CN117077801A - Model training method, wind speed prediction method, device, medium and equipment - Google Patents
Model training method, wind speed prediction method, device, medium and equipment Download PDFInfo
- Publication number
- CN117077801A CN117077801A CN202310670340.4A CN202310670340A CN117077801A CN 117077801 A CN117077801 A CN 117077801A CN 202310670340 A CN202310670340 A CN 202310670340A CN 117077801 A CN117077801 A CN 117077801A
- Authority
- CN
- China
- Prior art keywords
- wind speed
- model
- climate
- determining
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000012549 training Methods 0.000 title claims abstract description 57
- 238000004590 computer program Methods 0.000 claims description 12
- 230000002123 temporal effect Effects 0.000 claims description 12
- 230000008878 coupling Effects 0.000 claims description 7
- 238000010168 coupling process Methods 0.000 claims description 7
- 238000005859 coupling reaction Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 16
- 238000012216 screening Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 238000013475 authorization Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Environmental & Geological Engineering (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present disclosure relates to the field of climate prediction, and in particular, to a model training method, a wind speed prediction method, a device, a medium, and equipment. The method comprises the following steps: acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model; determining a first climate factor according to the historical data; determining a significant climate factor and a weight parameter of the significant climate factor in a preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed; and updating the model parameters of the preset model according to the significant climate factors and the weight parameters. Therefore, the weather process of the height of the target area can be well simulated through the determined weather factors, and the accuracy of the output result of the wind speed prediction model is further improved; and through the screening of the obvious climate factors, the training efficiency of the wind speed prediction model can be improved, and the use experience of users is improved.
Description
Technical Field
The present disclosure relates to the field of climate prediction, and in particular, to a model training method, a wind speed prediction method, a device, a medium, and equipment.
Background
The novel power system mainly using new energy is gradually opened in the power market of each province. The new energy power station needs to declare the middle-long-term electric quantity of the month and forecast the average month wind speed. The accurate prediction of the average monthly wind speed can improve the accuracy of reporting the middle-long-term electric quantity in the month.
At present, the main stream statistical prediction scheme takes the corrected historical average state as a prediction result, and the prediction result does not consider the wind speed change caused by different climate processes in different years, so that the prediction result lacks obvious fluctuation characteristics, and the accuracy of the prediction wind speed result is insufficient.
Disclosure of Invention
The invention aims to provide a training method, a wind speed prediction method, a device, a medium and equipment for a model so as to improve the accuracy of wind speed prediction.
To achieve the above object, a first aspect of the present disclosure provides a wind speed prediction model training method, including:
acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model;
determining a first climate factor according to the historical data;
according to the first climate factor and the historical contemporaneous wind speed data of the target area, determining a significant climate factor in a preset model and a weight parameter of the significant climate factor, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed;
and updating the model parameters of the preset model according to the significant climate factors and the weight parameters.
Optionally, the obtaining, by using a numerical weather forecast model, historical data of weather forecast of a plurality of preset areas includes:
and acquiring historical data with spatial resolution smaller than a spatial resolution threshold and temporal resolution larger than a temporal resolution threshold according to the numerical weather forecast model for each preset area.
Optionally, the historical data includes at least one of sea surface temperature, sea heat content, surface temperature, barometric pressure field, altitude field, atmospheric temperature;
the determining a first climate factor according to the historical data comprises the following steps:
and inputting the historical data into a pre-trained climate factor determination model, obtaining at least one climate factor output by the climate factor determination model, and determining the climate factor determined based on the historical data as the first climate factor.
Optionally, the determining the significant climate factor and the weight parameter of the significant climate factor in the preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area includes:
determining pearson correlation coefficients for each first climate factor and the historical contemporaneous wind speed data of the target region;
determining the salient climate factor from the first climate factor according to the pearson correlation coefficient;
training a model of a preset model by using the historical contemporaneous wind speed data until a training ending condition is met, and determining the weight parameters of the remarkable climate factors.
Optionally, the method further comprises:
and according to the wave model and the dynamic ocean model, coupling to obtain a numerical weather forecast model.
A second aspect of the present disclosure provides a wind speed prediction method, comprising:
acquiring prediction data of weather forecast of a plurality of preset areas in prediction time through a numerical weather forecast model;
determining a second climate factor according to the prediction data;
and inputting the second climate factors into a wind speed prediction model to obtain a predicted wind speed output by the wind speed prediction model, wherein the wind speed prediction model is trained by the wind speed prediction model training method in the first aspect of the disclosure.
A third aspect of the present disclosure provides a wind speed prediction model training apparatus, comprising:
the first acquisition module is used for acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model;
the first determining module is used for determining a first climate factor according to the historical data;
the second determining module is used for determining a significant climate factor in a preset model and a weight parameter of the significant climate factor according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed;
and the updating module is used for updating the model parameters of the preset model according to the significant climate factors and the weight parameters.
Optionally, the first obtaining module is configured to obtain historical data of weather forecast of a plurality of preset areas by:
and acquiring historical data with spatial resolution smaller than a spatial resolution threshold and temporal resolution larger than a temporal resolution threshold according to the numerical weather forecast model for each preset area.
Optionally, the historical data includes at least one of sea surface temperature, sea heat content, surface temperature, barometric pressure field, altitude field, atmospheric temperature;
the first determination module is to determine a first climate factor by:
and inputting the historical data into a pre-trained climate factor determination model, obtaining at least one climate factor output by the climate factor determination model, and determining the climate factor determined based on the historical data as the first climate factor.
Optionally, the second determining module includes:
a first determination submodule for determining pearson correlation coefficients of each first climate factor and the historical contemporaneous wind speed data of the target area;
a second determining sub-module for determining the salient climate factor from the first climate factor according to the pearson correlation coefficient;
and the third determining submodule is used for training a model of a preset model by utilizing the historical contemporaneous wind speed data until the training ending condition is met, and determining the weight parameters of the remarkable climate factors.
Optionally, the apparatus further comprises:
and the coupling module is used for coupling to obtain a numerical weather forecast model according to the wave model and the dynamic ocean model.
A fourth aspect of the present disclosure provides a wind speed prediction apparatus, comprising:
the second acquisition module is used for acquiring the prediction data of the weather forecast of a plurality of preset areas in the prediction time through the numerical weather forecast model;
the third determining module is used for determining a second climate factor according to the prediction data;
and a fourth determining module, configured to input the second climate factor into a wind speed prediction model, and obtain a predicted wind speed output by the wind speed prediction model, where the wind speed prediction model is obtained by training according to the wind speed prediction model training method according to the first aspect of the present disclosure.
A fifth aspect of the present disclosure provides a computer readable medium having stored thereon a computer program which when executed by a processing device performs the steps of the method of the first or second aspect of the present disclosure.
A sixth aspect of the present disclosure provides an electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the method of the first or second aspect of the disclosure.
According to the technical scheme, historical data of weather forecast of a plurality of preset areas are obtained through a numerical weather forecast model; determining a first climate factor according to the historical data; determining a significant climate factor and a weight parameter of the significant climate factor in a preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed; and updating the model parameters of the preset model according to the significant climate factors and the weight parameters. Therefore, the weather process of the height of the target area can be well simulated through the determined weather factors, and the accuracy of the output result of the wind speed prediction model is further improved; and through the screening of the obvious climate factors, the training efficiency of the wind speed prediction model can be improved, and the use experience of users is improved.
Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of the disclosure, and are incorporated in and constitute a part of this specification, illustrate the disclosure and together with the description serve to explain, but do not limit the disclosure.
FIG. 1 is a flow chart of a wind speed predictive model training method provided in accordance with one embodiment of the present disclosure.
FIG. 2 is a flowchart of a method of wind speed prediction provided in accordance with one embodiment of the present disclosure.
FIG. 3 is a block diagram of a wind speed predictive model training apparatus provided in accordance with one embodiment of the present disclosure.
FIG. 4 is a block diagram of a wind speed prediction apparatus provided in accordance with one embodiment of the present disclosure.
Fig. 5 is a schematic structural view of an electronic device provided according to an embodiment of the present disclosure.
Detailed Description
Specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the disclosure, are not intended to limit the disclosure.
It should be noted that, all actions for acquiring signals, information or data in the present disclosure are performed under the condition of conforming to the corresponding data protection rule policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.
FIG. 1 is a flow chart of a wind speed predictive model training method provided in accordance with one embodiment of the present disclosure. In the embodiment of the disclosure, the training of the wind speed prediction model can be realized in a parallel updating mode. The method can be applied to terminal equipment and a server. As shown in fig. 1, the method may include S101 to S104.
S101, acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model.
For example, the numerical weather forecast model may be a machine learning model that is pre-trained. The numerical weather forecast model may refer to the european central numerical weather forecast model (ECMWF), which may be based on a wave model (ECWAM) coupled with a dynamic ocean model (NEMO). Historical weather observations relayed from radiosondes, weather satellites and other observation systems may be used as inputs to a numerical weather forecast model, in which case the corresponding output may be a historical numerical weather forecast, i.e., historical data of the weather forecast referred to above.
Climate change is not merely an isolated evolution of the atmosphere itself. It interacts with sea, ice and snow rings, land surface (rock rings) biospheres, etc. Climate formation and change is the overall manifestation of the feedback interactions of the various subsystems in the climate system. Therefore, in the process of establishing the numerical weather forecast model, the numerical weather forecast model can be trained by combining various physical elements and physical processes. The physical elements may be air temperature, humidity, air pressure, sunlight, precipitation, etc., and the physical process may be an atmospheric ring flow forming process, etc.
To increase the efficiency of operation, different coupling models (numerical weather forecast models) may employ different resolutions but with the same physical settings (including physical elements and physical processes). The new energy power station needs to declare the middle-long-term electric quantity of the month, so that the duration of the predicted time of the predicted wind speed is taken as an example of one month, the meteorological element fields on the corresponding different terrains are formed by spatial interpolation, wherein the spatial resolution of the meteorological elements (namely, the historical data) provided by the numerical weather forecast model finally can be 0.25 degrees, and the time resolution can be 6 hours. Thus, the numerical weather forecast model can provide high-resolution weather forecast history data meeting the requirement of forecasting wind speed, and accuracy of the weather factors determined later is improved.
The high-rise atmosphere and the large-scale physical process in the ocean in a long time range can be well simulated through the numerical weather forecast model, so that accurate historical data can be obtained.
For example, the plurality of preset areas may include each area in the global scope, and the plurality of areas may be selected as the preset areas in the global scope according to actual requirements. If a plurality of regions are selected as preset regions in the global scope, in order to ensure the reliability of the wind speed prediction model obtained by training, the determined preset regions may include a target region and other regions associated with the target region. The other region associated with the target region may be a region adjacent to the target region, or may be another region within a predetermined range centered on the target region.
Climate change is interactive with sea, ice and snow circles, land surface (rock circle) biospheres, etc. Therefore, in order to improve the accuracy of the subsequently determined climate factors, it is preferable that each region on the global scale be determined as a plurality of preset regions.
S102, determining a first climate factor according to the historical data.
The historical data may include, among other things, many of sea surface temperature, sea heat content, surface temperature, barometric pressure field, altitude field, atmospheric temperature.
Climate factors are the dominant factors forming climate and may include three types of climate factors, radiation factors, circulation factors and geographic factors. The radiation factor is related to solar radiation; the atmospheric circulation factors govern the transportation of heat, moisture and mass at different times and places, and play roles in regulating and redistributing energy; the geographical factors comprise geographical latitude, sea-land distribution, sea vibration height, land surface property, terrain orientation and the like, and have intricate and complex influence on the first two factors. The climate factors corresponding to different areas can also influence each other. Therefore, based on the climate factors, a good simulation of complex physical processes in the near-surface, boundary layer can be performed.
The numerical weather forecast model well simulates the high-level atmosphere and the large-scale physical process in the ocean, but is limited by the operation performance, and the complex physical process near the ground and in the boundary layer in the numerical model is less considered. To solve this problem, the first climate factor may be further determined based on historical data obtained by a numerical weather forecast model to simulate well the complex physical processes in the boundary layer, the middle ground, by the climate factor.
Therefore, the large-scale atmosphere and ocean element fields fed back by the historical data of the numerical weather forecast can be combined, and the historical weather factors (namely the first weather factors) of all the global areas responding to the weather system can be constructed, so that the subsequently determined wind speed prediction model can better simulate the weather process of the target area, such as the height of a power plant fan, through application of the weather factors, and the accuracy of the output result of the wind speed prediction model is improved.
For example, the historical data may be input into a pre-trained climate factor determination model, at least one climate factor output by the climate factor determination model may be obtained, and the climate factor determined based on the historical data may be determined as the first climate factor.
The climate factor determination model may be established based on a technology for constructing the climate factor, and the technology for constructing the climate factor is not limited, so long as the model can be used for realizing the construction of the climate factor. In addition, the technology for constructing climate factors is common knowledge in the art and will not be described in detail here.
The first climate factor may be processed based on a duration of the predicted time corresponding to the predicted wind speed, e.g., the first climate factor may be unified as a mean value over the duration. If the predicted time length corresponding to the wind speed to be predicted is one month, the first climate factors can be unified into month average values so that the climate factors are adapted to the time length corresponding to the predicted wind speed.
S103, determining the significant climate factors and the weight parameters of the significant climate factors in the preset model according to the first climate factors and the historical contemporaneous wind speed data of the target area.
Wherein the historical contemporaneous wind speed data is determined based on a predicted time corresponding to the predicted wind speed.
For example, the target area may be preset according to actual requirements, for example, a location of a power plant fan may be used as the target area, so as to obtain a wind speed prediction model for predicting a wind speed of a power plant station through training.
As described above, the new energy power station needs to report the middle-long-term electric quantity of the month, so the predicted time may be a specific month to be measured, and the historical contemporaneous wind speed data may include wind speed data of months to be measured in a plurality of years. Taking the example that the predicted time corresponding to the predicted wind speed is 6 months, that is, the wind speed to be predicted is the wind speed of 6 months in the current year, the historical contemporaneous wind speed data can include the wind speed data of 6 months in the last year and the wind speed data of 6 months in the previous year.
Wind speeds in a period of time often have similarity, so that historical wind speed data of other months in the same season of the same year can be used as historical contemporaneous wind speed data. Still taking the predicted time corresponding to the predicted wind speed as 6 months as an example, the historical contemporaneous wind speed data may include historical wind speed data for 4 months and 5 months of the same year.
The significant climate factors can be screened out from the first climate factors through feature engineering, so that training efficiency of the wind speed prediction model is improved, and user experience is improved. Under the condition that the significant climate factors are determined, the historical contemporaneous wind speed data of the target area can be used as a training sample, the preset model is trained until the training ending condition is met, and the weight parameter corresponding to each significant climate factor can be determined.
For example, the training end condition may include at least one of: the output value of the loss function of the model is smaller than or equal to a preset threshold value, and the iteration times reach the preset times threshold value. Thus, the accuracy of the wind speed prediction model obtained by training can be improved.
And S104, updating model parameters of a preset model according to the significant climate factors and the weight parameters.
For example, after determining the significant climate factor and the weight parameter of the significant climate factor, a mode of updating the preset model according to the significant climate factor and the weight parameter may be a model updating mode commonly used in the art, which is not described herein. The input of the trained wind speed prediction model may comprise the latest climate factor determined based on the predicted data of the weather forecast and the output may comprise the predicted wind speed. Therefore, a wind speed prediction model obtained based on the historical contemporaneous wind speed data training of the power plant can provide a predicted wind speed with higher accuracy, and further the accuracy of reporting the middle-long-term electric quantity in the month is improved.
According to the technical scheme, historical data of weather forecast of a plurality of preset areas are obtained through a numerical weather forecast model; determining a first climate factor according to the historical data; determining a significant climate factor and a weight parameter of the significant climate factor in a preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed; and updating model parameters of the preset model according to the significant climate factors and the weight parameters. Therefore, the weather process of the height of the target area can be well simulated through the determined weather factors, and the accuracy of the output result of the wind speed prediction model is further improved; and through the screening of the obvious climate factors, the training efficiency of the wind speed prediction model can be improved, and the use experience of users is improved.
Optionally, in S101, obtaining, by a numerical weather forecast model, historical data of weather forecast of a plurality of preset areas may include:
and acquiring historical data with spatial resolution smaller than a spatial resolution threshold and temporal resolution larger than a temporal resolution threshold according to each preset area through a numerical weather forecast model.
The spatial resolution threshold and the temporal resolution threshold may be preset according to actual requirements. For example, the spatial resolution threshold and the temporal resolution threshold may be adapted to the duration of the predicted time of the predicted wind speed. If the spatial resolution is smaller than the spatial resolution threshold, determining that the remote sensing image corresponding to the preset area is clearer; if the time resolution is greater than the time resolution threshold, it may be determined that the time span corresponding to the historical data is longer. Therefore, the determined historical data has higher resolution, and the numerical weather forecast model can provide the historical data of the weather forecast with high resolution meeting the prediction requirement so as to improve the accuracy of the weather factors determined later.
Optionally, in S103, determining the significant climate factor and the weight parameter of the significant climate factor in the preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area may include:
determining pearson correlation coefficients of each first climate factor and historical contemporaneous wind speed data of the target area;
determining a significant climate factor from the first climate factor according to the pearson correlation coefficient;
training a model of a preset model by using the historical contemporaneous wind speed data until the training ending condition is met, and determining the weight parameters of the remarkable climate factors.
Illustratively, the calculation of the pearson correlation coefficient is a conventional technical means in the art, and will not be described herein. The greater the pearson correlation coefficient, the greater the degree of linear correlation between the corresponding first climate factor and the historical contemporaneous wind speed data of the target region. The first climate factor corresponding to the pearson correlation coefficient greater than the coefficient threshold may be determined as the significant climate factor, or the preset number of first climate factors corresponding to the pearson correlation coefficient having the largest value may be determined as the significant climate factor. After the significant climate factors are determined, training a model of the preset model by using the historical contemporaneous wind speed data until the training ending condition is met, and determining the weight parameters of the significant climate factors. Therefore, the wind speed prediction model with high reliability can be trained efficiently.
FIG. 2 is a flowchart of a method of wind speed prediction provided in accordance with one embodiment of the present disclosure. As shown in fig. 2, the method may include S201 to S203.
S201, obtaining prediction data of weather forecast of a plurality of preset areas in prediction time through a numerical weather forecast model.
S202, determining a second climate factor according to the prediction data.
S203, inputting the second climate factors into a wind speed prediction model to obtain a predicted wind speed output by the wind speed prediction model, wherein the wind speed prediction model is obtained by training according to the wind speed prediction model training method in any embodiment.
The method for obtaining the weather forecast of the plurality of preset areas in the forecast time is similar to the method for obtaining the historical data. The difference is that the current weather observations relayed from radiosondes, weather satellites and other observation systems can be used as inputs to a numerical weather forecast model, where the corresponding outputs can be the most recently predicted numerical weather forecast, i.e. the predicted data of the weather forecast referred to above. According to the prediction data, the second climate factor is determined to be the same as the method in S102 described above, and will not be described here again. Thus, the predicted wind speed of the target area in the predicted time can be simply and efficiently determined.
Based on the same inventive concept, the present disclosure also provides a wind speed prediction model training device. FIG. 3 is a block diagram of a wind speed predictive model training apparatus 300 provided in accordance with one embodiment of the present disclosure. Referring to fig. 3, the wind speed prediction model training apparatus 300 may include:
the first obtaining module 301 is configured to obtain, through a numerical weather forecast model, historical data of weather forecast of a plurality of preset areas;
a first determining module 302, configured to determine a first climate factor according to the historical data;
a second determining module 303, configured to determine a significant climate factor in a preset model and a weight parameter of the significant climate factor according to the first climate factor and historical contemporaneous wind speed data of the target area, where the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed;
and the updating module 304 is configured to update the model parameters of the preset model according to the significant climate factors and the weight parameters.
According to the technical scheme, historical data of weather forecast of a plurality of preset areas are obtained through a numerical weather forecast model; determining a first climate factor according to the historical data; determining a significant climate factor and a weight parameter of the significant climate factor in a preset model according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed; and updating the model parameters of the preset model according to the significant climate factors and the weight parameters. Therefore, the weather process of the height of the target area can be well simulated through the determined weather factors, and the accuracy of the output result of the wind speed prediction model is further improved; and through the screening of the obvious climate factors, the training efficiency of the wind speed prediction model can be improved, and the use experience of users is improved.
Optionally, the first obtaining module 301 is configured to obtain historical data of weather forecast of a plurality of preset areas by:
and acquiring historical data with spatial resolution smaller than a spatial resolution threshold and temporal resolution larger than a temporal resolution threshold according to the numerical weather forecast model for each preset area.
Optionally, the historical data includes at least one of sea surface temperature, sea heat content, surface temperature, barometric pressure field, altitude field, atmospheric temperature;
the first determining module 302 is configured to determine the first climate factor by:
and inputting the historical data into a pre-trained climate factor determination model, obtaining at least one climate factor output by the climate factor determination model, and determining the climate factor determined based on the historical data as the first climate factor.
Optionally, the second determining module 303 includes:
a first determination submodule for determining pearson correlation coefficients of each first climate factor and the historical contemporaneous wind speed data of the target area;
a second determining sub-module for determining the salient climate factor from the first climate factor according to the pearson correlation coefficient;
and the third determining submodule is used for training a model of a preset model by utilizing the historical contemporaneous wind speed data until the training ending condition is met, and determining the weight parameters of the remarkable climate factors.
Optionally, the apparatus 300 further includes:
and the coupling module is used for coupling to obtain a numerical weather forecast model according to the wave model and the dynamic ocean model.
Based on the same inventive concept, the present disclosure also provides a wind speed prediction apparatus. FIG. 4 is a block diagram of a wind speed prediction apparatus 400 provided in accordance with one embodiment of the present disclosure. Referring to fig. 4, the wind speed prediction apparatus 400 may include:
a second obtaining module 401, configured to obtain, through a numerical weather forecast model, prediction data of weather forecast of a plurality of preset areas in a prediction time;
a third determining module 402, configured to determine a second climate factor according to the prediction data;
a fourth determining module 403, configured to input the second climate factor into a wind speed prediction model, and obtain a predicted wind speed output by the wind speed prediction model, where the wind speed prediction model is obtained by training according to the wind speed prediction model training method described in any of the foregoing embodiments.
The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.
Fig. 5 is a schematic structural diagram of an electronic device 1900 provided according to one embodiment of the disclosure. For example, electronic device 1900 may be provided as a server. Referring to fig. 5, the electronic device 1900 includes a processor 1922, which may be one or more in number, and a memory 1932 for storing computer programs executable by the processor 1922. The computer program stored in memory 1932 may include one or more modules each corresponding to a set of instructions. Further, processor 1922 may be configured to execute the computer program to perform the wind speed prediction model training method or wind speed prediction method described above.
In addition, the electronic device 1900 may further include a power component 1926 and a communication component 1950, the power component 1926 may be configured to perform power management of the electronic device 1900, and the communication component 1950 may be configured to enable communication of the electronic device 1900, e.g., wired or wireless communication. In addition, the electronic device 1900 may also include an input/output (I/O) interface 1958. Electronic device 1900 may operate based on an operating system stored in memory 1932.
In another exemplary embodiment, a computer readable storage medium is also provided comprising program instructions which, when executed by a processor, implement the steps of the wind speed prediction model training method or wind speed prediction method described above. For example, the non-transitory computer readable storage medium may be the memory 1932 including program instructions described above that are executable by the processor 1922 of the electronic device 1900 to perform the wind speed prediction model training method or the wind speed prediction method described above.
In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable apparatus, the computer program having code portions for performing the wind speed prediction model training method or the wind speed prediction method described above when executed by the programmable apparatus.
The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solutions of the present disclosure within the scope of the technical concept of the present disclosure, and all the simple modifications belong to the protection scope of the present disclosure.
In addition, the specific features described in the above embodiments may be combined in any suitable manner without contradiction. The various possible combinations are not described further in this disclosure in order to avoid unnecessary repetition.
Moreover, any combination between the various embodiments of the present disclosure is possible as long as it does not depart from the spirit of the present disclosure, which should also be construed as the disclosure of the present disclosure.
Claims (10)
1. A method of training a wind speed predictive model, comprising:
acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model;
determining a first climate factor according to the historical data;
according to the first climate factor and the historical contemporaneous wind speed data of the target area, determining a significant climate factor in a preset model and a weight parameter of the significant climate factor, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed;
and updating the model parameters of the preset model according to the significant climate factors and the weight parameters.
2. The method for training a wind speed prediction model according to claim 1, wherein the obtaining, by a numerical weather prediction model, the historical data of weather predictions for a plurality of preset areas includes:
and acquiring historical data with spatial resolution smaller than a spatial resolution threshold and temporal resolution larger than a temporal resolution threshold according to the numerical weather forecast model for each preset area.
3. The wind speed predictive model training method of claim 1, wherein the historical data includes at least one of sea surface temperature, sea heat content, surface temperature, barometric pressure field, altitude field, atmospheric temperature;
the determining a first climate factor according to the historical data comprises the following steps:
and inputting the historical data into a pre-trained climate factor determination model, obtaining at least one climate factor output by the climate factor determination model, and determining the climate factor determined based on the historical data as the first climate factor.
4. The method according to claim 1, wherein determining the significant climate factors and the weight parameters of the significant climate factors in the preset model according to the first climate factors and the historical contemporaneous wind speed data of the target area comprises:
determining pearson correlation coefficients for each of the first climate factors and the historical contemporaneous wind speed data of the target region;
determining the salient climate factor from the first climate factor according to the pearson correlation coefficient;
training a model of the preset model by using the historical contemporaneous wind speed data until a training ending condition is met, and determining the weight parameters of the remarkable climate factors.
5. The wind speed predictive model training method of claim 1, further comprising:
and according to the wave model and the dynamic ocean model, coupling to obtain a numerical weather forecast model.
6. A method of wind speed prediction, comprising:
acquiring prediction data of weather forecast of a plurality of preset areas in prediction time through a numerical weather forecast model;
determining a second climate factor according to the prediction data;
inputting the second climate factors into a wind speed prediction model to obtain a predicted wind speed output by the wind speed prediction model, wherein the wind speed prediction model is trained by the wind speed prediction model training method according to any one of claims 1-5.
7. A wind speed predictive model training apparatus, comprising:
the first acquisition module is used for acquiring historical data of weather forecast of a plurality of preset areas through a numerical weather forecast model;
the first determining module is used for determining a first climate factor according to the historical data;
the second determining module is used for determining a significant climate factor in a preset model and a weight parameter of the significant climate factor according to the first climate factor and the historical contemporaneous wind speed data of the target area, wherein the historical contemporaneous wind speed data is determined based on a prediction time corresponding to a predicted wind speed;
and the updating module is used for updating the model parameters of the preset model according to the significant climate factors and the weight parameters.
8. A wind speed prediction apparatus, comprising:
the second acquisition module is used for acquiring the prediction data of the weather forecast of a plurality of preset areas in the prediction time through the numerical weather forecast model;
the third determining module is used for determining a second climate factor according to the prediction data;
a fourth determining module, configured to input the second climate factor into a wind speed prediction model, and obtain a predicted wind speed output by the wind speed prediction model, where the wind speed prediction model is obtained by training by the wind speed prediction model training method according to any one of claims 1-5.
9. A computer readable medium on which a computer program is stored, characterized in that the program, when being executed by a processing device, carries out the steps of the method according to any one of claims 1-6.
10. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310670340.4A CN117077801A (en) | 2023-06-07 | 2023-06-07 | Model training method, wind speed prediction method, device, medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310670340.4A CN117077801A (en) | 2023-06-07 | 2023-06-07 | Model training method, wind speed prediction method, device, medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117077801A true CN117077801A (en) | 2023-11-17 |
Family
ID=88706784
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310670340.4A Pending CN117077801A (en) | 2023-06-07 | 2023-06-07 | Model training method, wind speed prediction method, device, medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117077801A (en) |
-
2023
- 2023-06-07 CN CN202310670340.4A patent/CN117077801A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363327B (en) | ConvLSTM and 3D-CNN-based short rainfall prediction method | |
| US10041475B1 (en) | Reducing curtailment of wind power generation | |
| US10598157B2 (en) | Reducing curtailment of wind power generation | |
| CN111027175B (en) | Method for evaluating social and economic influences of flood based on coupling model integrated simulation | |
| Wong et al. | Stochastic model output statistics for bias correcting and downscaling precipitation including extremes | |
| CN112285807B (en) | Meteorological information prediction method and device | |
| CN109472283B (en) | Dangerous weather prediction method and device based on multiple incremental regression tree model | |
| WO2020203854A1 (en) | Power generation amount prediction device, power generation amount prediction method, and program | |
| CN106803128B (en) | Photovoltaic power generation prediction method and device | |
| US20100100328A1 (en) | System and Method for Generating a Cloud Type and Coverage Prediction Database | |
| CN114493050A (en) | Multi-dimensional fusion new energy power parallel prediction method and device | |
| CN112100921A (en) | Method for acquiring wind resource and wind speed based on WRF and random forest | |
| CN112100922A (en) | Wind resource prediction method based on WRF and CNN convolutional neural network | |
| CN111505740A (en) | Weather prediction method, weather prediction device, computer equipment and storage medium | |
| CN116224473B (en) | Wind speed forecasting method, model training method, device and storage medium | |
| CN115079307A (en) | Wind power plant weather forecasting method and system based on multi-mode optimal integration | |
| CN112613633A (en) | Meteorological element prediction method and device for wind power plant | |
| CN114819264A (en) | Photovoltaic power station irradiance ultra-short term prediction method based on space-time dependence and storage medium | |
| CN117421871A (en) | Offshore wind power potential evaluation method and device and computer equipment | |
| CN116931129A (en) | Short-term precipitation prediction method, device, equipment and medium based on multi-mode set | |
| CN114418243B (en) | Distributed new energy cloud grid prediction method and system | |
| CN117077801A (en) | Model training method, wind speed prediction method, device, medium and equipment | |
| CN113222276B (en) | Ice-covering condition prediction method, device, computer equipment and storage medium | |
| CN115983104A (en) | Wind speed prediction method and device, storage medium and electronic equipment | |
| Seo et al. | Evolutionary nonlinear Model Output Statistics for wind speed prediction using Genetic Programming |
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 |