CN111830595A - Meteorological element prediction method and equipment - Google Patents
Meteorological element prediction method and equipment Download PDFInfo
- Publication number
- CN111830595A CN111830595A CN202010526785.1A CN202010526785A CN111830595A CN 111830595 A CN111830595 A CN 111830595A CN 202010526785 A CN202010526785 A CN 202010526785A CN 111830595 A CN111830595 A CN 111830595A
- Authority
- CN
- China
- Prior art keywords
- meteorological
- weather
- meteorological element
- graph
- sequence
- 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
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Environmental & Geological Engineering (AREA)
- Engineering & Computer Science (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 application provides a meteorological element prediction method and equipment, which can generate a sequence of meteorological element graphs according to meteorological element information of a plurality of meteorological stations, input the sequence of the meteorological element graphs into a pre-constructed meteorological element prediction model to obtain a predicted sequence of the meteorological element graphs, and determine the prediction sequences of the meteorological elements corresponding to the meteorological stations according to the prediction sequences, thereby realizing the prediction of the meteorological elements in the future, improving the accuracy of the meteorological element prediction, reducing the computational complexity of the meteorological element prediction and improving the real-time performance of the meteorological element prediction.
Description
Technical Field
The present application relates to the field of weather forecasting, and in particular, to a method and an apparatus for forecasting meteorological parameters.
Background
Currently, China has gone ahead in the world in terms of meteorological data acquisition, and by the end of 2018, China has had 6 thousands of automatic meteorological stations covering 95.6% of villages and towns, and the number and density of the meteorological stations have reached world first. The meteorological data acquired in a high-density and large-scale manner provides abundant basic data for weather forecast, and also provides higher requirements for meteorological data processing and weather forecast methods.
The existing weather forecasting method is mainly a numerical calculation method, numerical solution of a massive formula needs to be carried out on a high-performance calculation platform, the calculation amount is very large, the calculation time is long, and the real-time performance of weather forecasting is difficult to guarantee. Other weather forecasting methods such as weather forecasting through a traditional bayesian network have the problem that the accuracy of weather element forecasting is not high to different degrees, and the requirements of future weather forecasting are difficult to meet.
Disclosure of Invention
An object of the present application is to provide a meteorological element prediction method and apparatus, which are used to solve the problem of inaccurate meteorological element prediction in the prior art.
In order to achieve the above object, the present application provides a method for predicting meteorological elements, wherein the method comprises:
generating a sequence of a meteorological element graph according to meteorological element information of a plurality of meteorological stations;
inputting the sequence of the meteorological element graph into a meteorological element prediction model which is constructed in advance, and obtaining a predicted meteorological element graph sequence;
and determining a prediction sequence corresponding to the meteorological elements of the plurality of meteorological stations according to the predicted sequence of the meteorological element graph.
Further, generating a sequence of weather element maps based on the weather element information of the plurality of weather stations, comprising:
determining nodes and edges of a weather element graph according to geographic information of a plurality of weather stations, wherein the geographic information comprises any combination of one or more of the following: geographical location information, regional landform information;
generating a meteorological element graph corresponding to the acquisition time according to the acquisition time corresponding to the meteorological element information of a plurality of meteorological stations;
and sequencing the multiple meteorological element graphs according to the acquisition time to generate a time sequence of the meteorological element graphs.
Further, determining nodes and edges of the weather element graph according to the geographic information of the plurality of weather stations comprises:
taking a plurality of weather stations as nodes in the weather element graph, and taking the geographical position information of the weather stations as the position information of corresponding nodes;
and determining distance information between the two nodes according to the position information of the two nodes, and generating an edge between the two nodes in the meteorological element graph according to the distance information.
Further, determining nodes and edges of the weather element graph according to the geographic information of the plurality of weather stations comprises:
taking a plurality of weather stations as nodes in the weather element graph, and taking regional geomorphic information of the weather stations as geomorphic information of corresponding nodes;
and generating an edge between the two nodes in the meteorological element graph according to the geomorphic information of the two nodes.
Further, the step of sequencing a plurality of weather element graphs according to the acquisition time to generate a time sequence of the weather element graphs includes:
acquiring a first meteorological element diagram corresponding to first acquisition time, and adding the first meteorological element diagram into a time sequence of the meteorological element diagram;
acquiring a second meteorological element graph corresponding to second acquisition time, wherein the second acquisition time is a moment after the first acquisition time;
adding the second meteorological element graph to the time sequence of the meteorological element graphs, and determining the relative positions of the second meteorological element graph and the first meteorological element graph according to a preset sequencing mode.
Further, the step of pre-constructing the meteorological element prediction model comprises:
acquiring a sequence of meteorological element graphs for training;
sequentially inputting the sequence of the meteorological element diagram for training into a recurrent neural network, wherein a repeating module in the recurrent neural network predicts the sequence of the meteorological element diagram for training through a gate structure to obtain a prediction sequence of the meteorological element diagram for training;
comparing the prediction sequence with a corresponding real value sequence, and adjusting the parameters of the recurrent neural network according to the comparison result;
and when a preset training termination condition is met, taking the current parameters of the recurrent neural network as the parameters of the meteorological element prediction model.
Further, the recurrent neural network is a graph-based long-short term memory network or a graph-based gate cycle unit network.
Further, the repetition module uses the following formula:
ht＝Ot⊙tanh(ct)，
wherein itFor the output of the t-time input gate, ftFor the output of the forgetting gate at time t, ctOutput of the cell state at t time, OtOutput of the output gate for t time, htFor output of t-time hidden features, xtFor input of t time, Wxi，Whi，ωci，Wxf，Whf，ωcf，Wxc，Whc，Wxo，Who，ωcoIn order to be the parameters of the model,
for the operation of graph convolution, bi，bf，bc，boIs a bias parameter.
Based on another aspect of the present application, there is also provided an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, cause the apparatus to perform the method of predicting a meteorological element as described above.
The present application also provides a computer readable medium having computer readable instructions stored thereon, the computer readable instructions being executable by a processor to implement the aforementioned method of predicting meteorological elements.
Compared with the prior art, the scheme provided by the application can generate the sequence of the meteorological element map according to the meteorological element information of the meteorological stations, then inputs the sequence of the meteorological element map into the meteorological element prediction model which is constructed in advance to obtain the predicted sequence of the meteorological element map, and then determines the prediction sequences of the meteorological elements corresponding to the meteorological stations according to the prediction sequences, so that the future meteorological element prediction is realized, the accuracy of the meteorological element prediction can be improved, the computational complexity of the meteorological element prediction is reduced, and the real-time performance of the meteorological element prediction is improved.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of a method for predicting meteorological elements according to some embodiments of the present disclosure;
FIG. 2 is a schematic diagram of a meteorological element diagram provided in some embodiments of the present application.
Detailed Description
The present application is described in further detail below with reference to the attached figures.
In a typical configuration of the present application, the terminal and the network device each include one or more processors (CPUs), input/output interfaces, network interfaces, and memories.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
Fig. 1 illustrates a meteorological element prediction method according to some embodiments of the present application, which may specifically include the following steps:
step S101, generating a sequence of weather element graphs according to weather element information of a plurality of weather stations;
step S102, inputting the sequence of the meteorological element graph into a meteorological element prediction model which is constructed in advance, and acquiring the sequence of the predicted meteorological element graph;
and step S103, determining a prediction sequence corresponding to the meteorological elements of the plurality of meteorological stations according to the predicted sequence of the meteorological element graph.
The scheme is particularly suitable for a scene which is expected to carry out weather forecast according to weather element data of a plurality of weather stations, can generate a sequence of weather element diagrams according to the weather element information of the plurality of weather stations, inputs the sequence of the weather element diagrams into a weather element prediction model which is trained in advance to obtain the sequence of the predicted weather element diagrams, and finally obtains weather element prediction sequences corresponding to the weather stations according to the predicted diagram sequences.
In step S101, a sequence of weather element maps is first generated based on weather element information of a plurality of weather stations. Here, the weather station is a station for collecting weather element data, and is located at different geographical positions, and may have different distribution densities in different geographical areas, and the distance between the weather stations may be different. Meteorological elements may include, but are not limited to: temperature, humidity, wind speed, pressure and other meteorological data.
In some embodiments of the present application, the generating of the sequence of the weather element map according to the weather element information of the plurality of weather stations may include the following specific steps:
1) determining nodes and edges of a weather element graph according to geographic information of a plurality of weather stations, wherein the geographic information comprises any combination of one or more of the following: geographical location information, regional landform information; here, the geographic information of the weather station refers to various information related to the geography of the weather station, such as the geographic location, the landform (such as forest or desert, etc.) of the area; the nodes and the edges of the weather element graph can be determined according to one kind of geographic information, and the nodes and the edges of the weather element graph can also be determined according to the combination of various kinds of geographic information;
2) generating a meteorological element graph corresponding to the acquisition time according to the acquisition time corresponding to the meteorological element information of the plurality of meteorological stations; here, the meteorological element information collected by the meteorological station may include meteorological element data collected at a plurality of times, for example, the meteorological element is temperature, and the meteorological station a is configured to, at 15: 00 temperature data collected was 23.6 degrees, at 15: 30 at 23.4 degrees, weather station B at 15: 00 temperature data collected was 22.9 degrees, at 15: 30 is 22.8 degrees, then according to the acquisition time 15: 00, determining 23.6 degrees of corresponding temperature data of the meteorological station A and 22.9 degrees of corresponding temperature data of the meteorological station B, and generating and acquiring time 15: 00 corresponding temperature map;
3) and sequencing the multiple weather element graphs according to the acquisition time to generate a time sequence of the weather element graphs.
Here, a Graph refers to a Graph (Graph) in Graph theory, and is a Graph composed of a plurality of nodes (nodes) and edges (edges) connecting two nodes, and is used for depicting relationships between different nodes. The meteorological element graph comprises a graph structure and specific data, wherein the graph structure comprises a node set and an edge set, the node set comprises a plurality of nodes, the edge set comprises a plurality of edges, the nodes can comprise a plurality of attributes, and the edges can also comprise a plurality of attributes. The sequence of the weather element graph is a sequence composed of a plurality of weather element graphs, the structure of the weather element graphs in the sequence is the same, and specific data in the graphs are different, for example, the weather element graph a and the weather element graph B may both include the same node set and the same edge set, but values of node attributes therein are different.
Fig. 2 shows a structure of a weather element graph in some embodiments, where one weather station is used as one node in the graph, 4 nodes are available from 4 weather stations, whether an edge exists between the nodes is determined according to a preset condition, an edge is generated between the nodes meeting the condition, 4 edges are generated between 4 nodes in the graph, and 2 nodes connected by the edge have a certain association, where the preset condition may be determined according to geographic information of the weather station, for example.
In some embodiments of the present application, the nodes and edges of the weather element graph are determined according to the geographic information of the plurality of weather stations, and one way may specifically include the following steps:
1) taking a plurality of weather stations as nodes in a weather element graph, and taking the geographical position information of the weather stations as the position information of corresponding nodes;
2) and determining distance information between the two nodes according to the position information of the two nodes, and generating an edge between the two nodes in the meteorological element graph according to the distance information.
Specifically, each weather station is used as one node in the weather element graph, the position information of the weather station is used as the position information of the corresponding node, and the position information of the node can be represented by the node attribute. Furthermore, meteorological elements of the meteorological station may also be used as node attributes.
In addition, the edges between the nodes in the meteorological element graph are generated according to the distance information between the two nodes, the distance information between the nodes is determined according to the position information of the two nodes, the position information of the nodes is still the longitude and latitude, the distance information between the two nodes can be the straight line distance of the longitude and latitude of the two nodes, the edges between the two nodes are the straight lines connecting the two nodes at the moment, and the attribute of the edges can include the distance information between the nodes.
In some embodiments of the present application, generating an edge between two nodes in a meteorological element graph according to the distance information may specifically include the following steps: and if the distance information is smaller than the preset threshold value, generating an edge between two nodes in the meteorological element graph, otherwise, not generating the edge between the two nodes. Here, since the distance affects the relevance of the meteorological element data in the meteorological stations, not all meteorological stations generate one edge, and only meteorological stations with a distance smaller than a certain preset threshold value generate an edge, which indicates that the meteorological element data between the meteorological stations may have mutual influence, and meteorological stations with a distance larger than the preset threshold value consider that the meteorological element data do not have mutual influence. The predetermined threshold may be set based on user experience, for example, 20 km, an edge is generated between weather stations within 20 km, and no edge is generated between weather stations over 20 km.
In some embodiments of the present application, the nodes and edges of the weather element graph are determined according to the geographic information of the plurality of weather stations, and another manner may specifically include the following steps:
1) taking a plurality of weather stations as nodes in a weather element graph, and taking regional geomorphic information of the weather stations as geomorphic information of corresponding nodes;
2) and generating an edge between the two nodes in the meteorological element graph according to the geomorphic information of the two nodes. Here, a variety of different preset rules may be used according to whether the geomorphic information generates edges between nodes. For example, one preset rule is that an edge is generated between two nodes only if the geomorphic information is completely the same, and an edge is not generated between two nodes if the geomorphic information is different; another preset rule is to generate an edge between nodes that meet the connection rule, and if the connection rule is to generate an edge between a node a whose landscape information is a forest and a node B whose landscape information is a plain, an edge may be generated between a node a whose landscape information is a forest and a node B whose landscape information is a plain.
Here, whether to generate an edge may also be determined based on various kinds of geographic information, and whether to generate an edge between nodes or the like may be comprehensively considered based on, for example, location information and geomorphic information of nodes.
In some embodiments of the present application, the sorting the multiple weather element maps according to the acquisition time to generate a time sequence of the weather element maps may specifically include the following steps:
1) acquiring a first meteorological element diagram corresponding to first acquisition time, and adding the first meteorological element diagram into a time sequence of the meteorological element diagram; if the time sequence of the weather element graph is empty, the first weather element graph is used as the first graph in the time sequence, if the time sequence of the weather element graph is not empty, the weather element graph corresponding to the time closest to the time corresponding to the first weather element graph in the time sequence can be found, then the first weather element graph is determined to be added before or after the weather element graph according to the preset sorting mode of the graphs in the sequence, for example, if the graph sorting mode is ascending, the first weather element graph is added after the weather element graph, and if the graph sorting mode is descending, the first weather element graph is added before the weather element graph;
2) acquiring a second meteorological element graph corresponding to second acquisition time, wherein the second acquisition time is a moment after the first acquisition time;
3) adding the second meteorological element graph into the time sequence of the meteorological element graphs, and determining the relative positions of the second meteorological element graph and the first meteorological element graph according to a preset sequencing mode; here, the predetermined sorting manner may be ascending or descending according to the corresponding time of the graphs in the time series, for example, if the predetermined sorting manner is ascending, the relative position of the second weather pixel graph and the first weather pixel graph is determined as the rear, and the second weather pixel graph is added behind the first weather pixel graph; and if the preset sorting mode is descending, determining that the relative position of the second meteorological element graph and the first meteorological element graph is before, and adding the second meteorological element graph to the front of the first meteorological element graph.
Thereafter, the remaining weather element maps are added to the time series of the weather element maps according to the above method, and after all the weather element maps are added, the time series of the final weather element maps is obtained.
In step S102, the sequence of the weather element map is input to a weather element prediction model constructed in advance, and the predicted sequence of the weather element map is acquired. Here, the pre-constructed meteorological element prediction model is a model constructed through a neural network, and in some embodiments of the present application, the construction of the model may specifically include the following steps:
1) acquiring a sequence of meteorological element graphs for training; here, the sequence of the weather element map used for training may also be generated according to a similar method to the aforementioned method for generating the sequence of the weather element map, except that the sequence of the weather element map used for training corresponds to a sequence of real values with a subsequent time, the sequence of real values also being a sequence composed of the weather element map, wherein the weather element map is also generated according to the real weather element data collected by the weather station, except that the corresponding time is later than the time corresponding to the weather element map used for training;
2) sequentially inputting the sequences of the meteorological element graphs for training into a recurrent neural network, wherein a repeating module in the recurrent neural network predicts the sequences of the meteorological element graphs for training through a gate structure to obtain a prediction sequence of the meteorological element graphs for training; in some embodiments of the present application, a model is constructed using a Receivable Neural Network (RNN) for input, including a repetition module for generating output data for a current time from input data and output data of the repetition module from a previous time;
in addition, the Gate (Gate) structure in the duplicated module is a way of optionally passing through information, and there may be a plurality of Gate structures, such as an input Gate, a forgetting Gate and an output Gate, where the input Gate is used to determine input data of the duplicated module, the forgetting Gate is used to selectively forget the input data, and the output Gate is used to output final data;
3) comparing the prediction sequence with the corresponding real value sequence, and adjusting the parameters of the recurrent neural network according to the comparison result; here, the comparison result between the predicted sequence and the actual value sequence can be calculated through a loss function, and then parameters of a network layer in the recurrent neural network are gradually adjusted through a certain optimization method such as a gradient descent method;
4) when a preset training termination condition is met, taking the parameters of the current cyclic neural network as the parameters of the meteorological element prediction model; here, the preset training termination condition may be determined according to the user requirement, for example, the training may be performed for a certain number of times, the loss function is smaller than a certain threshold, and the like; after the training termination condition is met, the currently obtained parameters of the recurrent neural network are the parameters of the meteorological element prediction model, and the model is determined after the model parameters are determined.
In some embodiments of the present application, the recurrent neural network may be a graph-based long-short term memory network or a graph-based gate-cycle cell network. The Long-short term memory network (LSTM) is a special RNN, mainly aims to solve the problems of gradient elimination and gradient explosion in the Long sequence training process, and compared with a common recurrent neural network, the LSTM can have better performance in a longer sequence. The Gate Recycling Unit (GRU) network is a variant of the LSTM, which is simpler than the LSTM network in structure, and is easier to train, and can improve the training efficiency to a great extent.
In some embodiments of the present application, the repetition module may use the following formula:
ht＝Ot⊙tanh(ct)，
wherein itFor the output of the t-time input gate, ftFor the output of the forgetting gate at time t, ctOutput of the cell state at t time, OtOutput of the output gate for t time, htFor output of t-time hidden features, xtFor input of t time, Wxi，Whi，ωci，Wxf，Whf，ωcf，Wxc，Whc，Wxo，Who，ωcoIn order to be the parameters of the model,
for the operation of graph convolution, bi，bf，bc，boIs a bias parameter.
Here, the graph convolution operation is a convolution operation performed on the graph, and since the graph is a non-euclidean space, the number of neighbors of a node in the graph is not fixed, and the number of neighbors of a node (or a pixel point) in the euclidean space represented by an image is fixed, the ordinary convolution applied to the euclidean space cannot be directly applied to the non-euclidean space.
Here, the repeating module may use an LSTM-based neural network structure, and the repeating module includes an input gate, a forgetting gate and an output gate, and the input and output of each gate are determined by specific parameters in the above formula, as a Hadamard product (Hadamardproduct), which is a multiplication.
In step S103, a prediction sequence corresponding to the weather elements of the plurality of weather stations is determined based on the predicted sequence of the weather element map. Here, each of the predicted weather element maps in the sequence of the predicted weather element maps includes the predicted values of the weather elements of all the weather stations at the current time, and in order to obtain the predicted value of the weather element corresponding to each weather station, the predicted value of the weather element may be obtained from the node attribute in the predicted weather element map, and then the predicted value of the weather element is determined as the predicted value of the weather element of the weather station corresponding to the node.
Some embodiments of the present application also provide an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, cause the apparatus to perform the aforementioned method of predicting meteorological elements.
Some embodiments of the present application also provide a computer readable medium having computer readable instructions stored thereon, the computer readable instructions being executable by a processor to implement the aforementioned method for predicting meteorological elements.
In summary, the scheme provided by the application can generate the sequence of the meteorological element graph according to the meteorological element information of a plurality of meteorological stations, then input the sequence of the meteorological element graph into the meteorological element prediction model which is constructed in advance to obtain the predicted sequence of the meteorological element graph, and then determine the prediction sequence of the meteorological element corresponding to the meteorological stations according to the prediction sequence, thereby realizing the prediction of the meteorological elements in the future, improving the accuracy of the meteorological element prediction, reducing the computational complexity of the meteorological element prediction and improving the real-time performance of the meteorological element prediction.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application comprises a device comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the device to perform a method and/or a solution according to the aforementioned embodiments of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Claims (10)
1. A meteorological element prediction method, wherein the method comprises the following steps:
generating a sequence of a meteorological element graph according to meteorological element information of a plurality of meteorological stations;
inputting the sequence of the meteorological element graph into a meteorological element prediction model which is constructed in advance, and obtaining a predicted meteorological element graph sequence;
and determining a prediction sequence corresponding to the meteorological elements of the plurality of meteorological stations according to the predicted sequence of the meteorological element graph.
2. The method of claim 1, wherein generating a sequence of weather element maps from weather element information of a plurality of weather stations comprises:
determining nodes and edges of a weather element graph according to geographic information of a plurality of weather stations, wherein the geographic information comprises any combination of one or more of the following: geographical location information, regional landform information;
generating a meteorological element graph corresponding to the acquisition time according to the acquisition time corresponding to the meteorological element information of a plurality of meteorological stations;
and sequencing the multiple meteorological element graphs according to the acquisition time to generate a time sequence of the meteorological element graphs.
3. The method of claim 2, wherein determining the nodes and edges of the weather element graph based on the geographic information of the plurality of weather stations comprises:
taking a plurality of weather stations as nodes in the weather element graph, and taking the geographical position information of the weather stations as the position information of corresponding nodes;
and determining distance information between the two nodes according to the position information of the two nodes, and generating an edge between the two nodes in the meteorological element graph according to the distance information.
4. The method of claim 2, wherein determining the nodes and edges of the weather element graph based on the geographic information of the plurality of weather stations comprises:
taking a plurality of weather stations as nodes in the weather element graph, and taking regional geomorphic information of the weather stations as geomorphic information of corresponding nodes;
and generating an edge between the two nodes in the meteorological element graph according to the geomorphic information of the two nodes.
5. The method of claim 2, wherein sorting the plurality of weather element maps according to the acquisition time to generate a time series of weather element maps comprises:
acquiring a first meteorological element diagram corresponding to first acquisition time, and adding the first meteorological element diagram into a time sequence of the meteorological element diagram;
acquiring a second meteorological element graph corresponding to second acquisition time, wherein the second acquisition time is a moment after the first acquisition time;
adding the second meteorological element graph to the time sequence of the meteorological element graphs, and determining the relative positions of the second meteorological element graph and the first meteorological element graph according to a preset sequencing mode.
6. The method of claim 1, wherein the pre-building of the meteorological element prediction model comprises:
acquiring a sequence of meteorological element graphs for training;
sequentially inputting the sequence of the meteorological element diagram for training into a recurrent neural network, wherein a repeating module in the recurrent neural network predicts the sequence of the meteorological element diagram for training through a gate structure to obtain a prediction sequence of the meteorological element diagram for training;
comparing the prediction sequence with a corresponding real value sequence, and adjusting the parameters of the recurrent neural network according to the comparison result;
and when a preset training termination condition is met, taking the current parameters of the recurrent neural network as the parameters of the meteorological element prediction model.
7. The method of claim 6, wherein the recurrent neural network is a graph-based long-short term memory network or a graph-based gate-cycle cell network.
8. The method of claim 6, wherein the repetition module uses the following formula:
ht＝Ot⊙tanh(ct)，
wherein itFor the output of the t-time input gate, ftFor the output of the forgetting gate at time t, ctOutput of the cell state at t time, OtOutput of the output gate for t time, htFor output of t-time hidden features, xtFor input of t time, Wxi，Whi，ωci，Wxf，Whf，ωcf，Wxc，Whc，Wxo，Who，ωcoIn order to be the parameters of the model,
for the operation of graph convolution, bi，bf，bc，boIs a bias parameter.
9. An apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, cause the apparatus to perform the method of any of claims 1 to 8.
10. A computer readable medium having computer readable instructions stored thereon which are executable by a processor to implement the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010526785.1A CN111830595A (en) | 2020-06-09 | 2020-06-09 | Meteorological element prediction method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010526785.1A CN111830595A (en) | 2020-06-09 | 2020-06-09 | Meteorological element prediction method and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111830595A true CN111830595A (en) | 2020-10-27 |
Family
ID=72899148
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010526785.1A Pending CN111830595A (en) | 2020-06-09 | 2020-06-09 | Meteorological element prediction method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111830595A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033901A (en) * | 2021-03-30 | 2021-06-25 | 上海眼控科技股份有限公司 | Meteorological element prediction method, device, equipment and storage medium |
| CN115391746A (en) * | 2022-10-28 | 2022-11-25 | 航天宏图信息技术股份有限公司 | Interpolation method, device, electronic device and medium for meteorological element data |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110082636A1 (en) * | 2006-03-03 | 2011-04-07 | Inrix, Inc. | Dynamic time series prediction of future traffic conditions |
| CN102169194A (en) * | 2010-08-25 | 2011-08-31 | 清华大学 | System for perceiving and forecasting regional weather |
| CN103426026A (en) * | 2013-09-10 | 2013-12-04 | 信阳师范学院 | Method for forecasting hybrid neural network and recognizing scenic spot meteorological elements |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN109274727A (en) * | 2018-08-31 | 2019-01-25 | 中国联合网络通信集团有限公司 | Based on the meteorological data sharing method of block chain, apparatus and system |
| CN109543295A (en) * | 2018-11-21 | 2019-03-29 | 国网青海省电力公司 | The meteorological element data processing method and device of numerical weather forecast |
| CN110687619A (en) * | 2019-09-18 | 2020-01-14 | 北京农业信息技术研究中心 | Meteorological data verification method and system for farmland meteorological station |
| CN110689179A (en) * | 2019-09-18 | 2020-01-14 | 北京工商大学 | Water bloom prediction method based on space-time sequence mixed model |
| CN110908014A (en) * | 2019-12-11 | 2020-03-24 | 国网湖南省电力有限公司 | Galloping refined correction forecasting method and system |
| CN110956677A (en) * | 2019-12-13 | 2020-04-03 | 上海眼控科技股份有限公司 | Radar map prediction method, radar map prediction device, computer equipment and storage medium |
| CN111062410A (en) * | 2019-11-05 | 2020-04-24 | 复旦大学 | Star information bridge weather prediction method based on deep learning |
| CN111126704A (en) * | 2019-12-25 | 2020-05-08 | 河海大学常州校区 | Multi-region precipitation prediction model construction method based on multi-graph convolution and memory network |
| CN111158068A (en) * | 2019-12-31 | 2020-05-15 | 哈尔滨工业大学（深圳） | Short-term prediction method and system based on simple convolutional recurrent neural network |
| 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 |
-
2020
- 2020-06-09 CN CN202010526785.1A patent/CN111830595A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110082636A1 (en) * | 2006-03-03 | 2011-04-07 | Inrix, Inc. | Dynamic time series prediction of future traffic conditions |
| CN102169194A (en) * | 2010-08-25 | 2011-08-31 | 清华大学 | System for perceiving and forecasting regional weather |
| CN103426026A (en) * | 2013-09-10 | 2013-12-04 | 信阳师范学院 | Method for forecasting hybrid neural network and recognizing scenic spot meteorological elements |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN109274727A (en) * | 2018-08-31 | 2019-01-25 | 中国联合网络通信集团有限公司 | Based on the meteorological data sharing method of block chain, apparatus and system |
| CN109543295A (en) * | 2018-11-21 | 2019-03-29 | 国网青海省电力公司 | The meteorological element data processing method and device of numerical weather forecast |
| CN110687619A (en) * | 2019-09-18 | 2020-01-14 | 北京农业信息技术研究中心 | Meteorological data verification method and system for farmland meteorological station |
| CN110689179A (en) * | 2019-09-18 | 2020-01-14 | 北京工商大学 | Water bloom prediction method based on space-time sequence mixed model |
| CN111062410A (en) * | 2019-11-05 | 2020-04-24 | 复旦大学 | Star information bridge weather prediction method based on deep learning |
| CN110908014A (en) * | 2019-12-11 | 2020-03-24 | 国网湖南省电力有限公司 | Galloping refined correction forecasting method and system |
| CN110956677A (en) * | 2019-12-13 | 2020-04-03 | 上海眼控科技股份有限公司 | Radar map prediction method, radar map prediction device, computer equipment and storage medium |
| CN111126704A (en) * | 2019-12-25 | 2020-05-08 | 河海大学常州校区 | Multi-region precipitation prediction model construction method based on multi-graph convolution and memory network |
| CN111158068A (en) * | 2019-12-31 | 2020-05-15 | 哈尔滨工业大学（深圳） | Short-term prediction method and system based on simple convolutional recurrent neural network |
| 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 |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033901A (en) * | 2021-03-30 | 2021-06-25 | 上海眼控科技股份有限公司 | Meteorological element prediction method, device, equipment and storage medium |
| CN115391746A (en) * | 2022-10-28 | 2022-11-25 | 航天宏图信息技术股份有限公司 | Interpolation method, device, electronic device and medium for meteorological element data |
| CN115391746B (en) * | 2022-10-28 | 2023-01-31 | 航天宏图信息技术股份有限公司 | Interpolation method, interpolation device, electronic device and medium for meteorological element data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113792929B (en) | Traffic flow prediction method, electronic device and storage medium | |
| CN102496187B (en) | Method for tracking contour line to boundary and fault based on triangular mesh | |
| CN112270259B (en) | SAR image ship target rapid detection method based on lightweight convolutional neural network | |
| WO2019019653A1 (en) | Device and method for extracting topographical boundary | |
| CN111830595A (en) | Meteorological element prediction method and equipment | |
| CN111461423A (en) | High-precision gridding air quality inference method, system, terminal equipment and storage medium | |
| Bermingham et al. | Mining place-matching patterns from spatio-temporal trajectories using complex real-world places | |
| Li et al. | Online real-time trajectory analysis based on adaptive time interval clustering algorithm | |
| CN111505738A (en) | Method and equipment for predicting meteorological factors in numerical weather forecast | |
| Sabarish et al. | Clustering of trajectory data using hierarchical approaches | |
| Zhang et al. | SolarGAN: Synthetic annual solar irradiance time series on urban building facades via Deep Generative Networks | |
| CN114220017A (en) | Remote sensing data scale self-adaptive adjusting method and device, storage medium and equipment | |
| CN113537143A (en) | 3D shape recognition method and device based on graph convolution neural network | |
| Chen et al. | Plane segmentation for a building roof combining deep learning and the RANSAC method from a 3D point cloud | |
| CN114566048B (en) | Traffic control method based on multi-view self-adaptive space-time diagram network | |
| CN116150694A (en) | Dynamic graph anomaly detection method | |
| Chen et al. | Dynamic path flow estimation using automatic vehicle identification and probe vehicle trajectory data: A 3D convolutional neural network model | |
| Liu | Multidimensional nonlinear landscape planning based on parameter feature extraction and multimedia technology | |
| Liu et al. | The “M” in digital elevation models | |
| CN117312396A (en) | Lane positioning method and device, electronic equipment and readable storage medium | |
| CN111486847B (en) | Unmanned aerial vehicle navigation method and system | |
| CN114565730A (en) | Method and device for processing map vector sample and computer readable storage medium | |
| Zhang | Optimal planning algorithm of forest wetland tourism path based on GIS | |
| CN113077112A (en) | Data acquisition method and device, computer equipment and storage medium | |
| CN118445284B (en) | Beidou space-time traffic data recovery method based on transformation-induced low-rank tensor |
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 | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20221104 |