purposes selected, simple output signals, such as the mentioned binary signals 26 . 28 , the control unit can 18 and especially the combination of LIDAR system 20 and signal conditioning unit 22 also basically provide arbitrarily complex output signals as a basis for power plant management. Individual examples are given below, but expressly without any claim to completeness:

• - For example, in a LIDAR signal encoding a percentage decrease in wind force, a time to activate a diesel generator or the like is determined based on the LIDAR signal or an expected amount of energy determined based on the LIDAR signal;
• - And / or at a, for example, a percentage decrease in the wind force encoding LIDAR signal is based on the LIDAR signal or on the basis of the LIDAR signal determined amount of energy required amount of energy and a number required for feeding the necessary amount of energy, for example Diesel generators determined and the or each diesel generator is started;
• - And / or at a, for example, a percentage decrease in the wind force encoding LIDAR signal based on the LIDAR signal a minimum amount of reserve energy required and / or a minimum necessary flywheel and based on their preferred operating modes (network-guided, network-forming, off) of in the power grid unit 10 existing converters of, for example, photovoltaic systems or battery systems determined in order to achieve a cost-optimal operation in terms of auxiliary power consumption and / or component aging.

The illustrations in 3 . 4 and 5 illustrate in principle the determination of a wind direction and a wind speed by means of a LIDAR system 20 , The representation in 3 shows in this respect the LIDAR system 20 , one from the LIDAR system 20 outgoing laser beam 36 and one on an aerosol 38 reflected beam 40 , From the term of the two rays 36 . 40 the distance to the reflection point can be determined. To determine the wind speed, a frequency change between the transmitted and the moving at a wind speed aerosol 38 reflected beam 36 . 40 evaluated (Doppler shift), as schematically simplified in the representation in 4 is shown. The representation in 5 illustrates that to determine a respective wind direction at different times, several measuring points, here three measuring points, and the respective reflected beams 40 to be viewed as. In fact, a vector is a measure of wind speed and wind direction.

The representation in 6 illustrates the approach proposed here in another way. Shown are seen from top to bottom four snapshots with changing wind conditions. On the left is the wind power plant 12 shown in the form of the turbines of individual wind turbines included. The wind conditions are due to the wind power plant 12 pointing block arrows shown. Below each snapshot is also a wind speed graph 42 shown, where on the abscissa the wind speed and on the ordinate the distance from the wind power plant 12 are worn off.

In the uppermost illustration in 6 are uniform wind conditions both at the site of the wind power plant 12 as shown throughout the range of coverage covered for illustration. The graph immediately below shows that the wind conditions change (smaller block arrows, falling flank in the graph 42 ). This change has no effect on the location of the wind power plant 12 and is not there yet recognizable. In the two other subsequent illustrations, the front moves with less wind power ever further to the wind power plant 12 to.

With the concept presented here succeeds by means of the LIDAR system 20 also a determination of one at a distance from the location of the wind power plant 12 prevailing wind strength. For example, the place taken into account corresponds to the place which is in the third of the four representations in 6 is marked by the vertical line. If there by means of the LIDAR system 20 the decrease in wind speed shown here can be detected by means of the control unit 18 automatic activation, for example, of the diesel generator 14 respectively. Based on the last two of the four representations in 6 is recognizable that at this time at the place of the wind power plant 12 still the prevailing wind conditions prevail, so that the energy feed into the power grid unit 10 has not changed while at the same time the automatically activated diesel generator 14 already starts up and thus at the latest by the time at which the reduced wind speed also leads to a reduced energy supply, to compensate for the reduced feed amount of the wind power plant 12 is available.

The representation in 7 shows one compared to the one in 2 shown supplemented control unit 18 , Thereafter, the control unit includes 18 , for example, as functionality of the signal conditioning unit 22 , an optimization unit 44 and from the optimization unit 44 processed additional data 46 or has access to such data. For the additional data 46 For example, this is historical data 46 , namely historical LIDAR signals on the one hand and historical, temporally correlated actual wind conditions and / or data based thereon, such as a momentary energy contribution or an energy contribution within a predetermined or predefinable time period etc. at the location of the wind power plant 12 on the other hand. The signal conditioning unit 22 then acts as a filter and generated by means of the optimization unit 44 and considering, for example, the historical data 46 a control signal 30 for activation, for example, of the diesel generator 14 , The data used 46 allow in a conventional manner training the optimization unit 44 or the signal conditioning unit 22 so that from the signal conditioning unit 22 generated output signals as training progressively represent the actual future conditions. Additionally or alternatively to data in the form of historical data 46 can be the optimization unit 44 and / or the signal conditioning unit 22 For example, consider data of the following type: Topology of the wind power plant environment 12 ; Vegetation in the vicinity of the wind power plant 12 , if necessary including seasonal changes; Influences of the wind power plant 12 included wind turbines, for example, depending on an orientation of the wind turbines; longer-term weather forecasts or weather models. Specifically, such data as topological data, vegetation data, year and daytime data, and / or data relative to prevailing weather conditions at each site of the wind turbine also allow extrapolation of the LIDAR signal 24 or another one from the LIDAR system 20 supplied signals. In a particular embodiment of the optimization unit 44 this includes an implementation of one of the data used in each case 46 , for example, historical data 46 , trained neural network.

Alternatively, the optimization unit comprises 44 For example, an implementation of a physical model, for example a model based on a simplified schematic representation in FIG 8th shown fluid dynamics simulation. The room around the wind power plant 12 or any of these included the wind turbines is calculated using a grid of cells 50 . 52 overdrawn. With the cells 50 . 52 they are cells 50 a meso-scaled model with, for example, a diameter / an edge length of 30 m as well as adjacent cells 52 a macro-scale model with, for example, a diameter / edge length of 3 km. For every cell 50 . 52 At least data on air pressure and wind speed are stored. Initial values for a prediction of air movement based on physical equations, namely partial differential equations of fluid dynamics, could be given by LIDAR measurements. The boundary conditions of the meso-scaled model can be obtained by information from a macroscopic weather model of the larger lattice cells 52 be determined. By prediction of the physical equations (partial differential equations of fluid dynamics) of the air movement, the effect of local disturbances / wind barriers, for example in the form of a piece of woodland, can now be considered 54 , a hill, buildings and the like, are calculated on the future wind speeds. In the example shown, the indicated forest 54 cause at least part of the forest 54 apt wind around the forest 54 would flow around like this by means of the drawn flow lines 56 is illustrated. At about the same pressure this would due to the conservation of mass to a significantly increased wind speed at the site of the wind power plant 12 to lead. Accordingly, changes in wind speed before such or similar obstruction would cause significantly larger / smaller changes in the location of the wind power plant 12 cause.

Based on LIDAR signals 24 , for example LIDAR signals 24 with respect to the indicated in the illustration along a horizontal line block arrows indicated places / cells 50 (Scanning locations, the direction and the length of the block arrows illustrates the wind direction or the wind speed), can use the model, the expected future wind conditions, ie within the forecast horizon at the location of the wind power plant 12 or at the location of individual or all of the wind power plant 12 covered wind turbines. The calculated expected wind conditions and / or the expected energy supply of the wind turbine or wind turbines then form the basis for an automatic power plant management. In the context of such an automatic power plant management, for example, the automatic activation of another energy generator takes place 14 or another energy producer 14 ,

While the invention has been further illustrated and described in detail by the exemplary embodiment, the invention is not limited by the disclosed or disclosed examples, and other variations can be derived therefrom by those skilled in the art without departing from the scope of the invention.

Individual aspects of the description submitted here can be briefly summarized as follows: The use of a LIDAR system is suggested 20 for short-term forecasting in a forecast horizon of 1-10 minutes, in particular a 1-5 minute forecast horizon, expected wind conditions at the location of a wind power plant 12 and power plant management based on the estimated expected wind conditions and a power plant management process of a power grid unit 10 , in which at least one wind power plant 12 feeds electrical energy and being to the power grid unit 10 at least one other energy producer 14 activated, with the following steps: By means of the LIDAR system 20 a short-term forecast takes place within a forecast horizon of 1-10 minutes, in particular a 1-5 minute forecast horizon, expected wind conditions at the location of the wind power plant 12 and on the basis of the calculated expected wind conditions, an activation of further energy producers takes place 14 or another energy producer 14 ,

Measurements based on a LIDAR system 20 are well suited for the power plant management described here, because the usual coverage of a LIDAR system 20 of up to 12 km, assuming a mean wind speed of 25 m / s, a forecast horizon of up to 16 minutes is allowed and thus the repeatedly mentioned critical time window for a timely activation of additional power generators 14 good coverage.

Significant advantages of the approach described here consist especially in the possibility of reducing ongoing operating costs (OPEX) because more energy producers 14 , such as diesel generators 14 or the like, not permanently, or at least do not need to run unnecessarily early and now each can be selectively connected as needed. In the same way, energy stores, such as batteries and the like, can be switched on as needed in order to specifically absorb, for example, increased feed quantities resulting from strong wind conditions or gusts of wind and to conserve them for periods of future infeed. Just as the current operating costs can be reduced with the approach presented here, the investment costs (CAPEX) can also be reduced, because on the basis of better predictions of the variation of the feed quantities of a wind power plant dependent on the wind conditions 12 is also an adapted to the actual compensation quantities design of other energy producers 14 , such as diesel generators 14 , and / or energy storage possible. An exactly matching diesel generator in this respect 14 is cheaper than a diesel generator, which for safety reasons larger than such a precisely matching diesel generator 14 is designed.

LIST OF REFERENCE NUMBERS

10

Mains unit

12

Wind power plant

14

Power generator / diesel generator

16

consumer

18

control unit

20

LIDAR system

22

Signal conditioning unit

24

LIDAR signal

26

(first) binary signal

28

(second) binary signal

30

control signal

32

Wind strength signal

34

threshold

36

Beam, laser beam

38

aerosol

40

reflected beam

42

Graph (the wind speed)

44

optimization unit

46

Data / historical data

48

(free)

50

Cell (of a meso-scaled model)

52

Cell (of a marko-scale model)

54

Wind barrier / woodland

56

flow line

Claims (10)

Use of a LIDAR system ( 20 ) for the short-term prediction of expected wind conditions at the location of a wind power plant ( 12 ) or several wind power plants in a forecast horizon of one minute to ten minutes, in particular a forecast horizon of one minute to five minutes, and power plant management based on the calculated expected wind conditions. Use of a LIDAR system ( 20 ) according to claim 1, wherein the power plant management on the basis of the calculated expected wind conditions, the automatic activation of another energy generator ( 14 ) or other energy producers ( 14 ) which together with at least one consumer ( 16 ) and the wind power plant ( 12 ) activatable to a wind power plant ( 12 ), the or each consumer ( 16 ) and the or each energy producer ( 14 ) comprehensive power grid unit ( 10 ) is connected or are. Use of a LIDAR system ( 20 ) according to one of claims 1 or 2, wherein the power plant management based on the determined expected wind conditions comprises the automatic determination of a necessary energy reserve. Power plant management process, whereby a power grid unit ( 10 ) at least one wind power plant ( 12 ) feeds electrical energy and to the power grid unit ( 10 ) at least one other energy producer ( 14 ) is activably connected, with the following steps: by means of a LIDAR system ( 20 ), a short-term forecast is made within a forecast horizon of one minute to ten minutes, in particular a forecast horizon of one minute to five minutes, expected wind conditions at the location of the wind power plant ( 12 ), on the basis of the calculated expected wind conditions, an activation of further energy generators ( 14 ) or another energy producer ( 14 ). Power plant management method according to claim 4, wherein the LIDAR system ( 20 ) a functional unit of a control unit ( 18 ) and another functional unit of the control unit ( 18 ) a LIDAR system ( 20 ) downstream signal conditioning unit ( 22 ) and wherein the signal conditioning unit ( 22 ) from one of the expected wind conditions at the location of the wind power plant ( 12 ) LIDAR signal ( 24 ) of the LIDAR system ( 20 ) a control signal ( 30 ) for activating further energy producers ( 14 ) or another energy producer ( 14 ) generated. Power plant management method according to claim 5, wherein the signal conditioning unit ( 22 ) acts as a filter and by means of an optimization unit ( 44 ) and historical data ( 46 ) a control signal ( 30 ) for activating further energy producers ( 14 ) or another energy producer ( 14 ) generated. Power plant management method according to claim 6, wherein the optimization unit ( 44 ) an implementation of one using the historical data ( 46 ) comprises trained neural network. Power plant management method according to claim 5, 6 or 7, wherein the signal conditioning unit ( 22 ) an optimization unit ( 44 ) with an implementation of a physical model of an environment of the wind power plant ( 12 ) and the Signal conditioning unit ( 22 ) based on the LIDAR signal ( 24 ) and the model a control signal ( 30 ) for activating further energy producers ( 14 ) or another energy producer ( 14 ) generated. A computer program comprising program code means for carrying out all the steps of any one of claims 4 to 8 when the computer program is stored on a control unit ( 18 ) is executed for power plant management. Control unit ( 18 ) for power plant management with a processing unit and a memory in which a computer program is loaded according to claim 9, which in the operation of the control unit ( 18 ) for power plant management by the processing unit.

Images