CN112613633A - Meteorological element prediction method and device for wind power plant - Google Patents
Meteorological element prediction method and device for wind power plant Download PDFInfo
- Publication number
- CN112613633A CN112613633A CN202011056321.5A CN202011056321A CN112613633A CN 112613633 A CN112613633 A CN 112613633A CN 202011056321 A CN202011056321 A CN 202011056321A CN 112613633 A CN112613633 A CN 112613633A
- Authority
- CN
- China
- Prior art keywords
- meteorological
- target
- point
- meteorological element
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000004927 fusion Effects 0.000 claims abstract description 34
- 238000005259 measurement Methods 0.000 claims abstract description 8
- 238000004590 computer program Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 230000002123 temporal effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Provided are a meteorological element prediction method and a device for a wind power plant, wherein the meteorological element prediction method comprises the following steps: obtaining point location predicted values of target meteorological elements of a plurality of representative point locations in a wind power plant in a preset time period; inputting the point location predicted value of the target meteorological element of each representative point location into a meteorological fusion prediction model to obtain a wind field predicted value of the target meteorological element in the preset time period; the representative point location is a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period. By adopting the meteorological element prediction method and device for the wind power plant in the exemplary embodiment of the invention, the prediction value of the target meteorological element which can represent the wind power plant most can be obtained through fusion calculation, so that the prediction accuracy of the target meteorological element of the wind power plant is improved.
Description
Technical Field
The present invention relates generally to the field of wind power generation, and more particularly, to a method and an apparatus for predicting meteorological parameters of a wind farm.
Background
The wind speed, wind direction and other meteorological elements of the numerical weather forecast can be converted into the output power forecast of the wind power plant and the photovoltaic through the forecasting algorithm, so that the accurate forecast of the numerical weather forecast can provide important decision support for power dispatching and is one of important decision factors of the new energy power generation power forecasting precision.
At present, meteorological prediction data adopted by power prediction of a wind power plant is a predicted value of wind speed of a central longitude and latitude of the wind power plant calculated by numerical weather prediction, namely, the wind speed of the central longitude and latitude represents the average wind speed of the whole plant, so that a power model is established. However, the occupied area of the wind power plant is usually large, the point positions of the wind generation sets are not uniformly distributed (determined by fluid calculation and not equidistantly distributed), and meanwhile, the wind power plant is mostly built in a mountainous area with complex topography and topography, and the wind speed changes violently in the horizontal direction, so that the average wind speed of the whole wind power plant represented by the wind speed of the center longitude and latitude of the wind power plant brings large errors to power prediction.
Disclosure of Invention
An object of an exemplary embodiment of the present invention is to provide a method and apparatus for predicting meteorological elements of a wind farm, so as to overcome at least one of the above disadvantages.
In one general aspect, there is provided a meteorological element prediction method for a wind farm, the meteorological element prediction method comprising: obtaining point location predicted values of target meteorological elements of a plurality of representative point locations in a wind power plant in a preset time period; inputting the point location predicted value of the target meteorological element of each representative point location into a meteorological fusion prediction model to obtain a wind field predicted value of the target meteorological element in the preset time period; the representative point location is a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period.
Optionally, the step of obtaining the point location prediction values of the target meteorological elements of a plurality of representative point locations in the wind farm in a predetermined time period may include: and extracting point location predicted values of the target meteorological elements of the plurality of representative point locations in the preset time period from meteorological forecast data provided by a meteorological source.
Optionally, the meteorological fusion prediction model may be trained by: acquiring point location predicted values of the target meteorological elements of the plurality of representative point locations within a preset time period; acquiring an observed value of a target meteorological element of the wind power plant in the preset time period; and taking the obtained point location prediction values of the plurality of representative point locations as the input of a meteorological fusion prediction model, taking the obtained observation value of the wind power plant as the output of the meteorological fusion prediction model, and training the meteorological fusion prediction model.
Alternatively, the plurality of representative point locations may include a plurality of point locations in a plurality of time units, which may be obtained by dividing a year according to climate change conditions of an area in which the wind farm is located, wherein the plurality of point locations in each time unit may be determined by: acquiring historical weather forecast data of a plurality of weather grid points in a preset range with a target point as a center in the time unit, wherein the target point is any one of all the weather grid points included in a target area, and the target area covers an area where a wind power plant is located; determining the similarity index of the historical meteorological forecast data of the meteorological grid point and the historical observation data of the meteorological grid point aiming at each meteorological grid point in a preset range; searching minimum value grid points of similar indexes from the plurality of meteorological grid points in a preset range; and determining the minimum value lattice points of which the similar indexes are smaller than those of the target point positions in the searched minimum value lattice points as the point positions in the time unit.
Alternatively, a plurality of representative points within a year of the wind farm may be obtained by: the plurality of representative point locations are obtained by performing deduplication processing on each point location under all time units in one year.
Alternatively, the predetermined range at each time unit may be determined by: acquiring a plurality of contour maps centering on the target point under the time unit; searching candidate meteorological grid points which are closest to the position of the highest value of the historical meteorological forecast data of the target meteorological element from the contour map aiming at each contour map, and determining the distance between the candidate meteorological grid points and the target point; selecting a maximum distance from the distances determined for the plurality of contour maps; determining the predetermined range centered on a target point position based on the maximum distance.
Alternatively, the number of contour maps per time unit may be determined based on the duration of the time unit and the time resolution of the historical weather forecast data, and/or the distance determined for each contour map may refer to the distance in the longitude or latitude direction between the candidate weather grid point and the target point.
In another general aspect, there is provided a meteorological element prediction apparatus for a wind farm, the meteorological element prediction apparatus comprising: the data acquisition module is used for acquiring point location predicted values of target meteorological elements of a plurality of representative point locations in the wind power plant in a preset time period; the meteorological prediction module is used for inputting the acquired point location prediction value of the target meteorological element of each representative point location into a meteorological fusion prediction model to obtain a wind field prediction value of the target meteorological element in the preset time period; the representative point location is a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period.
In another general aspect, there is provided a controller comprising: a processor; a memory for storing a computer program which, when executed by the processor, implements the above-described method of meteorological element prediction for a wind farm.
In another general aspect, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of meteorological element prediction for a wind farm as described above.
By adopting the meteorological element prediction method and device for the wind power plant in the exemplary embodiment of the invention, the prediction value of the target meteorological element which can represent the wind power plant most can be obtained through fusion calculation, so that the prediction accuracy of the target meteorological element of the wind power plant is improved.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings which illustrate exemplary embodiments.
FIG. 1 shows a flow chart of a meteorological element prediction method for a wind farm according to an exemplary embodiment of the present invention;
FIG. 2 shows a flowchart of the steps of determining a plurality of point locations per unit of time according to an example embodiment of the present invention;
FIG. 3 shows a schematic diagram of a contour plot centered at a target point in time units according to an exemplary embodiment of the present invention;
FIG. 4 illustrates a schematic diagram of a plurality of point locations in units of time according to an exemplary embodiment of the present invention;
FIG. 5 shows a flowchart of the steps of training a weather fusion prediction model, according to an exemplary embodiment of the present invention;
FIG. 6 is a schematic diagram showing the comparison of the observation values with the weather forecast data of the target meteorological element;
FIG. 7 is a schematic diagram illustrating comparison of weather forecast data and observations of a target meteorological element, according to an exemplary embodiment of the present invention;
FIG. 8 shows a block diagram of a meteorological element prediction apparatus for a wind farm according to an exemplary embodiment of the present invention;
fig. 9 illustrates a block diagram of a controller according to an exemplary embodiment of the present invention.
Detailed Description
Various example embodiments will now be described more fully with reference to the accompanying drawings, in which some example embodiments are shown.
FIG. 1 shows a flow chart of a meteorological element prediction method for a wind farm according to an exemplary embodiment of the present invention.
Referring to fig. 1, in step S10, point location prediction values of target meteorological elements at a predetermined time period for a plurality of representative point locations in a wind farm are acquired.
Here, the predetermined period of time may refer to a period of time in the future after the current time. By way of example, meteorological elements may include, but are not limited to, at least one of: the target meteorological element can be one of the meteorological elements mentioned above, such as temperature, humidity, wind speed, wind direction, and atmospheric pressure at different heights.
In an alternative example, the predicted point location values of the target meteorological elements of a plurality of representative point locations in a predetermined time period can be obtained through a numerical weather forecast, for example, the numerical weather forecast can refer to a method for performing numerical calculation through a large-scale computer under certain initial value and boundary value conditions according to actual conditions of the atmosphere, solving a system of equations describing hydrodynamics and thermodynamics of a weather evolution process, and predicting the atmospheric motion state and weather phenomena in a future certain time period, namely, a means for making the weather forecast by using the current weather conditions as input data.
That is, point location prediction values representing target meteorological elements of a plurality of point locations over a predetermined period of time may be extracted from weather forecast data (i.e., numerical weather forecast) provided from an arbitrary weather source. Here, the point location prediction value may refer to a prediction value of the target meteorological element at the representative point location for a predetermined period of time. In one example, the point location prediction value of the target meteorological element at a predetermined time (i.e., a time in the future) may also be obtained.
Here, the numerical weather forecast is grid data, the area where the wind farm is located includes a plurality of weather grid points, that is, the numerical weather forecast of the area where the wind farm is located can be obtained, a part of the weather grid points is selected from the plurality of weather grid points included in the area where the wind farm is located as a plurality of representative points, and point prediction values of target weather elements of the plurality of representative points are extracted from the obtained numerical weather forecast of the area where the wind farm is located.
In step S20, the wind field prediction value of the target meteorological element for the predetermined period of time is obtained by inputting the point location prediction value of the target meteorological element for each of the obtained representative point locations into the meteorological fusion prediction model.
Here, the representative point location may refer to a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period, or may refer to a point location representing the target meteorological element of the wind farm in different solar terms (in each predetermined time period). That is, in the meteorological element prediction method for a wind farm according to the exemplary embodiment of the present invention, the points at which the predicted value and the actual measurement value are closest to each other for each of the different solar terms are determined, and the determined points at each solar term are integrated to obtain a plurality of representative points in the wind farm.
In an alternative example, the plurality of representative point locations may include a plurality of point locations under a plurality of time units. In one example, a plurality of time units may be obtained by dividing a year according to climate change conditions of an area in which the wind farm is located.
For example, the more frequent the climate change of the area where the wind farm is located, the shorter the duration of the time unit, and here, the climate change situation may be determined based on the climate characteristics of the area where the wind farm is located. The duration of each time unit may be the same or different.
The first predetermined period of time (e.g., 10 days) may be determined as a time unit if the climate of the area in which the wind farm is located changes relatively frequently (e.g., the rate of climate change is less than a first set point), and the second predetermined period of time (e.g., one quarter) may be determined as a time unit if the climate of the area in which the wind farm is located changes infrequently (e.g., the rate of climate change is greater than a second set point). As an example, the second set point may be greater than or equal to the first set point, with the second predetermined length of time being greater than the first predetermined length of time.
In addition to the above-described manner of determining the duration of the time units based on the climate change condition of the area in which the wind farm is located, the duration of each time unit may be set to a default value, for example, a third predetermined duration (e.g., monthly) may be determined as the duration of each time unit.
Alternatively, a plurality of point locations under each time unit in the year may be determined, and a plurality of representative point locations in the wind farm may be obtained by performing deduplication processing on each point location under all time units in the year.
The process of determining a plurality of point locations at any one time unit is described below with reference to fig. 2. That is, a plurality of dots in each time unit can be obtained for each time unit in the manner shown in fig. 2.
Fig. 2 shows a flowchart of the steps of determining a plurality of point locations per time unit according to an exemplary embodiment of the present invention.
Referring to fig. 2, in step S101, a plurality of contour maps centered on a target point location at any one time unit are acquired.
Here, the target point may be any one of all the meteorological points included in the target area, the target area covers an area where the wind farm is located, and the contour map may refer to a contourr map of the target meteorological elements.
As an example, the number of contour plots at each time unit may be determined according to the duration of the time unit and the temporal resolution of the historical weather forecast data.
For example, the number of contour plots at any unit of time can be determined using the following equation:
a is time resolution x time unit
Where a represents the number of contour plots at any time unit.
In step S102, the distance between the candidate weather grid point and the target point in each contour map is determined.
For example, for each contour map, the weather grid candidate point closest to the position of the highest value of the historical weather forecast data of the target weather element is searched from the contour map, that is, the weather grid point closest to the position of the highest value of the historical weather forecast data of the target weather element in the contour map is determined as the weather grid candidate point, and the distance between the weather grid candidate point and the target point is determined.
Fig. 3 shows a schematic diagram of a contour map centered at a target point in time units according to an exemplary embodiment of the present invention.
As shown in fig. 3, point a represents a target point, and point B represents a candidate weather grid point that is searched in the isobaric chart and is closest to the position where the highest value of the historical weather forecast data of the target weather element is located.
In a preferred example, for each contour map, the distance in the longitudinal direction and the latitudinal direction between the candidate weather lattice point and the target point location may be determined, and the maximum value among the distance in the longitudinal direction and the distance in the latitudinal direction is determined as the distance between the candidate weather lattice point and the target point location in the contour map.
In an example, the distance between the candidate weather grid point and the target point may also be converted to the grid point number H at the spatial resolution of the current numerical weather forecast.
Returning to fig. 2, in step S103, the maximum distance among the distances determined for the plurality of contour maps is selected.
In step S104, a predetermined range centered on the target point is determined based on the maximum distance.
For example, assuming that the maximum distance between the candidate weather grid point and the target point location is the grid point number Hmax, the predetermined range may refer to a weather forecast result that Hmax weather grid points corresponding to the maximum distance are obtained outward with the target point location as the center, that is, 2Hmax × 2Hmax is obtained as 4Hmax2Forecasting results of the weather grid points.
In step S105, historical weather forecast data for a plurality of weather grid points within a predetermined range centered on the target point is acquired in the time unit.
In step S106, for each weather grid point in a predetermined range, a similarity index between the historical weather forecast data of the weather grid point and the historical observation data of the weather grid point is determined.
Here, for each meteorological site within a predetermined range, historical observation data of the meteorological site is acquired, the historical observation data being measured data at the same time period as the acquired meteorological forecast data.
In one example, the time resolution of the historical weather forecast data may be interpolated to be the same as the historical observation data, and the weather elements and altitudes of the historical weather forecast data are interpolated from the historical observation data.
By way of example, the similarity indicator may include, but is not limited to, at least one of: RMSE (root mean square error), R (correlation coefficient), MAE (mean absolute error). Here, the method for determining the above-mentioned similarity index is common knowledge in the art, and the disclosure will not be repeated for this part.
In step S107, minimum-value grid points of the similarity index are searched from a plurality of weather grid points within a predetermined range.
For example, if the similar indicators of a weather grid point are all smaller than the similar indicators of other weather grid points located around the weather grid point, the weather grid point is determined as the minimum grid point of the similar indicators, and if the similar indicator of a weather grid point is greater than or equal to the similar indicators of at least one other weather grid point located around the weather grid point, the weather grid point does not belong to the minimum grid point.
Here, if the minimum value lattice point of the similarity index is not searched within the predetermined range, the area covered by the predetermined range may be expanded, and steps S105 to S107 may be repeatedly performed to search for the minimum value lattice point.
In step S108, the minimum-value lattice point of the similar index whose similar index is smaller than that of the target point location among the searched minimum-value lattice points is determined as a point location in the time unit.
For example, historical meteorological forecast data and historical observation data of the target point location may be obtained, similarity indexes of the historical meteorological forecast data and the historical observation data of the target point location are determined, the similarity indexes of the minimum value grid points searched in a predetermined range are compared with the similarity indexes of the target point location, and the minimum value grid point of which the similarity index is smaller than the similarity index of the target point location is determined as the best point location for representing the target meteorological element of the wind farm in the time unit.
The method for predicting meteorological elements of a wind farm according to an exemplary embodiment of the invention is described as an example.
In this example, it is assumed that weather forecast data of a one-year-history european numerical forecast center of a certain wind farm is acquired, the spatial resolution of the weather forecast data being 0.1 ° × 0.1 ° (degrees), and the temporal resolution being 1 hour.
The method comprises the steps of obtaining the machine head wind speed of each wind turbine generator in the wind power plant in one year, carrying out arithmetic average on the machine head wind speed to obtain the average machine head wind speed of the wind power plant, and using the average machine head wind speed as an observed value of the wind speed of the wind power plant.
The U component (component of wind speed in the latitudinal direction) and the V component (component of wind speed in the longitudinal direction) of 10 meters and 100 meters in the weather forecast data are extracted, the wind speed at the height of 70 meters is obtained by interpolation, and further, the time resolution can be linearly interpolated to 15 minutes so as to be consistent with the time resolution of the observed value of the wind speed.
Assuming that the duration of the time unit is a month, the processing can be performed for each month.
A contourr diagram of a target meteorological element (i.e., wind speed) centered on a target point is drawn, and assuming that the range of the contourr diagram is 10 meteorological points in size, i.e., 110 (kilometers) × 110 (kilometers) in size, a total of 4 × 24 × 30 contourr diagrams are obtained.
Taking June as an example, in each contour graph in June, a weather grid point candidate (30, 46) closest to the position of the highest value of the target weather element is searched, the maximum distance between the grid point and the nearest weather grid point (32, 51) of the target point is recorded, and the distance is converted into the grid point number 5 by the spatial resolution of the current numerical weather forecast.
In the above way, different months are processed, and the following results are obtained: one month: lattice number 5, february: lattice number 8, March: the number of lattice points is 4, April: 7, May points: lattice number 4, june: the number of lattice points is 5, July: lattice number 8, august: grid number 10, september: the number of lattice points is 3, October: lattice number 4, november: the number of lattice points is 5, December: the number of lattice points is 2.
The point locations at each month are determined for each month. For example, in june, taking the target point location as the center, the weather forecast results of 5 weather grid points whose numerical weather forecast is closest to the location of the target point location are obtained outwards, that is, the forecast results of 100 weather grid points are obtained.
And for each meteorological grid point, determining a similar index of the wind speed predicted value and the observed value of the meteorological grid point, searching a minimum grid point (such as the point circled in fig. 4) of the similar index, and determining the minimum grid point, in the minimum grid point, of the similar index, of which the similar index is smaller than that of the target point, as the optimal point under the time unit. Taking the example shown in fig. 4, june has three optimal points of (41.8 ° N,106.1 ° E), (41.9 ° N,106.5 ° E), and (41.4 ° N, 106.8 ° E), respectively.
By the above method, all months are processed, and the following 7 representative points are obtained by performing the deduplication processing:
1：41.5°N,106.4°E
2：41.4°N,106.4°E
3：41.5°N,106.4°E
4：41.8°N,106.1°E
5：41.3°N,106.8°E
6：41.9°N,106.5°E
7：41.4°N,106.8°E
and extracting the point location predicted values of the target meteorological elements of the 7 representative point locations from the meteorological forecast data each time the target meteorological elements of the wind power plant are predicted, and inputting the point location predicted values into a meteorological fusion prediction model to obtain the wind field predicted value of the target meteorological elements of the wind power plant after fusion.
In an exemplary embodiment of the present invention, the meteorological fusion prediction model for predicting the wind farm predicted value of the target meteorological element of the wind farm at the predetermined time is a pre-trained prediction model, and a process of training the meteorological fusion prediction model is described below with reference to fig. 5.
FIG. 5 shows a flowchart of the steps of training a meteorological fusion prediction model, according to an example embodiment of the present invention.
Referring to fig. 5, in step S30, point location prediction values of target meteorological elements for a plurality of representative point locations within a predetermined period of time are acquired.
In an example, the predetermined time period for performing the meteorological fusion prediction model training may refer to a time period before the predetermined time period (or the predetermined time) in step S10. For example, when predicting the mth day of the wind farm, all available machine learning models can be used for training and modeling by using data of m-32 to m-2 days (the training duration can be adjusted according to local conditions and is default to 30 days), then the target meteorological elements of the mth-1 day of the wind farm are predicted, and the algorithm with the optimal effect is used as the algorithm for predicting the current meteorological elements of the mth day of the wind farm.
In step S40, an observed value of a target meteorological element of the wind farm over a predetermined period of time is acquired.
In step S50, the point prediction values of the plurality of representative points obtained are input as a weather fusion prediction model, and the observation values of the wind farm obtained are output as a weather fusion prediction model, and the weather fusion prediction model is trained.
Fig. 6 is a schematic diagram showing comparison between observation values and weather forecast data of a conventional target weather element. FIG. 7 is a schematic diagram illustrating comparison of weather forecast data and observations of a target meteorological element, according to an exemplary embodiment of the present invention. In the figure, the abscissa is the date, the ordinate is the value of the target meteorological element, the curve 1 is the observed value curve of the target meteorological element, the curve 2 is the meteorological forecast data curve of the target meteorological element obtained by the existing meteorological prediction method, and the curve 3 is the meteorological forecast data curve of the target meteorological element obtained by the meteorological element prediction method according to the exemplary embodiment of the present invention.
As can be seen from fig. 6 and 7, the correlation optimization result (R ═ 0.49) of fig. 7 is better than (R ═ 0.40) of fig. 6, the theoretical power accuracy optimization result (TPA ═ 0.6948) of fig. 7 is better than (TPA ═ 0.6940) of fig. 6, and the root mean square error optimization result (RMSE ═ 2.63) of fig. 7 is better than (RMSE ═ 2.71) of fig. 6.
FIG. 8 shows a block diagram of a meteorological element prediction apparatus for a wind farm according to an exemplary embodiment of the present invention.
As shown in fig. 8, a meteorological element prediction apparatus 100 of a wind farm according to an exemplary embodiment of the present invention includes: a data acquisition module 101 and an weather prediction module 102.
Specifically, the data acquisition module 101 acquires point location prediction values of target meteorological elements of a plurality of representative point locations in the wind farm over a predetermined period of time.
Here, the predetermined period of time may refer to a period of time in the future after the current time. By way of example, meteorological elements may include, but are not limited to, at least one of: the target meteorological element can be one of the meteorological elements mentioned above, such as temperature, humidity, wind speed, wind direction, and atmospheric pressure at different heights.
That is, point location prediction values representing target meteorological elements of a plurality of point locations over a predetermined period of time may be extracted from weather forecast data (i.e., numerical weather forecast) provided from an arbitrary weather source. Here, the point location prediction value may refer to a prediction value of the target meteorological element at the representative point location for a predetermined period of time.
The meteorological prediction module 102 obtains a wind field prediction value of the target meteorological element in a predetermined time period by inputting the point location prediction value of the target meteorological element of each representative point location into the meteorological fusion prediction model.
Here, the representative point location may refer to a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period, or may refer to a point location representing the target meteorological element of the wind farm under different solar terms. That is, the data acquisition module 101 may determine, for different solar terms, the point locations with the predicted values and the measured values closest to each other under each solar term, respectively, and synthesize the determined point locations under each solar term to obtain a plurality of representative point locations in the wind farm.
In an alternative example, the plurality of representative point locations may include a plurality of point locations under a plurality of time units. In one example, the data acquisition module 101 may obtain a plurality of time units by dividing a year according to climate change conditions of an area where the wind farm is located.
For example, the more frequent the climate change of the area where the wind farm is located, the shorter the duration of the time unit, and here, the climate change situation may be determined based on the climate characteristics of the area where the wind farm is located. The duration of each time unit may be the same or different.
The data acquisition module 101 may determine a plurality of point locations in each time unit in a year, and perform deduplication processing on each point location in all time units in a year to obtain a plurality of representative point locations in the wind farm.
For example, the data acquisition module 101 may determine a plurality of point locations at any one of each unit of time in the following manner.
The method comprises the steps of obtaining a plurality of contour maps which take a target point location as a center in any time unit, determining the distance between a candidate weather grid point and the target point location in each contour map, selecting the maximum distance in the distances determined for the plurality of contour maps, and determining a preset range which takes the target point location as the center based on the maximum distance.
Acquiring historical weather forecast data of a plurality of weather grid points in a preset range with a target point as a center in the time unit, determining similar indexes of the historical weather forecast data of the weather grid points and historical observation data of the weather grid points aiming at each weather grid point in the preset range, searching minimum value grid points of the similar indexes from the plurality of weather grid points in the preset range, and determining the minimum value grid points of the similar indexes, of the searched minimum value grid points, of which the similar indexes are smaller than those of the target point, as the point positions in the time unit.
As an example, the number of contour plots at each time unit may be determined according to the duration of the time unit and the temporal resolution of the historical weather forecast data.
In one example, for each contour map, the candidate weather grid point closest to the position of the highest value of the historical weather forecast data of the target weather element is searched from the contour map, that is, the weather grid point closest to the position of the highest value of the historical weather forecast data of the target weather element in the contour map is determined as the candidate weather grid point, and the distance between the candidate weather grid point and the target point is determined.
In an exemplary embodiment of the invention, the meteorological fusion prediction model for predicting the wind farm predicted value of the target meteorological element of the wind farm at the preset moment is a pre-trained prediction model. In one example, the meteorological element prediction apparatus 100 of a wind farm according to an exemplary embodiment of the present invention may further include: and the model training module 103 is used for training the meteorological fusion prediction model.
For example, the model training module 103 obtains the point prediction values of the target meteorological elements of a plurality of representative points in a predetermined time period, obtains the observation values of the target meteorological elements of the wind farm in the predetermined time period, uses the obtained point prediction values of the plurality of representative points as the input of the meteorological fusion prediction model, uses the obtained observation values of the wind farm as the output of the meteorological fusion prediction model, and trains the meteorological fusion prediction model. Here, the predetermined period of time for performing the weather fusion prediction model training may refer to a period of time before the predetermined period of time for performing the target meteorological element prediction.
Fig. 9 illustrates a block diagram of a controller according to an exemplary embodiment of the present invention.
As shown in fig. 9, the controller 200 according to an exemplary embodiment of the present invention includes: a processor 201 and a memory 202.
In particular, the memory 202 is used for storing a computer program which, when being executed by the processor 201, implements the above-mentioned method for meteorological element prediction of a wind farm.
Here, the meteorological element prediction method for a wind farm shown in fig. 1 may be executed in the processor 201 shown in fig. 9. That is, each module shown in fig. 8 may be implemented by a general-purpose hardware processor such as a digital signal processor or a field programmable gate array, may be implemented by a special-purpose hardware processor such as a special chip, or may be implemented entirely in software by a computer program, for example, may be implemented as each module in the processor 201 shown in fig. 9.
There is also provided, in accordance with an exemplary embodiment of the present invention, a computer-readable storage medium storing a computer program. The computer readable storage medium stores a computer program which, when executed by a processor, causes the processor to perform the above-described method of meteorological element prediction for a wind farm. The computer readable recording medium is any data storage device that can store data read by a computer system. Examples of the computer-readable recording medium include: read-only memory, random access memory, read-only optical disks, magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data transmission through the internet via wired or wireless transmission paths).
According to the meteorological element prediction method and device for the wind power plant, a plurality of representative point positions capable of representing the whole target meteorological element of the wind power plant can be selected according to different seasons, the predicted value capable of representing the target meteorological element of the wind power plant can be calculated through algorithm fusion such as machine learning, the target meteorological element is taken as the wind speed, the wind speed predicted value obtained through the mode can be taken as the input of power prediction, and therefore the power prediction accuracy is improved.
The method and the device for predicting the meteorological elements of the wind power plant according to the exemplary embodiment of the invention are suitable for the wind power plant with wide distribution range and complex terrain, and can correct the spatial forecast deviation of the numerical weather forecast.
In addition, the method and the device for predicting meteorological elements of the wind farm in the exemplary embodiment of the invention optimize the method for using the numerical weather forecast result (i.e. by selecting a plurality of representative points), and can optimize the weather forecast result (any meteorological element) of any point forecast.
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims.
Claims (10)
1. A meteorological element prediction method for a wind farm is characterized by comprising the following steps:
obtaining point location predicted values of target meteorological elements of a plurality of representative point locations in a wind power plant in a preset time period;
inputting the point location predicted value of the target meteorological element of each representative point location into a meteorological fusion prediction model to obtain a wind field predicted value of the target meteorological element in the preset time period;
the representative point location is a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period.
2. The meteorological element prediction method according to claim 1, wherein the step of obtaining a point location prediction value of the target meteorological element for a plurality of representative point locations in the wind farm over a predetermined period of time includes:
and extracting point location predicted values of the target meteorological elements of the plurality of representative point locations in the preset time period from meteorological forecast data provided by a meteorological source.
3. The meteorological element prediction method according to claim 1, wherein the meteorological fusion prediction model is trained by:
acquiring point location predicted values of the target meteorological elements of the plurality of representative point locations within a preset time period;
acquiring an observed value of a target meteorological element of the wind power plant in the preset time period;
and taking the obtained point location prediction values of the plurality of representative point locations as the input of a meteorological fusion prediction model, taking the obtained observation value of the wind power plant as the output of the meteorological fusion prediction model, and training the meteorological fusion prediction model.
4. The meteorological element prediction method according to claim 1, wherein the plurality of representative points include a plurality of points in a plurality of time units obtained by dividing a year according to climate change conditions of an area where a wind farm is located,
wherein the plurality of points under each time unit is determined by:
acquiring historical weather forecast data of a plurality of weather grid points in a preset range with a target point as a center in the time unit, wherein the target point is any one of all the weather grid points included in a target area, and the target area covers an area where a wind power plant is located;
for each meteorological grid point in the preset range, determining a similarity index of historical meteorological forecast data of the meteorological grid point and historical observation data of the meteorological grid point;
searching minimum grid points of similar indexes from the plurality of meteorological grid points in the preset range;
and determining the minimum value lattice points of which the similar indexes are smaller than those of the target point positions in the searched minimum value lattice points as the point positions in the time unit.
5. The meteorological element prediction method according to claim 4, wherein the plurality of representative points in the wind farm over the year are obtained by:
the plurality of representative point locations are obtained by performing deduplication processing on each point location under all time units in one year.
6. The meteorological element prediction method according to claim 4, wherein the predetermined range at each time unit is determined by:
acquiring a plurality of contour maps centering on the target point under the time unit;
searching candidate meteorological grid points which are closest to the position of the highest value of the historical meteorological forecast data of the target meteorological element from the contour map aiming at each contour map, and determining the distance between the candidate meteorological grid points and the target point;
selecting a maximum distance from the distances determined for the plurality of contour maps;
determining the predetermined range centered on a target point position based on the maximum distance.
7. The meteorological element prediction method according to claim 6, wherein the number of contour maps at each time unit is determined based on a time duration of the time unit and a time resolution of the historical meteorological forecast data,
and/or, the distance determined for each contour map refers to the distance in the longitudinal direction or the latitudinal direction between the candidate weather grid point and the target point.
8. A meteorological element prediction device for a wind farm, comprising:
the data acquisition module is used for acquiring point location predicted values of target meteorological elements of a plurality of representative point locations in the wind power plant in a preset time period;
the meteorological prediction module is used for inputting the acquired point location prediction value of the target meteorological element of each representative point location into a meteorological fusion prediction model to obtain a wind field prediction value of the target meteorological element in the preset time period;
the representative point location is a point location where a predicted value and an actual measurement value of the target meteorological element are closest to each other in each predetermined time period.
9. A controller, comprising:
a processor;
memory for storing a computer program which, when executed by the processor, implements a method of meteorological element prediction for a wind farm according to any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that the computer program, when being executed by a processor, carries out a method of meteorological element prediction for a wind farm according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011056321.5A CN112613633B (en) | 2020-09-30 | 2020-09-30 | Weather element prediction method and device for wind power plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011056321.5A CN112613633B (en) | 2020-09-30 | 2020-09-30 | Weather element prediction method and device for wind power plant |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112613633A true CN112613633A (en) | 2021-04-06 |
| CN112613633B CN112613633B (en) | 2024-10-29 |
Family
ID=75224411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011056321.5A Active CN112613633B (en) | 2020-09-30 | 2020-09-30 | Weather element prediction method and device for wind power plant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112613633B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724280A (en) * | 2021-09-15 | 2021-11-30 | 南京信息工程大学 | Automatic identification method for ground weather map high-voltage system |
| CN115905997A (en) * | 2022-10-28 | 2023-04-04 | 安徽省国家电投和新电力技术研究有限公司 | Wind turbine generator meteorological disaster early warning method and system based on prediction deviation optimization |
| CN116955939A (en) * | 2023-09-20 | 2023-10-27 | 南京气象科技创新研究院 | Meteorological element structural feature forecast error quantization expression method based on graph similarity |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140344209A1 (en) * | 2010-08-23 | 2014-11-20 | Institute Of Nuclear Energy Research, Atomic Energy Council, Executive Yuan | Wind energy forecasting method with extreme wind speed prediction function |
| CN104616087A (en) * | 2015-03-11 | 2015-05-13 | 国家电网公司 | Wind power plant output power forecasting method taking weather factors and wind speed continuity into consideration |
| CN104699936A (en) * | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN106897481A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新系统控制技术有限公司 | The wind power integrated prediction method and device of utilization space multiple spot power prediction feature |
| CN106897790A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新系统控制技术有限公司 | A kind of wind power prediction method and device of selection Optimal predictor lattice point |
| CN108520269A (en) * | 2018-03-10 | 2018-09-11 | 华北电力大学（保定） | A kind of wind speed forecasting method and forecasting wind speed system |
| CN109583588A (en) * | 2018-12-17 | 2019-04-05 | 华北电力大学（保定） | A kind of short-term wind speed forecasting method and system |
| CN110263998A (en) * | 2019-06-19 | 2019-09-20 | 大唐（赤峰）新能源有限公司 | Multi-source numerical weather forecast set bilayer modification method |
| CN110705115A (en) * | 2019-10-10 | 2020-01-17 | 北京北科融智云计算科技有限公司 | Meteorological forecasting method and system based on deep belief network |
| CN111325376A (en) * | 2018-12-14 | 2020-06-23 | 北京金风科创风电设备有限公司 | Wind speed prediction method and device |
| CN111709569A (en) * | 2020-06-11 | 2020-09-25 | 国网冀北电力有限公司电力科学研究院 | Method and device for predicting and correcting output power of wind power plant |
-
2020
- 2020-09-30 CN CN202011056321.5A patent/CN112613633B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140344209A1 (en) * | 2010-08-23 | 2014-11-20 | Institute Of Nuclear Energy Research, Atomic Energy Council, Executive Yuan | Wind energy forecasting method with extreme wind speed prediction function |
| CN104699936A (en) * | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN104616087A (en) * | 2015-03-11 | 2015-05-13 | 国家电网公司 | Wind power plant output power forecasting method taking weather factors and wind speed continuity into consideration |
| CN106897481A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新系统控制技术有限公司 | The wind power integrated prediction method and device of utilization space multiple spot power prediction feature |
| CN106897790A (en) * | 2017-01-03 | 2017-06-27 | 北京国能日新系统控制技术有限公司 | A kind of wind power prediction method and device of selection Optimal predictor lattice point |
| CN108520269A (en) * | 2018-03-10 | 2018-09-11 | 华北电力大学（保定） | A kind of wind speed forecasting method and forecasting wind speed system |
| CN111325376A (en) * | 2018-12-14 | 2020-06-23 | 北京金风科创风电设备有限公司 | Wind speed prediction method and device |
| CN109583588A (en) * | 2018-12-17 | 2019-04-05 | 华北电力大学（保定） | A kind of short-term wind speed forecasting method and system |
| CN110263998A (en) * | 2019-06-19 | 2019-09-20 | 大唐（赤峰）新能源有限公司 | Multi-source numerical weather forecast set bilayer modification method |
| CN110705115A (en) * | 2019-10-10 | 2020-01-17 | 北京北科融智云计算科技有限公司 | Meteorological forecasting method and system based on deep belief network |
| CN111709569A (en) * | 2020-06-11 | 2020-09-25 | 国网冀北电力有限公司电力科学研究院 | Method and device for predicting and correcting output power of wind power plant |
Non-Patent Citations (3)
| Title |
|---|
| 徐晶晶;胡非;肖子牛;李军;: "风能模式预报的相似误差订正", 应用气象学报, no. 06, 15 December 2013 (2013-12-15) * |
| 白永祥;房大中;侯佑华;朱长胜;: "内蒙古电网区域风电功率预测系统", 电网技术, no. 10, 5 October 2010 (2010-10-05) * |
| 高金兰;李豪;段玉波;王宏建;: "基于Stacking多GRU模型的风电场短期功率预测", 吉林大学学报(信息科学版), no. 04, 15 July 2020 (2020-07-15) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724280A (en) * | 2021-09-15 | 2021-11-30 | 南京信息工程大学 | Automatic identification method for ground weather map high-voltage system |
| CN113724280B (en) * | 2021-09-15 | 2023-12-01 | 南京信息工程大学 | Automatic identification method for ground weather map high-voltage system |
| CN115905997A (en) * | 2022-10-28 | 2023-04-04 | 安徽省国家电投和新电力技术研究有限公司 | Wind turbine generator meteorological disaster early warning method and system based on prediction deviation optimization |
| CN116955939A (en) * | 2023-09-20 | 2023-10-27 | 南京气象科技创新研究院 | Meteorological element structural feature forecast error quantization expression method based on graph similarity |
| CN116955939B (en) * | 2023-09-20 | 2023-11-24 | 南京气象科技创新研究院 | Meteorological element structural feature forecast error quantization expression method based on graph similarity |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112613633B (en) | 2024-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112613633B (en) | Weather element prediction method and device for wind power plant | |
| CN110009037B (en) | Short-term engineering wind speed prediction method and system based on physical information coupling | |
| CN107621263B (en) | Geomagnetic positioning method based on road magnetic field characteristics | |
| CN107193060B (en) | A kind of multipath Typhoon Storm Surge Over method for quick predicting and system | |
| CN113222019B (en) | Meteorological forecast data processing method, device and equipment for transmission line tower | |
| CN109978275B (en) | Extreme strong wind speed prediction method and system based on mixed CFD and deep learning | |
| CN112700349B (en) | Method and device for selecting site of anemometer tower | |
| US20240054267A1 (en) | Method for planning a layout of a renewable energy site | |
| CN114595876A (en) | Regional wind field prediction model generation method and device and electronic equipment | |
| CN111325376A (en) | Wind speed prediction method and device | |
| CN105868443B (en) | Construction method of micro-terrain microclimate element field | |
| CN113392365A (en) | High-resolution meteorological grid data generation method and system | |
| CN111489063B (en) | Method and system for evaluating influence of wind driven generator on surrounding environment | |
| CN110555538B (en) | Wind power plant wind speed prediction method and prediction system | |
| CN113593006A (en) | Meteorological data spatial interpolation refining method and system based on deep learning | |
| CN115759483B (en) | Photovoltaic electric field solar irradiance prediction method, electronic equipment and storage medium | |
| CN115860248A (en) | Wind speed numerical assimilation short-term prediction method and system | |
| CN114325877B (en) | Assessment method and device for weather forecast data | |
| CN116245032A (en) | Wind power plant simulation generating capacity correction method and system considering space-time correlation | |
| Ganesan et al. | Assessment of wind energy potential using topographical and meteorological data of a site in Central India (Bhopal) | |
| CN110413656B (en) | Rapid flow axis automatic identification method based on high spatial resolution numerical mode | |
| CN111400844B (en) | Parameter scheme set generation method and wind speed forecasting method of meteorological model | |
| KR102350075B1 (en) | A Numerical Method for Predicting High Resolution Near-Surface Air Temperature Using Fine-scale Geographical Information | |
| CN112765852B (en) | Method for calculating possible maximum rainstorm for ultra-small watershed | |
| CN117272182B (en) | Daily air temperature prediction method, device, medium and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |