CN118313273A - Short-term climate prediction method and system based on combination of regional mode and machine learning - Google Patents
Short-term climate prediction method and system based on combination of regional mode and machine learning Download PDFInfo
- Publication number
- CN118313273A CN118313273A CN202410707459.9A CN202410707459A CN118313273A CN 118313273 A CN118313273 A CN 118313273A CN 202410707459 A CN202410707459 A CN 202410707459A CN 118313273 A CN118313273 A CN 118313273A
- Authority
- CN
- China
- Prior art keywords
- data
- feature
- air temperature
- temperature data
- daily average
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000010801 machine learning Methods 0.000 title claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 22
- 239000005431 greenhouse gas Substances 0.000 claims abstract description 16
- 238000004088 simulation Methods 0.000 claims abstract description 9
- 230000004927 fusion Effects 0.000 claims description 17
- 238000011176 pooling Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 241000282414 Homo sapiens Species 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention provides a short-term climate prediction method and a short-term climate prediction system based on the combination of regional mode and machine learning, which are implemented by acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target region; determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data; according to the initial field, the boundary field and the greenhouse gas emission data, simulating the air temperature of the target area through an area climate mode to obtain a simulation value of daily average air temperature data; and correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data, so that the accuracy of prediction is improved.
Description
Technical Field
The invention relates to a short-term climate prediction method and a short-term climate prediction system, in particular to a short-term climate prediction method and a short-term climate prediction system based on combination of regional mode and machine learning.
Background
Existing short-term climate prediction methods rely mostly on physical models, such as Regional Climate Models (RCMs), which tend to be lacking in accuracy in local or specific regions. The invention patent CN114970222A carries out deviation correction on the simulation value of the daily average air temperature data through HASM, and along with the development of machine learning technology, particularly the strong capability of a neural network in the aspects of pattern recognition and nonlinear relation modeling, the neural network is combined with a regional climate model, so that the efficiency and the accuracy of short-term climate prediction are expected to be improved.
Disclosure of Invention
The invention aims to solve the problem of insufficient short-term climate prediction precision in the prior art, and provides a short-term climate prediction method and a short-term climate prediction system based on the combination of a regional mode and machine learning so as to improve the efficiency and the accuracy of short-term climate prediction.
The invention is realized by the following technical scheme:
the first aspect of the invention provides a short-term climate prediction method based on the combination of regional mode and machine learning, which specifically comprises the following steps:
acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
Determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
according to the initial field, the boundary field and the greenhouse gas emission data, simulating the air temperature of the target area through an area climate mode to obtain a simulation value of daily average air temperature data;
And correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data.
Further, the topographic data reads global topographic geographic information through a Terrain program, and determines the running area of the mode, geographic information in the area, such as elevation, land utilization type, map projection, ocean and land masks and the like.
Further, sea level temperature data is processed by SST (Sea Surface Temperature) program.
Further, the analysis count uses ERA-Interm data in the climate analysis data set.
Further, the land utilization data is data reflecting the state, characteristics, dynamic change and distribution characteristics of a land utilization system and a land utilization element, and development and utilization of land, improvement and improvement, management and protection, land utilization planning and the like of human beings. Land use data takes the form of data provided by the regional climate model RegCM, authorities.
Further, the correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data includes:
And inputting the analog value, the air pressure, the air speed, the relative humidity and the dew point temperature of the daily average air temperature data into an error correction neural network, and outputting the predicted value of the daily average air temperature data.
Further, the error correction neural network includes:
The method comprises the steps of marking the output of a third convolution module as a characteristic F1, obtaining a characteristic F2 by a fourth convolution module and a first convolution layer which are sequentially connected with the F1 input, obtaining a characteristic F3 by a fifth convolution module and a second convolution layer which are sequentially connected with the F1 input, and carrying out first fusion on the characteristic F2 and the characteristic F3 to obtain the characteristic F2 The saidAnd respectively inputting a maximum pooling layer and an average pooling layer to obtain a feature Fm and a feature Fa, performing first fusion on the feature Fm and the feature Fa to obtain a feature Fma, performing second fusion on the feature Fma and the feature F1 to obtain a feature F, and inputting the feature F into a third convolution layer and a LeakyRelu layer to obtain a predicted value of daily average air temperature data.
A second aspect of the present invention provides a short-term climate prediction system based on a combination of regional mode and machine learning, comprising in particular the following modules: the system comprises a data acquisition module, a preprocessing module, a prediction module and a correction module;
the data acquisition module is used for acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
The preprocessing module is used for determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
The prediction module is used for simulating the air temperature of the target area through an area climate mode according to the initial field, the boundary field and the greenhouse gas emission data to obtain a simulation value of daily average air temperature data;
the correction module corrects the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data.
Compared with the prior art, the invention has the following beneficial effects:
Acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area; determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data; according to the initial field, the boundary field and the greenhouse gas emission data, simulating the air temperature of the target area through an area climate mode to obtain a simulation value of daily average air temperature data; and correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data, so that the accuracy of prediction is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for the description of the embodiments or the prior art will be briefly described, and it is apparent that the drawings in the following description are only one embodiment of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a short-term climate prediction method based on a combination of regional mode and machine learning;
FIG. 2 is a block diagram of a short-term climate prediction system based on a combination of regional mode and machine learning;
Fig. 3 is a block diagram of an error correction neural network.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the present invention easy to understand, the technical solutions in the embodiments of the present invention are clearly and completely described below to further illustrate the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all versions.
The invention is described in further detail below with reference to the attached drawing figures:
as shown in fig. 1, a first aspect of the present invention provides a short-term climate prediction method based on a combination of region mode and machine learning, which specifically includes the following steps:
acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
Determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
according to the initial field, the boundary field and the greenhouse gas emission data, simulating the air temperature of the target area through an area climate mode to obtain a simulation value of daily average air temperature data;
And correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data.
Further, the topographic data reads global topographic geographic information through a Terrain program, and determines the running area of the mode, geographic information in the area, such as elevation, land utilization type, map projection, ocean and land masks and the like.
Further, sea level temperature data is processed by SST (Sea Surface Temperature) program.
Further, the analysis count uses ERA-Interm data in the climate analysis data set.
Further, the land utilization data is data reflecting the state, characteristics, dynamic change and distribution characteristics of a land utilization system and a land utilization element, and development and utilization of land, improvement and improvement, management and protection, land utilization planning and the like of human beings. Land use data takes the form of data provided by the regional climate model RegCM, authorities.
Further, the correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data includes:
And inputting the analog value, the air pressure, the air speed, the relative humidity and the dew point temperature of the daily average air temperature data into an error correction neural network, and outputting the predicted value of the daily average air temperature data.
Further, the error correction neural network includes:
The method comprises the steps of marking the output of a third convolution module as a characteristic F1, obtaining a characteristic F2 by a fourth convolution module and a first convolution layer which are sequentially connected with the F1 input, obtaining a characteristic F3 by a fifth convolution module and a second convolution layer which are sequentially connected with the F1 input, and carrying out first fusion on the characteristic F2 and the characteristic F3 to obtain the characteristic F2 The saidAnd respectively inputting a maximum pooling layer and an average pooling layer to obtain a feature Fm and a feature Fa, performing first fusion on the feature Fm and the feature Fa to obtain a feature Fma, performing second fusion on the feature Fma and the feature F1 to obtain a feature F, and inputting the feature F into a third convolution layer and a LeakyRelu layer to obtain a predicted value of daily average air temperature data.
Further, the feature F2 and the feature F3 are subjected to first fusion to obtain a featureComprising the following steps:
3。
further, performing the first fusion on the feature Fm and the feature Fa to obtain a feature Fma includes:
a。
further, performing the second fusion of the feature Fma and the feature F1 to obtain the feature F includes:
1。
further, the first through fourth convolution modules include conv+bn+ relu.
Further, the fifth convolution module includes an adaptive pooling layer +conv +bn + relu.
As shown in fig. 2, the application provides a short-term climate prediction system based on a combination of regional mode and machine learning, which specifically comprises the following modules: the system comprises a data acquisition module, a preprocessing module, a prediction module and a correction module;
the data acquisition module is used for acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
The preprocessing module is used for determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
The prediction module is used for simulating the air temperature of the target area through an area climate mode according to the initial field, the boundary field and the greenhouse gas emission data to obtain a simulation value of daily average air temperature data;
the correction module corrects the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data.
Having described the main technical features and fundamental principles of the present invention and related advantages, it will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, but may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The above detailed description is, therefore, to be taken in all respects as illustrative and not restrictive, the scope of the invention 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.
Furthermore, it should be understood that although the present disclosure describes embodiments in terms of various embodiments, not every embodiment is described in terms of a single embodiment, but rather that the descriptions of embodiments are merely provided for clarity, and that the descriptions of embodiments in terms of various embodiments are provided for persons skilled in the art on the basis of the description.
Claims (6)
1. The short-term climate prediction method based on the combination of regional mode and machine learning is characterized by comprising the following steps of:
acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
Determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
according to the initial field, the boundary field and the greenhouse gas emission data, simulating the air temperature of the target area through an area climate mode to obtain a simulation value of daily average air temperature data;
Correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data;
the correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data comprises the following steps:
Inputting the analog value, the air pressure, the air speed, the relative humidity and the dew point temperature of the daily average air temperature data into an error correction neural network, and outputting the predicted value of the daily average air temperature data;
The error correction neural network includes:
The method comprises the steps of marking the output of a third convolution module as a characteristic F1, obtaining a characteristic F2 by a fourth convolution module and a first convolution layer which are sequentially connected with the F1 input, obtaining a characteristic F3 by a fifth convolution module and a second convolution layer which are sequentially connected with the F1 input, and carrying out first fusion on the characteristic F2 and the characteristic F3 to obtain the characteristic F2 The saidInputting a maximum pooling layer and an average pooling layer to obtain a feature Fm and a feature Fa respectively, performing first fusion on the feature Fm and the feature Fa to obtain a feature Fma, performing second fusion on the feature Fma and a feature F1 to obtain a feature F, and inputting the feature F into a third convolution layer and a LeakyRelu layer to obtain a predicted value of daily average air temperature data;
The feature F2 and the feature F3 are subjected to first fusion to obtain features Comprising the following steps:
3;
The first fusing the feature Fm and the feature Fa to obtain a feature Fma includes:
a;
performing the second fusing of the feature Fma and the feature F1 to obtain a feature F includes:
1。
2. The short-term climate prediction method based on the combination of regional mode and machine learning according to claim 1, wherein the topographic data reads global topographic geographic information through a Terrain program to determine the operating region of the mode and the geographic information in the region.
3. A short-term climate prediction method based on a combination of regional model and machine learning as claimed in claim 1 wherein sea level temperature data is processed by SST (Sea Surface Temperature) program.
4. A short-term climate prediction method based on a combination of regional mode and machine learning as claimed in claim 1 wherein the analysis count uses ERA-inter data in the climate analysis data set.
5. A short-term climate prediction method based on a combination of regional mode and machine learning as claimed in claim 1 wherein land utilization data is data provided by regional climate mode RegCM4 authorities.
6. A short-term climate prediction system based on a combination of regional mode and machine learning, which is characterized by comprising the following modules: the system comprises a data acquisition module, a preprocessing module, a prediction module and a correction module;
the data acquisition module is used for acquiring topographic data, sea level temperature data, analysis data, land utilization data and greenhouse gas emission data of a target area;
The preprocessing module is used for determining an initial field and a boundary field of the regional climate mode according to the topographic data, the sea level temperature data, the analysis data and the land utilization data;
The prediction module is used for simulating the air temperature of the target area through an area climate mode according to the initial field, the boundary field and the greenhouse gas emission data to obtain a simulation value of daily average air temperature data;
The correction module corrects the analog value of the daily average air temperature data based on the error correction neural network to obtain a predicted value of the daily average air temperature data;
the correcting the analog value of the daily average air temperature data based on the error correction neural network to obtain the predicted value of the daily average air temperature data comprises the following steps:
Inputting the analog value, the air pressure, the air speed, the relative humidity and the dew point temperature of the daily average air temperature data into an error correction neural network, and outputting the predicted value of the daily average air temperature data;
The error correction neural network includes:
The method comprises the steps of marking the output of a third convolution module as a characteristic F1, obtaining a characteristic F2 by a fourth convolution module and a first convolution layer which are sequentially connected with the F1 input, obtaining a characteristic F3 by a fifth convolution module and a second convolution layer which are sequentially connected with the F1 input, and carrying out first fusion on the characteristic F2 and the characteristic F3 to obtain the characteristic F2 The saidInputting a maximum pooling layer and an average pooling layer to obtain a feature Fm and a feature Fa respectively, performing first fusion on the feature Fm and the feature Fa to obtain a feature Fma, performing second fusion on the feature Fma and a feature F1 to obtain a feature F, and inputting the feature F into a third convolution layer and a LeakyRelu layer to obtain a predicted value of daily average air temperature data;
The feature F2 and the feature F3 are subjected to first fusion to obtain features Comprising the following steps:
3;
The first fusing the feature Fm and the feature Fa to obtain a feature Fma includes:
a;
performing the second fusing of the feature Fma and the feature F1 to obtain a feature F includes:
1。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410707459.9A CN118313273A (en) | 2024-06-03 | 2024-06-03 | Short-term climate prediction method and system based on combination of regional mode and machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410707459.9A CN118313273A (en) | 2024-06-03 | 2024-06-03 | Short-term climate prediction method and system based on combination of regional mode and machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118313273A true CN118313273A (en) | 2024-07-09 |
Family
ID=91727498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410707459.9A Pending CN118313273A (en) | 2024-06-03 | 2024-06-03 | Short-term climate prediction method and system based on combination of regional mode and machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118313273A (en) |
-
2024
- 2024-06-03 CN CN202410707459.9A patent/CN118313273A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108227041B (en) | Horizontal visibility forecasting method based on site measured data and mode result | |
| US20190338758A1 (en) | Method and device for calculating power generation of wind farm | |
| US6128019A (en) | Real-time multi-sensor synthetic environment created from a feature and terrain database using interacting and updatable abstract models | |
| CN108986207A (en) | A kind of road based on true road surface data and emulation modelling method is built along the line | |
| CN109191561B (en) | Urban three-dimensional building rapid modeling method for traffic environment pollution simulation | |
| CN111382507B (en) | Global troposphere delay modeling method based on deep learning | |
| CN114840900B (en) | Automatic generation method of derivative BIM (building information modeling) component based on i-GBDT (generic block transformation) technology | |
| CN116842877B (en) | Small-scale three-dimensional wind field reconstruction algorithm based on multi-source data comprehensive utilization | |
| CN109143408A (en) | Combine short-term precipitation forecasting procedure in dynamic area based on MLP | |
| CN109784544A (en) | A kind of noise prediction of level terrain wind power plant and Optimal Deployment Method | |
| CN106874602A (en) | Process meteorological data method and apparatus | |
| CN109493360B (en) | Method for automatically generating building outline by artificially scattering shape points of ground objects | |
| CN113627096B (en) | Fine wind field simulation method based on spatial correlation and monitoring data | |
| CN113743027B (en) | Method and device for drawing wind resource map based on CFD technology | |
| CN115017679A (en) | Simulation method and device for atmospheric pollution diffusion, storage medium and electronic equipment | |
| CN115879498A (en) | Marine environment parameter rapid refined forecasting method based on graph neural network | |
| CN111882677B (en) | Method and device for editing three-dimensional plant model, electronic equipment and storage medium | |
| CN116702299B (en) | Garden design method based on garden simulation | |
| CN118313273A (en) | Short-term climate prediction method and system based on combination of regional mode and machine learning | |
| CN115905997B (en) | Wind turbine generator meteorological disaster early warning method and system based on prediction deviation optimization | |
| CN115586591A (en) | Automatic storm surge forecasting method and device | |
| CN115860248A (en) | Wind speed numerical assimilation short-term prediction method and system | |
| CN115034159A (en) | Power prediction method, device, storage medium and system for offshore wind farm | |
| CN113095012B (en) | Splicing and fusing method for numerical simulation calculation results of wind power plant flow field partitions | |
| CN112596127A (en) | Novel method for calculating typhoon potential generation index |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |