CN113420939A - Cloud picture forecasting method, equipment and storage medium - Google Patents
Cloud picture forecasting method, equipment and storage medium Download PDFInfo
- Publication number
- CN113420939A CN113420939A CN202110799416.4A CN202110799416A CN113420939A CN 113420939 A CN113420939 A CN 113420939A CN 202110799416 A CN202110799416 A CN 202110799416A CN 113420939 A CN113420939 A CN 113420939A
- Authority
- CN
- China
- Prior art keywords
- cloud picture
- forecasting
- historical
- forecast
- target
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a cloud picture forecasting method, equipment and a storage medium, wherein the cloud picture forecasting method comprises the following steps: acquiring meteorological element forecast information within a preset time length from the current moment in a target area; and inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in a target area output by the target cloud picture forecasting model. The cloud picture forecasting method can combine meteorological element forecasting information and a target cloud picture forecasting model trained by a machine learning method to obtain a forecasting cloud picture, on one hand, only the meteorological element forecasting information needs to be obtained and is input into the target cloud picture forecasting model to obtain the forecasting cloud picture, and the cloud picture forecasting method is simple to implement and low in complexity; on the other hand, the target cloud picture forecasting model can obtain stable output without a starting process, so that the forecasting cloud picture with higher accuracy can be obtained at the initial stage of forecasting.
Description
Technical Field
The embodiment of the invention relates to the field of weather forecasting, in particular to a cloud picture forecasting method, equipment and a storage medium.
Background
The cloud picture is an image of cloud cover and ground surface features on the earth, as viewed from top to bottom by a meteorological satellite. Cloud pictures received in the current meteorological service mainly comprise infrared cloud pictures, visible light cloud pictures, vapor pictures and the like. Different weather systems can be identified by using the cloud pictures, the positions of the weather systems are determined, the strength and the development trend of the weather systems are estimated, and a basis is provided for weather analysis and weather forecast. Especially in the areas such as oceans, deserts, plateaus and the like which lack meteorological observation stations, the data provided by the satellite cloud pictures make up for the defects of conventional detection data and play an important role in improving the prediction accuracy. However, like other observation means, meteorological satellites can only provide past and near real-time meteorological observations, and cannot provide meteorological information for a future period of time. If the weather cloud picture of a period of time in the future can be forecasted by a technical means, the weather cloud picture is linked with the observed satellite cloud picture in time, and the weather forecasting method has important reference value for weather forecasters to analyze weather conditions and forecast weather trends.
Currently, cloud picture forecasting can be performed in the following way: forecasting meteorological element information by using a numerical model, inputting the forecasted meteorological element information into a radiation transmission model, calculating radiation brightness temperature, and generating a forecasted cloud picture.
However, the above method requires a complicated procedure: acquiring forecasted meteorological element information; inputting the forecasted meteorological element information into a radiation transmission mode to obtain radiation brightness temperature; according to the radiation brightness temperature, a cloud picture is generated, and therefore the cloud picture forecasting method is high in complexity. Meanwhile, the method based on the radiation transmission mode needs to go through a start-up process to obtain stable output, resulting in low accuracy of the predicted cloud picture at the initial period of prediction.
Disclosure of Invention
The invention provides a cloud picture forecasting method, equipment and a storage medium, which are used for solving the technical problems of higher complexity and lower accuracy of a forecasting initial period in the conventional cloud picture forecasting method.
In a first aspect, an embodiment of the present invention provides a cloud picture forecasting method, including:
acquiring meteorological element forecast information within a preset time length from the current moment in a target area;
inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecasting cloud picture within a preset time length from the current time in the target area output by the target cloud picture forecasting model; the target cloud picture forecasting model is a model trained in advance by adopting a machine learning method according to historical meteorological element forecasting information and the historical cloud pictures in the target area.
In a second aspect, an embodiment of the present invention provides a cloud picture forecasting device, including:
the first acquisition module is used for acquiring meteorological element forecast information within a preset time length from the current moment in a target area;
the first determination module is used for inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecasting cloud picture within a preset time length from the current time in the target area output by the target cloud picture forecasting model; the target cloud picture forecasting model is a model trained in advance by adopting a machine learning method according to historical meteorological element forecasting information and the historical cloud pictures in the target area.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the cloud forecasting method as provided in the first aspect.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the cloud picture forecasting method provided in the first aspect.
The embodiment of the invention provides a cloud picture forecasting method, equipment and a storage medium, wherein the method comprises the following steps: acquiring meteorological element forecast information within a preset time length from the current moment in a target area; inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in the target area output by the target cloud picture forecasting model. The target cloud picture forecasting model is a model trained in advance according to historical meteorological element forecasting information and the historical cloud pictures in the target area. The cloud picture forecasting method can be used for obtaining a forecast cloud picture by combining meteorological element forecasting information and a target cloud picture forecasting model trained by adopting a machine learning method, and has the following beneficial effects compared with the current cloud picture forecasting method: on one hand, the forecast cloud picture can be obtained only by acquiring the meteorological element forecast information and inputting the meteorological element forecast information into the target cloud picture forecast model, and the method is simple to implement and low in complexity; on the other hand, the target cloud picture forecasting model can obtain stable output without a starting process, so that the cloud picture forecasting method provided by the embodiment can obtain a forecast cloud picture with high accuracy at the initial stage of forecasting.
Drawings
Fig. 1 is a schematic flow chart of a cloud picture forecasting method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a target cloud picture forecasting model according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a cloud picture forecasting method according to another embodiment of the present invention;
fig. 4 is a schematic structural diagram of a cloud image forecasting device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a cloud image forecasting device according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Different weather systems can be identified by using the cloud pictures, the positions of the weather systems are determined, the strength and the development trend of the weather systems are estimated, and a basis is provided for weather analysis and weather forecast. Therefore, it is very important to realize cloud picture prediction.
The embodiment provides a cloud picture forecasting method, which comprises the following steps: acquiring meteorological element forecast information within a preset time length from the current moment in a target area; and inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in a target area output by the target cloud picture forecasting model. The target cloud picture forecasting model is a model trained in advance according to historical meteorological element forecasting information and the historical cloud pictures in the target area. The cloud picture forecasting method can be used for obtaining a forecast cloud picture by combining meteorological element forecasting information and a target cloud picture forecasting model trained by adopting a machine learning method, and has the following beneficial effects compared with the current cloud picture forecasting method: on one hand, the forecast cloud picture can be obtained only by acquiring the meteorological element forecast information and inputting the meteorological element forecast information into the target cloud picture forecast model, and the method is simple to implement and low in complexity; on the other hand, the target cloud picture forecasting model can obtain stable output without a starting process, so that the cloud picture forecasting method provided by the embodiment can obtain a forecast cloud picture with high accuracy at the initial stage of forecasting.
Fig. 1 is a schematic flow chart of a cloud picture forecasting method according to an embodiment of the present invention. The cloud picture forecasting method and the cloud picture forecasting device are suitable for a scene of forecasting the cloud pictures. The embodiment is exemplified by applying the method to a computer device. It will be appreciated that the method may also be applied to a server, and may also be applied to a system comprising a computer device and a server, and be implemented by interaction of the computer device and the server. The computer device may be, but is not limited to, various personal computers, notebook computers, smart phones, and tablet computers. The server may be implemented as a stand-alone server or as a server cluster consisting of a plurality of servers. As shown in fig. 1, the cloud picture forecasting method provided in this embodiment includes the following steps:
step 101: acquiring meteorological element forecast information within a preset time length from the current moment in the target area.
Specifically, the target area in this embodiment is an area that needs cloud picture prediction. For convenience of description in this embodiment, it is referred to as a target region. The target area here may range from a certain village, a certain town, a certain district, a certain city, a certain province or a certain country. This embodiment is not limited thereto.
The preset time period from the current time refers to a future time period that is a preset time period from the current time. Illustratively, the preset time period here may be 12 hours, 24 hours, or the like.
The meteorological element forecast information in this embodiment includes at least one of the following parameters: the forecasted cloud volume, the forecasted cloud water volume, the forecasted cloud ice volume, the forecasted air temperature, the forecasted air pressure, and the forecasted humidity.
In step 101, the embodiment may acquire weather element forecast information based on a numerical forecast manner.
More specifically, one possible implementation procedure of step 101 may be: according to reference meteorological element forecast information in a reference area within a preset time length from the current time, and based on a Weather forecast (WRF) mode, acquiring meteorological element forecast information in a target area within a preset time length from the current time. Wherein the reference region is a region including the target region, and an area of the reference region is larger than an area of the target region.
Exemplarily, assuming that the target area is china, the reference area may be all areas in the global scope. Assuming that the target region is Jiangsu province, the reference region may be China.
Alternatively, the reference meteorological element forecast information may be acquired from a Global Forecasting System (GFS).
When acquiring meteorological element forecast information within a preset time from the current time in a target area based on a WRF mode, specifically, the reference meteorological element forecast information is used as a driving field, and the WRF mode is operated, so that the meteorological element forecast information can be obtained.
In one implementation, the weather element forecast information may be weather element forecast information corresponding to each forecast time. In this implementation, a plurality of forecast times are included within a preset time period from the current time, and in step 101, meteorological element forecast information at each forecast time may be obtained. For example, assuming that the preset time duration is 24 hours, and the forecast time is every 15 minutes within the 24 hours, 24 × 4 meteorological element forecast information may be acquired in step 101.
Step 102: and inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in a target area output by the target cloud picture forecasting model.
The target cloud picture forecasting model is a model trained by a machine learning method in advance according to historical meteorological element forecasting information and historical cloud pictures in a target area.
Specifically, in this embodiment, a machine learning method is adopted in advance to train a target cloud picture forecasting model. Illustratively, the target cloud picture forecasting model may be a neural network model or a model trained based on an eXtreme Gradient addition (XGBoost) algorithm. The XGboost algorithm is fast in training speed and good in training effect in the training process. The training process of the target cloud picture forecasting model will be described in detail in the next embodiment.
In one implementation, to improve the accuracy of cloud picture prediction, the target cloud picture prediction model includes a plurality of target cloud picture prediction submodels. Fig. 2 is a schematic structural diagram of a target cloud picture forecasting model according to an embodiment of the present invention. As shown in fig. 2, the target cloud image forecast model in this embodiment may include N target cloud image forecast submodels.
In a specific implementation manner, the preset time period may be 24 hours, and N may be 4. That is, the target cloud picture forecasting model includes 4 target cloud picture forecasting submodels.
Based on the implementation manner, optionally, the cloud picture forecasting method provided by this embodiment further includes: and determining a target cloud picture forecasting sub-model corresponding to each forecasting time according to the distance between each forecasting time and the current time. That is, in this embodiment, the distance between the forecast time and the current time and the mapping relationship between the target cloud picture forecast submodules are provided. Because the corresponding relation between the meteorological element forecast information and the forecast cloud picture at different forecast moments is different, the accuracy of the forecast cloud picture can be further improved by the implementation mode of arranging the target cloud picture forecast sub-modules.
Correspondingly, step 102 may specifically be: and inputting the meteorological element forecast information corresponding to each forecast time into the corresponding target cloud picture forecast sub-model to obtain the forecast cloud picture of each forecast time within a preset time length from the current time in the target area.
It should be noted that, in an implementation manner, the distance between the forecasting time and the current time, and the target cloud picture forecasting sub-modules may be in a one-to-one correspondence relationship. That is, different forecasting moments correspond to different target cloud picture forecasting submodules. The accuracy of the forecast cloud picture obtained by the implementation mode is high.
In another implementation, the distance between the forecast time and the current time, and the target cloud picture forecast sub-modules may be in a many-to-one relationship. That is, multiple forecast moments correspond to the same target cloud picture forecast submodule. By adopting the implementation mode, the number of target cloud picture forecasting sub-modules can be reduced, and the workload in the training process is reduced, so that the accuracy of forecasting the cloud picture and the training efficiency are balanced.
For example, assuming that the preset time period is 24 hours, the target cloud picture forecasting model includes 4 target cloud picture forecasting submodels: a target cloud picture forecast sub-model 1, a target cloud picture forecast sub-model 2, a target cloud picture forecast sub-model 3, and a target cloud picture forecast sub-model 4. When the distance between the forecasting time and the current time is 0-6 hours, forecasting the cloud picture according to the target cloud picture forecasting sub-model 1; when the distance between the forecasting time and the current time is 6-12 hours, forecasting the cloud picture according to the target cloud picture forecasting sub-model 2; when the distance between the forecasting time and the current time is 12 to 18 hours, forecasting the cloud picture according to the target cloud picture forecasting sub-model 3; and when the distance between the forecasting time and the current time is 18-24 hours, forecasting the cloud picture according to the target cloud picture forecasting sub-model 4.
Continuing with the above example, when the forecast time is every 15 minutes within the 24 hours, the cloud picture forecasting method provided by the present embodiment can obtain the forecast cloud pictures every 15 minutes within the 24 hours in the future.
Optionally, the cloud images in this embodiment may be infrared cloud images, visible cloud images, vapor maps, and the like.
Optionally, each pixel point on the forecast cloud chart in this embodiment may be the normalized infrared bright temperature.
The embodiment provides a cloud picture forecasting method, which comprises the following steps: acquiring meteorological element forecast information within a preset time length from the current moment in a target area; and inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in a target area output by the target cloud picture forecasting model. The target cloud picture forecasting model is a model trained in advance according to historical meteorological element forecasting information and the historical cloud pictures in the target area. The cloud picture forecasting method can be used for obtaining a forecast cloud picture by combining meteorological element forecasting information and a target cloud picture forecasting model trained by adopting a machine learning method, and has the following beneficial effects compared with the current cloud picture forecasting method: on one hand, the forecast cloud picture can be obtained only by acquiring the meteorological element forecast information and inputting the meteorological element forecast information into the target cloud picture forecast model, and the method is simple to implement and low in complexity; on the other hand, the target cloud picture forecasting model can obtain stable output without a starting process, so that the cloud picture forecasting method provided by the embodiment can obtain a forecast cloud picture with high accuracy at the initial stage of forecasting.
Fig. 3 is a schematic flow chart of a cloud picture forecasting method according to another embodiment of the present invention. The cloud picture forecasting method provided by this embodiment describes how to train the target cloud picture forecasting model in detail on the basis of the embodiment shown in fig. 1 and various optional implementation schemes. As shown in fig. 3, the cloud picture forecasting method provided in this embodiment includes the following steps:
step 301: and acquiring historical meteorological element forecast information and a historical cloud picture of the target area.
Specifically, the historical cloud image in this embodiment may be a cloud image that has been observed by an observation satellite. Alternatively, the satellite may be a wind cloud number four satellite FY-4A.
More specifically, historical clouds may be collected from a national satellite weather center officer network. Since the observation satellite outputs the cloud image at a preset sampling interval, which may be 15 minutes, in step 301, a historical cloud image every 15 minutes in a past period of time may be obtained. That is, the historical cloud map may be a historical cloud map corresponding to the historical forecast time (for convenience of description, a time obtained at every other sampling interval in a past period of time is referred to as the historical forecast time).
Optionally, each pixel point on the historical cloud image in this embodiment is a normalized infrared bright temperature. Illustratively, the past period of time here may be the past year.
In this embodiment, the historical reference meteorological element forecast information may be acquired from the GFS, and then the historical meteorological element forecast information of the target area may be acquired based on the WRF mode. Specifically, the historical reference meteorological element forecast information is used as a driving field, and the WRF mode is operated, so that the historical meteorological element forecast information can be obtained.
Alternatively, the historical meteorological element forecast information may also be historical meteorological element forecast information corresponding to the historical forecast time. In step 301, historical meteorological element forecast information corresponding to each historical forecast time in a past period of time may be acquired. For example, historical meteorological element forecast information is acquired at 15 minute intervals in the past year.
The historical meteorological element forecast information in this embodiment includes at least one of the following parameters: historically forecasted cloud cover, historically forecasted cloud water cover, historically forecasted cloud ice cover, historically forecasted air temperature, historically forecasted air pressure, and historically forecasted humidity.
More specifically, when acquiring the historical meteorological element forecast information, the historical meteorological element forecast information within a preset time length in the future may be acquired with 0 point of each day as the historical current time. Of course, any time of each day may be used as the current historical time, and this embodiment is not limited thereto.
Optionally, the resolution of the historical meteorological element forecast information and the historical cloud chart acquired in this embodiment may be 4 kilometers.
Step 302: inputting the historical meteorological element forecasting information into an initial cloud picture forecasting model for training to obtain an output result.
Step 303: and according to the output result and the corresponding historical cloud picture, performing feedback updating on the initial cloud picture forecasting model until the training is finished, and determining the initial cloud picture forecasting model at the training end as a target cloud picture forecasting model.
Specifically, the corresponding historical cloud image refers to the historical cloud image at the same time as the historical meteorological element forecast information.
In steps 302 and 303, a loss parameter may be determined based on the output and the corresponding historical cloud map. The loss parameter may be a numerical value, a vector, or a matrix.
Optionally, when the loss parameter does not satisfy the preset convergence condition, for example, the loss parameter is greater than a preset threshold, the network structure and the network parameter of the initial cloud picture prediction model may be adjusted, the initial cloud picture prediction model is updated, and the step of inputting the historical meteorological element prediction information into the initial cloud picture prediction model for training is performed to obtain the output result, until the loss parameter satisfies the convergence condition, and the initial cloud picture prediction model when the loss parameter satisfies the convergence condition is used as the target cloud picture prediction model.
Corresponding to the implementation manner in which the target cloud picture forecasting model includes a plurality of target cloud picture forecasting sub models in the embodiment shown in fig. 1 and various optional implementation manners, in this embodiment, the initial cloud picture forecasting model may include a plurality of initial cloud picture forecasting sub models.
After obtaining the historical meteorological element forecast information and the historical cloud chart, the embodiment further includes: according to the distance between the historical current time and the historical forecast time, dividing the historical meteorological element forecast information into a plurality of groups of historical meteorological element forecast information sets, and dividing the historical cloud picture into a plurality of groups of historical cloud picture sets.
More specifically, the distance between the historical current time and the historical forecast time may be divided into a plurality of ranges. For example, 0-6 hours, 6-12 hours, 12-18 hours, and 18-24 hours. When dividing the historical meteorological element forecast information and the historical cloud picture set, dividing the historical meteorological element forecast information with the distance of 0-6 hours between the historical current time and the historical forecast time into a group, and dividing the historical cloud picture with the distance of 0-6 hours into a group; dividing historical meteorological element forecast information with the distance of 6-12 hours between the historical current time and the historical forecast time into a group, and dividing historical cloud pictures with the distance of 6-12 hours into a group; dividing historical meteorological element forecast information with a distance of 12-18 hours between the historical current time and the historical forecast time into a group, and dividing historical cloud pictures with a distance of 12-18 hours into a group; the method comprises the steps of dividing historical meteorological element forecast information with the distance of 18-24 hours between the historical current time and the historical forecast time into a group, and dividing historical cloud pictures with the distance of 18-24 hours into a group. Based on the dividing process, a plurality of groups of historical meteorological element forecast information sets and a plurality of groups of historical cloud picture sets can be obtained.
Then, step 302 may specifically be: and inputting the information in each group of historical meteorological element forecast information sets into the corresponding initial cloud picture forecast submodel for training to obtain the output result of each group.
Correspondingly, step 303 may specifically be: and according to the output result of each group and the historical cloud pictures in the historical cloud picture set corresponding to the group, performing feedback updating on the initial cloud picture forecasting sub-model until the training is finished, and determining the initial cloud picture forecasting sub-model at the training end as the target cloud picture forecasting sub-model.
Therefore, a plurality of target cloud picture forecast submodels can be obtained.
It should be noted that, the above-mentioned manner of dividing the historical meteorological element forecast information and the historical cloud chart can increase the number of training samples when training the target cloud chart forecast sub-model, improve the training efficiency, and reduce the difficulty of obtaining the training samples.
Step 304: acquiring meteorological element forecast information within a preset time length from the current moment in the target area.
Step 304 is similar to the implementation process and technical principle of step 101, and is not described herein again.
Step 305: and inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time length from the current time in a target area output by the target cloud picture forecasting model.
The target cloud picture forecasting model is a model trained by a machine learning method in advance according to historical meteorological element forecasting information and historical cloud pictures in a target area.
Step 305 is similar to the implementation process and technical principle of step 102, and is not described herein again.
The cloud picture forecasting method provided by this embodiment realizes satellite cloud picture forecasting by using a machine learning algorithm. The method is simple in calculation, easy to operate and convenient to popularize to other kinds of satellite cloud pictures and other numerical modes.
On one hand, the cloud picture forecasting method provided by the embodiment can obtain the forecast cloud picture only by acquiring the meteorological element forecasting information and inputting the meteorological element forecasting information into the target cloud picture forecasting model, and is simple to implement and low in complexity; on the other hand, the target cloud picture forecasting model can obtain stable output without a starting process, so that the cloud picture forecasting method provided by the embodiment can obtain a forecast cloud picture with high accuracy at the initial stage of forecasting.
Fig. 4 is a schematic structural diagram of a cloud image forecasting device according to an embodiment of the present invention. As shown in fig. 4, the cloud picture forecasting device provided in this embodiment includes the following modules: a first obtaining module 41 and a first determining module 42.
The first obtaining module 41 is configured to obtain meteorological element forecast information within a preset time from the current time in the target area.
Optionally, the first obtaining module 41 is specifically configured to: acquiring meteorological element forecast information within a preset time length from the current moment in a target area based on a weather forecast WRF mode according to the reference meteorological element forecast information within the preset time length from the current moment in the reference area; wherein the reference region is a region including the target region, and an area of the reference region is larger than an area of the target region.
Optionally, the meteorological element forecast information comprises at least one of the following parameters: the forecasted cloud volume, the forecasted cloud water volume, the forecasted cloud ice volume, the forecasted air temperature, the forecasted air pressure, and the forecasted humidity.
The first determining module 42 is configured to input the meteorological element forecast information into a pre-trained target cloud picture forecasting model, so as to obtain a forecast cloud picture within a preset time period from the current time in a target area output by the target cloud picture forecasting model.
The target cloud picture forecasting model is a model trained by a machine learning method in advance according to historical meteorological element forecasting information and historical cloud pictures in a target area.
Optionally, the meteorological element forecast information is meteorological element forecast information corresponding to each forecast time. The target cloud picture forecasting model comprises a plurality of target cloud picture forecasting sub-models. The apparatus also includes a second determination module.
And the second determining module is used for determining the target cloud picture forecasting submodel corresponding to each forecasting time according to the distance between each forecasting time and the current time.
The first determining module 42 is specifically configured to: and inputting the meteorological element forecast information corresponding to each forecast time into the corresponding target cloud picture forecast sub-model to obtain the forecast cloud picture of each forecast time within a preset time length from the current time in the target area.
The cloud picture forecasting device provided by the embodiment of the invention can execute the cloud picture forecasting method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Fig. 5 is a schematic structural diagram of a cloud image forecasting device according to another embodiment of the present invention. In this embodiment, based on the embodiment shown in fig. 4 and various optional implementation schemes, a detailed description is made on other modules included in the cloud picture forecasting apparatus. As shown in fig. 5, the cloud picture forecasting device provided in this embodiment further includes the following modules: a second obtaining module 51, a third determining module 52 and a fourth determining module 53.
And a second obtaining module 51, configured to obtain historical meteorological element forecast information and a historical cloud map of the target area.
And the third determining module 52 is configured to input the historical meteorological element forecasting information into the initial cloud atlas forecasting model for training, so as to obtain an output result.
And the fourth determining module 53 is configured to perform feedback updating on the initial cloud picture prediction model according to the output result and the corresponding historical cloud picture until the training is finished, and determine the initial cloud picture prediction model at the end of the training as the target cloud picture prediction model.
Optionally, the initial cloud picture forecasting model comprises a plurality of initial cloud picture forecasting submodels. The device also includes: and the dividing module is used for dividing the historical meteorological element forecast information into a plurality of groups of historical meteorological element forecast information sets according to the distance between the historical current time and the historical forecast time, and dividing the historical cloud picture into a plurality of groups of historical cloud picture sets.
The third determining module 52 is specifically configured to: and inputting the information in each group of historical meteorological element forecast information sets into the corresponding initial cloud picture forecast submodel for training to obtain the output result of each group.
The fourth determining module 53 is specifically configured to: and according to the output result of each group and the historical cloud pictures in the historical cloud picture set corresponding to the group, performing feedback updating on the initial cloud picture forecasting sub-model until the training is finished, and determining the initial cloud picture forecasting sub-model at the training end as the target cloud picture forecasting sub-model.
The cloud picture forecasting device provided by the embodiment of the invention can execute the cloud picture forecasting method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention. As shown in fig. 6, the computer device comprises a processor 60 and a memory 61. The number of the processors 60 in the computer device may be one or more, and one processor 60 is taken as an example in fig. 6; the processor 60 and the memory 61 of the computer device may be connected by a bus or other means, as exemplified by the bus connection in fig. 6.
The memory 61 is a computer-readable storage medium, and can be used to store software programs, computer-executable programs, and modules, such as program instructions and modules corresponding to the cloud image forecasting method in the embodiment of the present invention (for example, the first obtaining module 41 and the first determining module 42 in the cloud image forecasting apparatus). The processor 60 executes various functional applications of the computer device and the cloud picture forecasting method, i.e., implements the cloud picture forecasting method described above, by executing software programs, instructions, and modules stored in the memory 61.
The memory 61 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the computer device, and the like. Further, the memory 61 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory 61 may further include memory located remotely from the processor 60, which may be connected to a computer device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The present invention also provides a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a cloud forecasting method, the method comprising:
acquiring meteorological element forecast information within a preset time length from the current moment in a target area;
inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecasting cloud picture within a preset time length from the current time in the target area output by the target cloud picture forecasting model; the target cloud picture forecasting model is a model trained in advance by adopting a machine learning method according to historical meteorological element forecasting information and the historical cloud pictures in the target area.
Of course, the storage medium provided by the embodiments of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the cloud picture forecasting method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a computer device, or a network device) to execute the cloud picture forecasting method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the cloud picture forecasting device, the included units and modules are only divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A cloud picture forecasting method, the method comprising:
acquiring meteorological element forecast information within a preset time length from the current moment in a target area;
inputting the meteorological element forecasting information into a pre-trained target cloud picture forecasting model to obtain a forecasting cloud picture within a preset time length from the current time in the target area output by the target cloud picture forecasting model; the target cloud picture forecasting model is a model trained in advance by adopting a machine learning method according to historical meteorological element forecasting information and the historical cloud pictures in the target area.
2. The method according to claim 1, wherein the meteorological element forecast information is meteorological element forecast information corresponding to each forecast time; the target cloud picture forecasting model comprises a plurality of target cloud picture forecasting sub models;
the method further comprises the following steps:
and determining a target cloud picture forecast sub-model corresponding to each forecast time according to the distance between each forecast time and the current time.
3. The method of claim 2, wherein the inputting the meteorological element forecast information into a pre-trained target cloud picture forecasting model to obtain a forecast cloud picture within a preset time period from the current time in the target area output by the target cloud picture forecasting model comprises:
and inputting the meteorological element forecast information corresponding to each forecast time into the corresponding target cloud picture forecast sub-model to obtain the forecast cloud picture of each forecast time within a preset time length from the current time in the target area.
4. The method of claim 1, wherein the obtaining of the meteorological element forecast information within the target area within a preset time period from the current time comprises:
acquiring weather element forecast information within a preset time from the current time in the target area based on a weather forecast WRF mode according to reference weather element forecast information within a preset time from the current time in the reference area; wherein the reference region is a region including the target region, and an area of the reference region is larger than an area of the target region.
5. The method according to any one of claims 2 to 4, wherein the target cloud picture forecasting model is generated in a manner that includes:
acquiring historical meteorological element forecast information and a historical cloud picture of a target area;
inputting the historical meteorological element forecasting information into an initial cloud picture forecasting model for training to obtain an output result;
and according to the output result and the corresponding historical cloud picture, performing feedback updating on the initial cloud picture forecasting model until the training is finished, and determining the initial cloud picture forecasting model at the training end as the target cloud picture forecasting model.
6. The method of claim 5, wherein the initial cloud picture forecasting model comprises a plurality of initial cloud picture forecasting sub-models;
after obtaining the historical meteorological element forecast information and the historical cloud chart of the target area, the method further comprises the following steps:
and dividing the historical meteorological element forecast information into a plurality of groups of historical meteorological element forecast information sets according to the distance between the historical current time and the historical forecast time, and dividing the historical cloud picture into a plurality of groups of historical cloud picture sets.
7. The method of claim 6, wherein the inputting the historical meteorological element forecast information into an initial cloud prediction model for training, and obtaining an output result comprises:
and inputting the information in each group of historical meteorological element forecast information sets into the corresponding initial cloud picture forecast submodel for training to obtain the output result of each group.
8. The method of claim 7, wherein the feedback updating of the initial cloud picture forecasting model according to the output result and the corresponding historical cloud pictures until the training is finished, and determining the initial cloud picture forecasting model at the training end as the target cloud picture forecasting model comprises:
and according to the output result of each group and the historical cloud pictures in the historical cloud picture set corresponding to the group, performing feedback updating on the initial cloud picture forecasting sub-model until the training is finished, and determining the initial cloud picture forecasting sub-model at the training end as the target cloud picture forecasting sub-model.
9. A computer device, characterized in that the computer device comprises:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the cloud picture forecasting method of any of claims 1-8.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a cloud picture forecasting method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110799416.4A CN113420939A (en) | 2021-07-15 | 2021-07-15 | Cloud picture forecasting method, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110799416.4A CN113420939A (en) | 2021-07-15 | 2021-07-15 | Cloud picture forecasting method, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113420939A true CN113420939A (en) | 2021-09-21 |
Family
ID=77721093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110799416.4A Pending CN113420939A (en) | 2021-07-15 | 2021-07-15 | Cloud picture forecasting method, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113420939A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114090266A (en) * | 2021-12-01 | 2022-02-25 | 中科三清科技有限公司 | Air quality forecast generation method and device |
| CN114185114A (en) * | 2022-02-16 | 2022-03-15 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Correction method and device for rainfall forecast falling area, electronic equipment and medium |
| CN114372212A (en) * | 2022-03-21 | 2022-04-19 | 国家超级计算天津中心 | Numerical weather forecasting method and device based on supercomputer |
| CN115453661A (en) * | 2022-11-14 | 2022-12-09 | 中科星图维天信(北京)科技有限公司 | Weather forecasting method, weather forecasting device, weather forecasting equipment and storage medium |
| CN115508917A (en) * | 2022-11-22 | 2022-12-23 | 中国民用航空局空中交通管理局航空气象中心 | Method, device, equipment and storage medium for forecasting airport meteorological elements |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242374A (en) * | 2020-01-10 | 2020-06-05 | 上海眼控科技股份有限公司 | Lightning prediction method, device, computer equipment and computer readable storage medium |
| CN111337929A (en) * | 2020-03-26 | 2020-06-26 | 上海眼控科技股份有限公司 | Meteorological cloud picture prediction method and device, computer equipment and storage medium |
| CN111507929A (en) * | 2020-04-15 | 2020-08-07 | 上海眼控科技股份有限公司 | Meteorological cloud picture prediction method and device, computer equipment and storage medium |
| CN112802063A (en) * | 2021-02-03 | 2021-05-14 | 阳光电源股份有限公司 | Satellite cloud picture prediction method and device and computer readable storage medium |
| CN112926789A (en) * | 2021-03-17 | 2021-06-08 | 阳光电源股份有限公司 | Satellite cloud picture prediction method, prediction device and readable storage medium |
-
2021
- 2021-07-15 CN CN202110799416.4A patent/CN113420939A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242374A (en) * | 2020-01-10 | 2020-06-05 | 上海眼控科技股份有限公司 | Lightning prediction method, device, computer equipment and computer readable storage medium |
| CN111337929A (en) * | 2020-03-26 | 2020-06-26 | 上海眼控科技股份有限公司 | Meteorological cloud picture prediction method and device, computer equipment and storage medium |
| CN111507929A (en) * | 2020-04-15 | 2020-08-07 | 上海眼控科技股份有限公司 | Meteorological cloud picture prediction method and device, computer equipment and storage medium |
| CN112802063A (en) * | 2021-02-03 | 2021-05-14 | 阳光电源股份有限公司 | Satellite cloud picture prediction method and device and computer readable storage medium |
| CN112926789A (en) * | 2021-03-17 | 2021-06-08 | 阳光电源股份有限公司 | Satellite cloud picture prediction method, prediction device and readable storage medium |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114090266A (en) * | 2021-12-01 | 2022-02-25 | 中科三清科技有限公司 | Air quality forecast generation method and device |
| CN114090266B (en) * | 2021-12-01 | 2022-12-09 | 中科三清科技有限公司 | Air quality forecast generation method and device |
| CN114185114A (en) * | 2022-02-16 | 2022-03-15 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Correction method and device for rainfall forecast falling area, electronic equipment and medium |
| CN114185114B (en) * | 2022-02-16 | 2022-04-29 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Correction method and device for rainfall forecast falling area, electronic equipment and medium |
| CN114372212A (en) * | 2022-03-21 | 2022-04-19 | 国家超级计算天津中心 | Numerical weather forecasting method and device based on supercomputer |
| CN114372212B (en) * | 2022-03-21 | 2022-07-08 | 国家超级计算天津中心 | Numerical weather forecasting method and device based on supercomputer |
| CN115453661A (en) * | 2022-11-14 | 2022-12-09 | 中科星图维天信(北京)科技有限公司 | Weather forecasting method, weather forecasting device, weather forecasting equipment and storage medium |
| CN115453661B (en) * | 2022-11-14 | 2023-01-10 | 中科星图维天信(北京)科技有限公司 | Weather forecasting method, weather forecasting device, weather forecasting equipment and storage medium |
| CN115508917A (en) * | 2022-11-22 | 2022-12-23 | 中国民用航空局空中交通管理局航空气象中心 | Method, device, equipment and storage medium for forecasting airport meteorological elements |
| CN115508917B (en) * | 2022-11-22 | 2023-04-28 | 中国民用航空局空中交通管理局航空气象中心 | Airport weather element forecasting method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107703564B (en) | Rainfall prediction method and system and electronic equipment | |
| CN113420939A (en) | Cloud picture forecasting method, equipment and storage medium | |
| US10795056B2 (en) | Local weather forecast | |
| US11555921B2 (en) | Method and apparatus for predicting severe convection weather | |
| Weekes et al. | Long-term wind resource assessment for small and medium-scale turbines using operational forecast data and measure–correlate–predict | |
| CN113255239B (en) | Deep learning method for predicting icing thickness of ultrahigh voltage transmission line tower | |
| CN114332977A (en) | Key point detection method and device, electronic equipment and storage medium | |
| JP2009192312A (en) | Device for predicting generation of thunder, and method therefor, and program | |
| CN116720156A (en) | Weather element forecasting method based on graph neural network multi-mode weather data fusion | |
| WO2022039675A1 (en) | Method and apparatus for forecasting weather, electronic device and storage medium thereof | |
| CN114330110A (en) | Model training method, photovoltaic power generation power prediction method and system | |
| FR3019408A1 (en) | METHOD FOR PREDICTING SOLAR RADIATION ON THE GROUND IN THE VICINITY OF A SITE | |
| CN112614207B (en) | Contour line drawing method, device and equipment | |
| US20210063603A1 (en) | Distributed computing system and method for generating atmospheric wind forecasts | |
| TWI717796B (en) | System, method and storage medium for estimating the amount of sunshine in geographic location by artificial intelligence | |
| CN115587664B (en) | Solar radiation prediction method and system based on graph convolution long-short-term memory network | |
| CN115909255A (en) | Image generation method, image segmentation method, image generation device, image segmentation device, vehicle-mounted terminal and medium | |
| CN115099318A (en) | Training method and application method of wind speed space-time prediction network and electronic equipment | |
| CN115564100A (en) | Photovoltaic power prediction method, system and equipment | |
| CN114943188A (en) | Snow depth inversion method and device based on random forest model | |
| CN115267948A (en) | Method for predicting meteorological satellite cloud picture at continent level | |
| CN113435118A (en) | Irradiance determination method, irradiance determination device, irradiance determination equipment and storage medium | |
| CN115186865A (en) | Mass distributed renewable energy power prediction method based on MapReduce and cloud picture | |
| Mengxuan et al. | Development of cloud recognition system for ground-based cloud images based on machine vision | |
| de Grado et al. | On the development of a digital meteorological model for simulating future air traffic management automation |
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 |